Book SCM

S upply Chain M anagement
Management
y
for C ompetitiv
Competitiv
ompetitiveeA dv
Adv antage
dvantage
nl
Concepts & C ases
Cases
O
n
io
at
ul
irc
C
d
ite
im
rL
Fo
ABOUT THE AUTHORS
Narayan Rangaraj is Professor of Industrial Engineering and Operations Research at

the Indian Institute of Technology, Bombay since 1990. He is a Ph.D. from the
Johns Hopkins University and a B.Tech from IIT Bombay.
y
He has worked in areas of transport operations management, inventory and other
nl
aspects of supply chain management, and applications of operations research
techniques to various sectors. His recent areas of work are railway operations,
O
containerisation, the aviation sector and pricing/revenue management.
n
G Raghuram is Indian Railways Chair Professor in Rail Transport and Infrastructure
Management at the Indian Institute of Management, Ahmedabad. He has been on
io
the faculty there since 1985. He is a Ph.D. from Northwestern University, PGDM
at
from IIM-Ahmedabad and holds a B. Tech degree from IIT-Madras.
He has specialised in logistics and supply chain management, and infrastructure
ul
and transportation systems. His areas of research, consultancy, case studies and
publications include railways, ports and shipping, air and road sector, service
irc
organisations and issues in logistics and supply chain management. Raghuram has
taught at Northwestern University and Tulane University, USA. He has been a
C
visiting faculty at universities in USA, Canada, Yugoslavia, Singapore, UAE and

several institutions in India. He has consulted for over 95 organisations in the area of
Logistics and Supply Chain Management, and Infrastructure Development. Also, he
d
has published over 30 papers and written over 100 case studies.
ite
Mandyam M Srinivasan is the Ball Corporation Distinguished Professor of Business

at the University of Tennessee, Knoxville, USA. He has taught at this university since
im
1992. He has a Ph.D. from Northwestern University, a PGDM from IIM-

Bangalore, and an M Tech degree from IIT-Madras.
rL
Srinivasan has many years of experience in the automobile industry. He has

consulted with, and trained senior executives from numerous U.S. Fortune 500
organisations on how to build and manage lean supply chains. He has also trained
Fo
senior executives in India, Germany, Taiwan, Sweden, Singapore, China, Mexico and
Romania. Srinivasan has published over forty papers in peer-reviewed journals and
has written many articles for trade journals and magazines. He received the Franz
Edelman Award for Achievement in Operations Research from the Institute for
Operations Research and Management Sciences in 2006. He is the Editor of IIE
Transactions on Design & Manufacturing.
Management
y
for C ompetitiv
Competitiv
ompetitiveeA dv
Adv antage
dvantage
nl
Concepts & C ases
Cases
O
n
io
at
Narayan Rangaraj
ul
Indian Institute of Technology, Bombay
G Raghuram
irc
Indian Institute of Management, Ahmedabad
Mandyam M Srinivasan
C
University of Tennessee, USA

d
ite
im
rL
Fo
Tata McGraw-Hill Publishing Company Limited

NEW DELHI
McGraw-Hill Offices
New Delhi New York St Louis San Francisco Auckland Bogotá Caracas
Kuala Lumpur Lisbon London Madrid Mexico City Milan Montreal
San Juan Santiago Singapore Sydney Tokyo Toronto
Tata McGraw-Hill
Published by Tata McGraw-Hill Publishing Company Limited,
7 West Patel Nagar, New Delhi 110 008.
y
Supply Chain Management for Competitive Advantage: Concepts & Cases
nl
Copyright © 2009, by Tata McGraw-Hill Publishing Company Limited.
No part of this publication may be reproduced or distributed in any form or by any means, electronic,
O
mechanical, photocopying, recording, or otherwise or stored in a database or retrieval system without the prior
written permission of the publishers. The program listings (if any) may be entered, stored and executed in a
computer system, but they may not be reproduced for publication.
n
This edition can be exported from India only by the publishers,
io
Tata McGraw-Hill Publishing Company Limited.
ISBN (13): 978-0-07-022163-5
at
ISBN (10): 0-07-022163-4
Managing Director: Ajay Shukla ul
General ManagerPublishing (B&E/HSSL and School): V Biju Kumar
Editorial ManagerB&E: Tapas K Maji
irc
Junior Editorial Executive: Hemant K Jha
Junior Editorial Executive: Rajneesh Roy
Executive (Editorial Services): Anubha Srivastava
C
Senior Production Manager: Manohar Lal

Senior Production Assistant: Neelangshu Santra
General ManagerMarketing (Higher Education & School): Michael J. Cruz
d
Asst. Product Manager: Vijay Sarathi Jagannathan

ControllerProduction: Rajender P Ghansela
ite
Asst. General ManagerProduction: B L Dogra
Information contained in this work has been obtained by Tata McGraw-Hill, from sources believed to be
im
reliable. However, neither Tata McGraw-Hill nor its authors guarantee the accuracy or completeness of any
information published herein, and neither Tata McGraw-Hill nor its authors shall be responsible for any
errors, omissions, or damages arising out of use of this information. This work is published with the under-
standing that Tata McGraw-Hill and its authors are supplying information but are not attempting to render
rL
engineeringor other professional services. If such services are required, the assistance of an appropriate
professional should be sought.
Typeset at Text-o-Graphics, B1/56 Arawali Apartment, Sector 34, Noida 201 301 and printed at
Fo
Avon Printers, Plot No. 16, Main Loni Road, Jawahar Nagar Industrial Area, Shahdara, Delhi 110 094
Cover Design: K Anoop
Cover printed at: Rashtriya Printers
RYRQBRYFRXDBR
Preface
y
This book is positioned to serve two audiences: the MBA and senior student in a
nl
business programme, and the professional working in industry who wishes to gain a
deeper understanding of supply chain management and logistics especially in an
O
Indian context. The book thus provides an overview of concepts as well as intensive
material for analysis and training in the area that has come to be known as Supply
n
Chain Management. The topics covered are far reaching in practical scope.
io
A special feature of the book is the collection of detailed cases. All cases are
situated in the Indian business environment, as it makes a transition from traditional
at
business practices to new ones in an arena of global competition. This attempts to fill
a stated gap in the academic and business literature in the area. The cases are all
ul
drawn from real organisations and many are presented with actual data and
descriptions that provide a realistic picture of issues in supply chain management.
irc
The conceptual material in the book presents a set of integrative viewpoints on the
area. Chapter 1 discusses the drivers of supply chain management, such as
C
technology, and increased consumer power in a global market. The chapter discusses
how supply chain management practices have evolved over time to cope with these
d
drivers and highlights organisations that have benefited from better supply chain
management. Chapter 2 analyses the dynamics in a supply chain, and shows how
ite
lead time reduction and better information management are vital to supply chain
performance. The importance of big picture thinking is underscored in this chapter.
im
Chapter 3 talks about supply chain design and presents a roadmap for designing and
managing the lean supply chain. It discusses how the design can be adapted based on
the products managed by the supply chain. The role of performance measurements is
rL
highlighted in Chapter 4. This chapter uses a case to demonstrate how organisations

within the supply chain can benefit by shared metrics. Chapter 5 discusses supply
Fo
chain efficiencies and the benefits gained by applying lean principles.

Manufacturing is presented as the primary driver of benefits in value addition in
material processing and transformation. The implications of this emphasis on supply
chain efficiencies on other parts of the supply chain are explored. From a single
supply chain player’s point of view, upstream and downstream interface management
are taken up in a symmetric manner in Chapters 6 and 7. A special focus of the book
is to look at transportation and transfer of material between supply chain entities in
vi Preface
the language of SCM, in Chapter 8. Chapter 9 presents the pervasive role of modern
information technology in supply chain management and finally, Chapter 10
presents a range of quantitative tools and techniques that are used in different facets
of decision making in this area.
y
The cases are then presented and are intended to be read as stand alone pieces of
analysis and as a tool for integrated learning. The very essence of supply chain
nl
management is a holistic, multi-actor, multi-departmental view of decision making
O
and the cases reflect that concern. In some cases, a particular sub-area of decision
making can be isolated, but in general, the scope of analysis and decision making has
to be inferred and proposed in an open way. This delimiting of the scope of analysis
n
in a coherent manner is part and parcel of the learning objectives of the course.
io
We have instructor suggestions and teaching notes as support for many of the
cases, to be provided through the publisher. We also provide a teaching plan for
at
some typical course offerings in this area.
ul
As a guide to students, a brief overview of each case, and some questions for
analysis and an approach is suggested. A case positioning matrix helps in selection of
irc
cases for reading, and puts them in context. We emphasise that in our opinion, the
cases, although extensive, need to be read in their entirety and instructors and
students need to absorb the business logic and propose areas of analysis and
C
appropriate techniques from a wide selection. This breadth of vision and judgement
in selection is part of the training that distinguishes a true supply chain manager from
d
a routine application oriented view of highly specific techniques within artificial

ite
functional boundaries.
The case material, while dealing with Indian companies, does have global
conceptual validity in the general concerns of supply chain management. However, in
im
appropriate places, the specific conditions of the Indian sub-continent in terms of

business practices and overall ethos, and of infrastructural realities are emphasised
rL
upon.
Finally, the book provides some links to contemporary developments in the area
Fo
and can even serve as the starting point for some applied research. Apart from its use
as a textbook and as a guide to practitioners, it is of some value to researchers as well.
NARAYAN RANGARAJ
G RAGHURAM
MANDYAM M SRINIVASAN
Acknowledgements
y
I wish to acknowledge a number of people for their help and contributions: Suchir
nl
Sinha, Sanjeev George, Manik Bahadur, Tilak Raj Singh, Piyush Verma and Manish
Singh for some of the exercises and collection of material and references, N.
O
Hemachandra and Jayendran Venkateswaran for professional inputs and Deepa Hari
for personal support during the writing of the book.
Narayan Rangaraj
n
I thank the following co-authors of the cases for allowing us to use the material for
io
publication in this book:
Tathagata Bandyopadhyay Bibek Banerjee
at
Faculty, IIMA Faculty, IIMA
Nishchal Chudasama ul Anusha Dhasarathy
Research Associate, IIMA PGP Student, IIMA
Rekha Jain Abraham Koshy
irc
Faculty, IIMA Faculty, IIMA
D Kumar S Manikutty
Faculty, Railway Staff College, Faculty, IIMA
C
Vadodhara
Dilip Mathew Saral Mukherjee
d
Research Associate, IIMA Faculty, IIMA

ite
Preeti Monippally Rohit Pathak

PGP Student, IIMA PGP Student, IIMA
Malavika Pillai R N Prasad
im
PGP Student, IIMA Faculty, Railway Staff College,

Vadodhara
Rohithari Rajan Neeraj Sisodia
rL
PGP Student, IIMA Research Associate, IIMA

Chetan Soman
Fo
Faculty, IIMA
The organisations that have rendered direct support in the material development
efforts include:
Rajashree Cement, Grasim Industries and South Central Railway
Titan Industries Limited
CONCOR
Spices Board of India
viii Acknowledgements
Bayer CropScience Limited

FoodWorld
Airfreight Limited
and four others (who choose to remain anonymous)
Laxmi Transformers
y
Western Oil Limited
nl
Farmaid Tractors Limited
Seth Dhaniram
O
We acknowledge the research assistance for case writing provided by:
Premlata Agarwal
n
Anita Basalingappa
Sushma Choudhary
io
Ameesh Dave
at
Kapil Jain
Sunil Jaryal
Vishal Kashyap
Darshana Padia
ul
irc
Biswajyoti Pal
Parvathy Raman
Shivani Shukla
C
We also thank Macmillan India Ltd for permitting us to print the cases Laxmi
Transformers and Airfreight Limited, and Vikalpa, the Journal for Decision Makers, for
d
the case Chilli in Soup (A).

G. Raghuram
ite
I thank the College of Business Administration at the University of Tennessee,

im
Knoxville, for supporting this book project, and Cengage Learning (formerly
Thomson Learning) for permission to use material from two chapters in my book,
Streamlined: 14 Principles for Building and Managing the Lean Supply Chain,
rL
published by Cengage in 2004. I wish to thank Dr. James Reeve for permission to use
the Integrity Motors case, and Dr. Kenneth Gilbert for his keen insights, support,
and encouragement.
Fo
Mandyam M Srinivasan
The authors wish to thank Milind Padalkar for his co-writitng of Chapter 9.
They also wish to thank and N. Kumar and N. Chandrasekaran, both associated
with the CII Institute of Logistics in Channai at the time, who played a big role in
bringing them (the authors) together to work on this book and thereby provided the
physical and mental space that sparked this joint initiative. The authors also acknowledge
the help provided by Tapas Maji, Anubha Srivastava and Hemant Jha of TMH.
Authors
Contents
y
Preface v
nl
Acknowledgements vii
List of Abbreviations xiii
O
List of Cases xv
Decision Areas and Positioning of Cases xvi
n
Chapter 1. An Overview of Supply Chain Management 2
io
Introduction 3
at
1.1 Ownership of the Supply Chain 4
1.2 Supply Chain Drivers 4
1.3
1.4
ul
Supply Chains in the Producer-Centric Era 6
Financial Management and Return on Investment 10
irc
1.5 Supply Chains in the Customer-Centric Era 12
1.6 Examples of Well-run Supply Chains 15
C
Summary 32
References 33
Exercises 33
d
ite
Chapter 2. Understanding Supply Chain Dynamics 36

Introduction 37
2.1 Supply Chain Dynamics in Action 38
im
2.2 The Bullwhip Effect 40

2.3 The Impact of Lead Times 46
rL
2.4 Inventory Management and Supply Chain Dynamics 49

2.5 Offshoring and Outsourcing: on Supply Chain Dynamics and Costs 52
2.6 Lessons Learned 54
Fo
Conclusions 57
References 58
Exercises 58
x Contents
Chapter 3. Designing the Supply Chain 60

Introduction 61
3.1 Some Issues in Supply Chain Design 62
3.2 Steps for Designing and Managing Lean Supply Chains 63
Conclusions 87
y
References 88
nl
Exercises 89
O
Chapter 4. Performance Measurement 90
Introduction 91
n
4.1 Measuring Supply Chain Performance 92
4.2 The Prisoner’s Dilemma 94
io
4.3 The Integrity Motors Case 96
at
4.4 Evolving Supply Chain Metrics 105
4.5 Performance Monitoring 108ul
4.6 Competences and Performance 117
Conclusions 118
irc
References 118
Exercises 119
C
Chapter 5. SCM Effectiveness and Lean Thinking 120

Introduction 121
d
5.1 The Toyota Production System 122

ite
5.2 Creating Flow: The Tools and Techniques of Lean Thinking 124
5.3 Continuous Improvement and the Pursuit of Perfection 149
im
Conclusions 150
References 151
Exercises 151
rL
Chapter 6. SCM Across Organisations: Upstream Interface 154

Introduction 155
Fo
6.1 Procurement 157

6.2 Strategic Issues in Procurement: Sourcing 157
6.3 Operational Issues in Procurement: Purchasing 161
6.4 Items of Procurement 162
6.5 Contracts 164
Contents xi
6.6 Markets and Auctions 169

References 171
Exercises 171
Chapter 7. SCM Across Organisations: Downstream Interface 174
y
Introduction 175
nl
7.1 Distribution Management 176
7.2 Strategic Decisions in Distribution Management 176
O
7.3 Operational Management of Distribution 179
7.4 Items of Distribution and Delivery 183
n
7.5 Transportation, Storage and Warehousing 185
io
7.6 The Role of Contracts in Distribution and Marketing 187
References 189
at
Exercises 189
Chapter 8. Transportation, Storage and Warehousing
ul 192
Introduction 193
irc
8.1 Transportation Mode Choice 194
8.2 Transport Operator Decisions 195
8.3 Trucking Sector in India 197
C
8.4 Rail Transport 198

8.5 Container Operations 199
d
8.6 Air Transport 199

ite
8.7 Water Transport 200

8.8 Transport Networks 201
im
8.9 Storage and Warehousing 204

8.10 Concluding Examples and Remarks 205
References 209
rL
Exercises 209
Chapter 9. Role of Information Technology in SCM 212
Fo
Introduction 213
9.1 Basic Requirements of Supply Chain Management 214
9.2 Elements of a Modern IT System 216
9.3 A Typical Architecture of an IT System 218
9.4 Elements of the Application Layer 219
9.5 Current Practice 224
xii Contents
9.6 The Promise of Information Technology 231

Summary 234
Exercises 235
Chapter 10. Quantitative Tools for Supply Chain Management 236
y
Introduction 237
nl
10.1 Forecasting 238
10.2 Management of Inventories in Supply Chains 243
O
10.3 Linear Programming or Linear Optimisation 250
10.4 Routing Models 260
n
10.5 Pricing Decisions 261
io
References 264
Exercises 264
at
Case 1 Laxmi Transformers 271
Case 2 Rajashree Cement: Engine On Load
ul 283
Case 3 Western Oil Limited (A) 317
irc
Case 4 Farmaid Tractors Limited 336
Case 5 Supply Chain Management at Titan Industries Limited 356
Case 6 CONCOR: Tea Transportation 390
C
Case 7 Chilli in Soup (A) 412

Case 8 Bayer Cropscience: Science for a Better Supply Chain 439
d
Case 9 FoodWorld (B): Supply Chain Strategy 456

ite
Case 10 Seth Dhaniram C&FA 493

Case 11 Airfreight Limited 502
im
Index 533
rL
Fo
List of Abbreviations
y
ABC Activity Based Costing
nl
ATO Assemble to Order
BTO Build to Order
O
BTS Build to Stock
CFA Carrying and Forwarding Agent
n
COGS Cost of Goods Sold
CPFR Collaborative Planning, Forecasting and Replenishment
io
DEA Data Envelopment Analysis
EOQ Economic Order Quantity
at
ERP Enterprise Resource Planning
ETO Engineer to Order
ul
EU European Union
FMCG Fast Moving Consumer Goods
irc
GM General Motors
HUL Hindustan Unilever Limited
C
IP Integer Programming
IR Indian Railways
IT Information Technology
d
JIT Just-in-time
ite
JITD Just-in-time Distribution

LP Linear Programming
LT Lead Time
im
MIP Mixed Integer Programming

NDDB National Dairy Development Board
rL
OLAP On Line Analytical Programming

OR/MS Operations Research / Management Science
POS Point of Sale
Fo
PPV Purchase Price Variance

RAP Raw as Possible
RAPS Request for Price Quotes
RERUNS Annual Renewal of Existing Policies
RFID Radio Frequency Identification
ROI Return on Investment
xiv List of Abbreviations
RUNS Requests for Underwriting

SKD Semi Knocked Down
TH Throughput Rate
TOC Theory of Constraints
TPM Total Productive Maintenance
y
TPS Toyota Production System
nl
TQM Total Quality Management
TSP Traveling Salesman Problem
O
VMI Vendor Managed Inventory
VRP Vehicle Routing Problem
WIP Work-in-process
n
io
Case Abbreviations
LT Laxmi Transformers
at
RCL Rajashree Cement: Engine on Load
WOL(A) Western Oil Limited (A)
ul
FTL FarmAid Tractors Limited
TITAN Supply Chain Management at Titan Industries Limited
irc
CONCOR CONCOR: Tea Transportation
CIS(A) Chilli in Soup (A)
C
BCS Bayer CropScience: Science for a Better Supply Chain

FW(B) FoodWorld (B)
SDR-C&FA Seth Dhaniram – C&FA
d
AFL Airfreight Limited

ite
im
rL
Fo
List of Cases
y
S No Case Context
nl
1. Laxmi Transformers Mode choice and logistics driven market
choice for a sponge iron manufacturer
O
2. Rajashree Cement: Engine on Implications of a bulk cement supply
Load chain for Railways
n
3. Western Oil Limited (A) Determination of optimal jetty capacity
io
for a petroleum supply chain
4. FarmAid Tractors Limited Distribution logistics for a tractor
at
manufacturer
5. Supply Chain Management at ul Integrated view of issues in supply chain
Titan Industries Limited management for a watch manufacturer
6. CONCOR: Tea Transportation Marketing strategy for containerized tea
irc
transportation
7. Chilli in Soup (A) Choices for the Spices Board of India to
C
improve the image and quality of chilli

exports in the context of detection of
banned substances in the product by EU
d
8. Bayer CropScience: Science for a Supply chain improvement possibilities

ite
Better Supply Chain for a pesticide firm

9. FoodWorld (B) Supply chain strategy for a large food
im
retailing company
10. Seth Dhaniram – C&FA Role of a C&FA in the context of whether
to continue a business relationship or not
rL
11. Airfreight Limited Strategy for developing a third party

integrated logistics business, based on
Fo
the strengths of a courier business

Decision Areas and Positioning of Cases
y
nl
LT RCL WOL (A) FTL TITAN CONCOR CIS (A) BCS FW (B) SDR AFL
(1990) (2004) (2007) (1999) (1996) (2000) (2003) (2005) (1998) (1999) (1997)
O
Choice of markets and contracting * * *
Choice of sources and contracting * *
Plant location
Production structure *
n
Production outsourcing and contracting * *
io
Distribution network design * * *
Procurement network design * *
at
Logistics outsourcing to 3PLs and contracting * * *
Warehouse location * * *
Allocation * * *
ul
Plant layout and logistics *
Procurement planning
irc
· Planning horizon * *
· Order quantity * *
Production planning
C
· Planning horizon *
· Batch sizes *
Despatch planning
d
· Planning horizon * * * *
· Order processing * * * * * *
ite
Inventory stock norms * * * *
Warehouse operations * * *
Transportation
im
· Mode choice * * *
· Investment planning * * *
· Shipment size * * * * *
rL
· Routing * * *
· Contracting * * * * * *
Packaging and material handling * *
Information issues * * *
Fo
Quality issues * *
y
nl
O
Management
for C ompetitiv
Competitiv
ompetitiveeA dv
Adv antage
dvantage
n
io
Concepts & C ases
Cases
at
ul
irc
C
d
ite
im
rL
Fo
y
CHAPTER
nl
1
An Overview of Supply Chain
Management
O
n
CHAPTER OUTLINE
io
Introduction
at
1.1 Ownership of the Supply Chain
1.2 Supply Chain Drivers ul
1.2.1 Technology
1.2.2 Competition
irc
1.2.3 Business and Social Environment
1.2.4 Policy and Regulation Environment
1.2.5 The Move from a Producer-Centric to a Customer-Centric Focus
C
1.3 Supply Chains in the Producer-Centric Era

1.3.1 The Rise of Batch Production
d
1.3.2 Management Structures in the Producer-Centric Era

1.3.3 The Rise of Cost Accounting
ite
1.4 Financial Management and Return on Investment

1.5 Supply Chains in the Customer-Centric Era
im
1.5.1 New Business Environments Need New Management Systems

1.6 Examples of Well-Run Supply Chains
Summary
rL
References
Exercises
Fo
An Overview of Supply Chain Management 3
INTRODUCTION
W
hat is a supply chain and cent. The impact of these numbers can
what is supply chain be understood from the fact that sup-
y
management? While re- ply chain management costs ac-
searchers and practitioners provide counted for approximately 9.5% of the
nl
multiple definitions to describe these Gross Domestic Product of the United
terms, a simple definition for a supply States that year. A reduction of 5 per
O
chain is that it is a set of organisations cent in these costs would result in a
engaged in the delivery of a product or savings of about US$50 billion—that is
a service to the customer or the end not small change.
user. Organisations in the supply chain
n
In addition, logistics is only one of
include the suppliers, intermediaries, the many cost components in the sup-
io
transportation and logistics providers, ply chain. If other costs such as order
and customers. Supply chain manage- processing, materials acquisition and
ment also requires the coordination
at
inventory, supply chain planning, sup-
and collaboration of all the ply chain financing, and information
organisations in the supply chain. It is management costs are considered, the
the management of all activities in-
volved in sourcing, procurement, con-
ul
potential savings from effective supply
chain management would be much
version and logistics management to
irc
higher. Furthermore, these figures do
deliver the right product or service to not consider a multitude of hidden or
the customer at the right time, in the unknown costs. They also do not con-
proper quantity, and in the most cost- sider the possible benefits. These fig-
C
effective manner. ures do not, for example, consider the

Organisations that successfully ap- increased market-share potential for
ply good supply chain practices usu- supply chains that are able to respond
d
ally incur lower costs than their com- faster to customer needs. A study by
ite
petitors. A survey conducted in 2001 [R. D’Avanzo, H. von Lewinski, and L.

by the Supply Chain Council in the N. Van Wassenhove, 2003], which
United States found that, on an aver- analysed the link between supply
age, organisations spent about 11 per chain and financial performance, re-
im
cent of revenue on supply chain man- vealed that virtually all winning busi-
agement. In contrast, the correspond- ness strategies have, at their core,
ing figure for best-in-class organisa- supply chain strategies that provide a
rL
tions was between 3 per cent and 6 per competitive advantage.

Fo
4 Supply Chain Management for Competitive Advantage: Concepts & Cases
1.1 OWNERSHIP OF THE SUPPLY CHAIN

Based on the definition of supply chain management provided above, it is clear that
multiple organisations are involved in the delivery of the product or service to the
end-user. The successful design and operation of the supply chain thus clearly
y
involves effective management of the interactions between organisations. That raises
nl
the question as to whether there is an obvious owner of a supply chain. In other
words, when we talk about supply chain management, does it implicitly assume a
O
single dominant entity that manages the entire supply chain? Although one often
thinks of a manufacturer as the dominant entity driving the operations of a supply
n
chain, coordinating its activities with its suppliers and distributors, this is often not
the case. In some cases, a large retailer could be driving the supply chain in a
io
significant way and in other cases, a powerful supplier could be controlling the
evolution of a supply chain.
at
It is therefore helpful to look at the gamut of possibilities, and to look at each
ul
entity (suppliers, manufacturers, distributors, retailers and even transporters and
warehouse operators) in general, to understand how supply chains function. It should
irc
also be evident that an entity could be part of more than one physical supply chain
and some of the effectiveness of supply chains could be influenced by the asset
utilisation policies of this entity. This aspect is explored in subsequent chapters.
C
d
1.2 SUPPLY CHAIN DRIVERS

ite
A number of forces, or drivers, have played a role in the evolution of supply chain
management strategies over time. A few important drivers are listed and discussed
below.
im
1.2.1 Technology
rL
Two technological developments that have had a big impact on supply chain
structure and operations are information and communication technology, and
Fo
flexible manufacturing technology. Information and communication technology

allows a host of possibilities simply not imaginable even a decade ago. Flexible
manufacturing technology facilitates efficient small batch manufacturing, greatly
enhancing the ability to customise product offerings to match demand in a better
way.
We must, however, also acknowledge that logistics and supply chain management
practices are significantly affected by technologies based on economies of scale that
continue to emerge. One example is process technology, specifically in oil refining.
Today’s refineries are bigger than ever before, and this has implications on the
logistics of movement of crude oil and finished products, and batch sizes in
y
production. Another example is of ships that carry bulk or container cargo across the
nl
ocean. The most cost efficient of them carry huge payloads, requiring very large batch
handling at ports (not all ports can receive such ocean liners). The attendant large
O
batch sizes for such operations present the inevitable synchronisation problems with
the rest of the supply chain.
n
io
1.2.2 Competitive Factors
at
The major competitive factors for supply chains typically relate to cost and
availability measures. Availability is related to inventories; organisations generally
ul
agree that they carry more inventory than they need to, and that lead times are
significantly more than the technological minimum. Surprisingly, organisations that
irc
focus on managing inventories end up with reduced inventory related costs, along
with improved stock-out performance, due to faster access to markets and other
process improvements.
C
d
1.2.3 Business and Social Environment

ite
While competition is severe, it is now recognised that the battleground has shifted
from competition between organisations to competition between supply chains.
Members of the supply chain are now more receptive to the idea of collaboration.
im
That said, such collaboration is not always successful, but it has led to a different
model of business. The other influential change is that of continuous improvement.
rL
1.2.4 Policy and Regulation

Fo
Many policy and regulatory acts have a direct bearing on supply chain initiatives. For
example, the phasing out of central sales tax in India and replacement with value
added tax regime will have an impact on distribution and retailing strategies of
companies, including warehouse location. Similarly, industry initiatives in standards
of technologies such as Radio Frequency Identification Device (RFID), container
dimension standards, and software protocols and standards, do affect supply chain
practices.
While the above four drivers have played a significant role in the evolution of
supply chain management practices, probably the biggest driver for change has been
the shift in focus from producer-centric supply chains to customer-centric supply
chains. This shift, with all its ramifications, is discussed in greater detail below.
y
nl
1.2.5 The Move from a Producer-Centric to a
Customer-Centric Focus
O
It should be noted that supply chains have been in existence for many centuries. The
earliest supply chains were trading supply chains that were predominant until around
n
1750 A.D. These supply chains primarily involved producing goods at one location,
io
and then moving them to other locations where they were sold or exchanged. The
Industrial Revolution that began in Great Britain in the 18th century and spread
at
throughout the world shortly thereafter, however, changed everything. A direct
consequence of the Industrial Revolution was that an economy based on manual
ul
labour was replaced by an economy dominated by industry and machinery. This led
to the next generation of supply chains: the supply chains in the producer-centric era.
irc
During this evolution, from 1880 to 1980, the customer greatly benefited because
the cost of delivering products and services decreased in a dramatic manner.
However, as we will see shortly, supply chains in this era developed numerous
C
practices that are no longer effective in the current era, the customer-centric era.
The customer-centric era is an era of global competition. Many organisations that
d
built their empire based on practices that worked very well in a producer-centric era
ite
are now finding that these practices significantly inhibit their ability to compete
successfully in the customer-centric era. In other words, the tools and techniques
im
used to manage supply chains in a customer-centric era are significantly different

from those used in a producer-centric era. To understand the significance of this last
statement, it is informative to briefly review supply chains in the producer-centric
rL
era.
Fo
1.3 SUPPLY CHAINS IN THE PRODUCER-CENTRIC ERA

A number of developments served as catalysts for the evolution of supply chains in
the producer-centric era. For instance, the utilisation of electric power to drive
machinery was a major force. This one development alone was responsible for most
of the advances that took place during this period. Other key catalysts were the
introduction of airplanes, automobiles, and telecommunication. Industries in the
West flourished during this period. In particular, the U.S. industry flourished during
this period.
An important point to note is that since numerous organisations are involved in a
supply chain, their activities have to be coordinated. If individual organisations in the
y
supply chain undertake decisions that optimise their own operations without much
thought given to the overall supply chain’s operation, the product or service will not
nl
be delivered as effectively with regard to cost, quality or delivery. Unfortunately,
O
coordination between supply chain partners was not a major concern in the
producer-centric era.
In the production-centric era, demand for goods and services often outstripped
n
production capacity. It was a time when the customer was typically willing to buy
io
anything that the manufacturers had to sell. The producers had the maximum clout
in the supply chain, charged what the market would bear, and operated businesses to
at
maximise the utilisation of their own scarce capacity. Organisations were able to run
their business in relative isolation, formulating strategies that optimised their own
ul
operations, with little regard for the effect of decisions on other organisations in their
supply chain.
irc
Many large organisations grew and flourished in the producer-centric era. This
period was perceived as a ‘golden era’ for the U.S. industry. However, this was actually
C
a problematic period in a number of ways. Lack of global competition created, in

effect, domestic cartels in many industries that dictated the price the customer paid
d
for the product. For instance, the big-three auto companies were effectively a cartel
ite
led by General Motors. But cartels, like their more integrated cousins, monopolies,
tend to get inefficient and ineffective because they can afford to be. Although the
U.S. industry, pioneered by individuals such as Henry Ford, had led the world in
im
innovation in its early days, it started to take a back seat after the war. For example, in
the automobile industry, automatic transmission, which first appeared in the 1940
rL
Oldsmobile, was the last major innovation. While there appeared to be an abundance
of popular features promoted with each new model, there was little to distinguish
between the cars that were made in the 1960s through 1973, from their counterparts
Fo
of the 1950s.
1.3.1 The Rise of Batch Production

The desire to maximise utilisation of scarce capacity led to products being produced
in large batches. In academic circles, there was a tremendous amount of research
expended on determining optimal batch sizes, and the concept of an economic order
quantity (EOQ) was born. While large batches increase the efficiency of the resource
producing the batches, it often negatively affects other resources in the supply chain.
For instance, in the automobile industry, batch production resulted in plenty of
inventory of finished cars, as well as work-in-process inventory (WIP) within the
y
factories. Large batches tend to hurt quality since defects found on the shop floor do
nl
not generate a sense of urgency to fix the problem so that it may not recur. This was
because there was plenty of WIP to buffer any production delays and so the real
O
impact of the quality problem was marginalised. Worse, quality problems often
escaped unnoticed and were not caught until after the product was sold. The
n
consumers suffered from poor quality since the cars were often riddled with defects
io
and needed frequent repairs. This topic is further explored in Chapter 2, in the
context of inventory and its affect on the supply chain.
at
Since a natural consequence of batch production is larger inventories of finished
goods, there was a need to find efficient ways of storing and distributing these goods.
ul
Organisations had to wrestle with the problem of how to select the combination of
distribution channels and supply chain partners that would provide the most
irc
profitable operation in the long run. Since distribution was often fragmented,
distribution costs were high. Consequently, a large number of organisations strove to
C
reduce distribution costs by viewing this activity as an integrated, rather than a

fragmented, activity that took place in various parts of the organisation. Hence
distribution experienced increasing importance during this period especially since
d
mass production and economies of scale were the mantras of the day. Logistical
ite
activities were increasingly apparent in U.S. industry. The automakers set up dealer
networks to stock and sell their cars. In a similar manner, the mass producers in other
im
industries fostered the growth of large warehouses to stock finished goods.

rL
1.3.2 Management Structures in the Producer-Centric Era

The emergence of the large industrial corporation owes much to the organisational
structure and management systems developed and implemented by the U.S. auto
Fo
industry during the producer-centric era. The partitioning of organisations into

divisions and sub-units of divisions is largely attributed to Alfred Sloan and General
Motors, and the accounting dominated control systems discussed in the next section
owe much to automobile manufacturing organisations, primarily the Ford Motor
Company in the years after World War II. The core objective for these organisational
and control structures was to design a system in which the people responsible for
managing the smaller units could focus solely on their domain of authority;
moreover, all decisions and actions that crossed domain boundaries were to be the
responsibility of managers covering both domains. Units could be treated as either
‘cost centres’ or ‘profit centres’ and assigned targets. Managers of these units would
manage their domains to achieve and even beat these targets. The result was an
y
organisational structure in which each unit was charged with and focused on
nl
improving its local performance. The logic went that if every unit improved then, ipso
facto, the entire organisation would improve.
O
Consequently, organisations operated with the belief that local optimisation
would result in global optimisation. This belief facilitated management of large
n
complex organisations by breaking them down into smaller units that, in turn, could
io
be broken down into yet smaller units. It was recognised that this would result in
mismatches at the boundaries of the units. However, the costs in time, money and
at
customer satisfaction were smaller than the costs that would result from being totally
disorganised. This system of operation was phenomenally successful in the sense that
ul
most post World War II organisations that were organised around this model grew
their business significantly. Policies, procedures, performance metrics and all other
irc
elements that define organisational structure and function were based on this implied
belief in local optima. Since financial data (actually cost data) was among the few
C
quantitative data that were available and deemed important by accounting

departments, local optimisation came to be understood as cost minimisation or cost
control. This philosophy drove the management of most operational functions such
d
as production, distribution, and procurement.

ite
Thus, during the producer-centric era, organisations were managed typically

through conflicting departmental goals that resulted from the emphasis on local
im
optimisation. Each department attempted to maximise its own objectives, objectives

that often went counter to the objectives for other departments. Commercial
activities external to the organisation, such as buyer-supplier relationships, were
rL
essentially adversarial in nature. The conflicting departmental goals and adversarial

relationships between the members in the supply chain hindered the move towards
Fo
building more efficient supply chains.
1.3.3 The Rise of Cost Accounting

With the increase in the size and scope of industrial organisations there was a need for
the owners and managers of these large organisations to obtain information that
could help them undertake better decisions. The discipline of accounting was ideally
suited for this purpose. As defined by the American Accounting Association,
accounting is ‘the process of identifying, measuring and communicating economic
information to permit informed decision making by the receivers of the
information.’ Accounting is aimed at providing decision makers with information on
y
the economics of providing goods or service, allowing them to manage organisations
nl
from remote locations. The railroads, for instance, were examples of the need for
large business organisations to be managed from a different location.
O
Knowing what money was spent by which group and for what purpose was
deemed to be extremely important by the accounting profession. As the Industrial
n
Revolution generated demands on machinery, bankers and other lenders needed to
io
know how the machines they were paying for would perform and help the business
payback these loans. For an organisation that produced a variety of products,
at
determining product costs was important in deciding whether the product was
profitable or not. Products costs are typically composed of three components: direct
ul
labour, direct material and overheads. The first two were relatively easy to attribute
directly to a specific product. Overheads, on the other hand were difficult to trace to
irc
specific products, and hence cost accounting focused on tracking these costs
accurately, allocating them to specific products.
C
Oddly enough, while cost accounting owes much to Ford Motor Company in the
days after World War II, it was not a major issue during the days of Henry Ford I.
d
During his reign, overheads were relatively low, and typically only a single product
was being manufactured. Hence, it was easy to allocate the overheads to determine
ite
product costs. Thus, whereas the accounting profession dates back to early trading
networks, cost accounting gained prominence only as organisations diversified their
im
product offerings. Indeed, the post World War II rise of the U.S. auto industry
coincided with the rise of cost accounting as the primary decision making tool in all
industries. The concept of Standard Costs and Allocation was born. Chapter 4 briefly
rL
discusses cost accounting, specifically standard costing and activity based costing.
The discussion in that chapter will serve to highlight the benefits and potential
Fo
drawbacks of these techniques in the context of supply chain management.
1.4 FINANCIAL MANAGEMENT AND RETURN ON INVESTMENT

Even as cost accounting was gaining ground, the management of most large
industrial organisations still resided with the entrepreneurs who founded the
organisations. While they reviewed the accounting information, they did not manage
solely by these numbers. Identifying and seizing opportunities were the main vehicles
for them to make money. A new tool for evaluating opportunities was required.
Pierre Du Pont refined the technique of Return on Investment (ROI) to manage the
y
Du Pont Powder Company, one of the first modern American manufacturing
organisations. Du Pont eventually succeeded Durant as president of General Motors
nl
and collaborated with Alfred P. Sloan to reorganise General Motors. The two men
O
surmised that General Motors, with its multiple organisations, required a centralised
organisational structure similar to that in use at Du Pont Powder Company. Therefore,
they crafted a plan to structure General Motors as a collection of autonomous operating
n
divisions coordinated by a strong general office. Under the reorganisation, General
io
Motor’s general office borrowed the ROI methods from Du Pont Powder Company to
evaluate the financial efficiency of their processes and product lines.
at
The multidivisional, decentralised structure developed by these two entrepreneurs
serves as the model for many organisations to this day. Indeed, the ROI model
ul
developed by Du Pont provides an effective framework for determining how to
irc
Figure 1.1: The Du Pont Model for Leveraging Return on Investment
C
d
ite
im
rL
Fo
leverage an organisation’s competitive advantage into effective shareholder value even

in the customer-centric era. The figure below captures the Du Pont model for ROI.
Essentially, the model shows the immense value of effective supply chain and
operations strategies and how these strategies translate to a competitive advantage,
paving the way for improved ROI.
y
As shown in this figure, effective supply chain and operations strategies can address
nl
all the elements presented on the extreme right hand side. These strategies will
O
translate to improved attributes such as materials costs, conversion costs, quality,
delivery, etc., leading to three important financial measures: Sales, Expenses, and
Asset Use. Given sales and expenses, it is possible to calculate the return on sales as
n
simply (sales – expenses)/sales, which is expressed in the above figure as profit/sales.
io
The asset turns is calculated as shown in the figure. Multiplying return on sales by
asset turns results in return on assets or return on investment (ROI). In Chapter 4, we
at
provide a case study in which we guide the reader through a detailed numerical exercise
that demonstrates the affect of good supply chain management practices on ROI.
ul
We next discuss the customer-centric era and indicate why the management
structure developed in the producer-centric era is inadequate to address the needs of
irc
the customer-centric era. However, the Du Pont model, which was developed during
the producer-centric era remains valid in the customer-centric era as well, as we have
C
noted earlier.
d
1.5 SUPPLY CHAINS IN THE CUSTOMER-CENTRIC ERA

ite
The advent of the customer-centric era took place in the early 1960s, arguably during
the heyday of the producer-centric era. This period, coincidentally enough, also
im
marked the transition of the world economy from a U.S. dominated economy to a
more global economy engendered, in large part, by the Japanese automakers,
especially Toyota.
rL
In 1960, Japan’s auto manufacturers produced about half-a million vehicles. That
same year America’s auto industry built eight million vehicles. Who could have
Fo
predicted that just 20 years later the Japanese auto industry would produce more than
11 million vehicles—three million more than their American counterparts
manufactured in that same year—and become the world’s leading auto producer for
the next 10 years? And how did the Japanese manufacturers manage to overtake their
American counterparts? The answer was that they adopted a number of management
practices that were quite different from their American competitors.
In fact, the 1960s were pivotal years for Japanese manufacturers. They dedicated
themselves as an industry to developing compact and sub-compact cars with excellent
fuel economy. They succeeded in producing those vehicles at low cost by adopting
the just-in-time (JIT) production management philosophy and system. The JIT
management system had its roots in Toyota Motor Corporation, and owes its
y
conception and subsequent development to two individuals, Kiichiro Toyoda and
nl
Taiichi Ohno.
O
Kiichiro Toyoda’s goal was to initiate a production run when it was needed, rather
than making a production run in anticipation of a demand. This was the pull method
of production where a demand pull initiated a production run. The pull emphasis
n
was a fundamental departure from traditional manufacturing methods that pushed
io
production throughout the supply chain based on demand forecasts. He then began
convincing suppliers to cooperate with his just-in-time system. Kiichiro Toyoda also
at
changed the traditional physical layout of the plant so that machine tools were
organised in a flow line. That made the supply line shorter and meant parts could get
ul
into the assembly process sooner. Taiichi Ohno embellished the techniques pioneered
by Kiichiro Toyoda to make cars in small batches more efficiently than the big U.S.
irc
enterprises were able to.
A number of the TPS concepts used by Ohno were inspired by another U.S.
C
automaker who had played a major role in the producer-centric era. This was Henry
Ford, whose relentless attention to detail were greatly admired by Ohno. Like Ford,
d
Ohno emphasised waste reduction. However, Ohno was also able to build on Ford’s
ideas. For example, Henry Ford had offered automobiles to his customers in only one
ite
color to reduce setup times when changing between paint colors. The Japanese were
able to utilise Ford’s ideas to provide product variety in their offerings.
im
Although Japanese autos had now become competitive against both U.S. and other
foreign cars in quality and cost, the U.S. market was reluctant to accept these cars.
rL
They were perceived to be too small and of suspect quality. It took an unanticipated
international crisis to serve as the catalyst for the acceptance of Japanese autos in
America. The Arab-Israeli war in 1973 and the accompanying oil price increases
Fo
caught the U.S. consumers, who favoured large cars with large engines, by surprise.
The ‘gas guzzlers’ traditionally built by the Big Three lost their appeal and Japanese
cars were suddenly in demand. Suddenly the world’s largest market was clamouring
for the small, fuel-efficient Japanese cars. Japanese cars flooded the U.S. market. The
Japanese cars also found acceptance in the U.S. market because consumers quickly
found much better quality than they had learned to expect from automobiles. This
was due, in no small measure, to a handful of individuals such as Edwards Deming

and Joseph Juran who had spread the quality movement from the U.S. to Japan.
Although the advent of the customer-centric era was arguably during the early
1960s, this era really started to manifest itself in the early 1980s. The 1980s also
y
provided many organisations with the power of new management methods and
techniques, such as those pioneered by Taiichi Ohno. However, most organisations
nl
were either reluctant to embrace these methods, or were ignorant of them.
O
1.5.1 New Business Environments Need New Management
n
Systems
io
When it was first introduced, cost accounting was a powerful tool since it provided
managers with the ability to make decisions that dramatically improved the
at
performance of their areas and plants. However, as pointed out in [Lepore and
Cohen, 1999], ‘powerful solutions tend to make themselves obsolete … some of the
ul
basic assumptions of cost accounting became invalid in the 1940s, but as most
companies were using the same concepts, the negative impact was not so noticeable.’
irc
Lepore and Cohen argue that the Japanese did not, and still do not, use cost
accounting, and that those who tried to compete with them using cost accounting
C
methods were forced into breaking the cost accounting rules in order to survive.
The standard cost accounting system was originally developed as a financial tool for
d
controlling diverse activities. While industrial and mechanical engineers were

developing standards for the discrete activities within production processes,
ite
accountants were working to create standard costing principles to allocate overhead

costs. These indirect costs were typically allocated evenly across all the products.
im
When product lines and indirect production costs were small, the distortions created
by this simple allocation technique were not critical. Moreover, the market demand
was enough to consume all products produced, allowing producers a considerable
rL
control of product prices.

Nonetheless, during the past decade the business landscape has changed
Fo
dramatically:
➣ Consumers, not producers, have become the dominant component. Today
the production capacity is greater than product demand for the majority of
industries. Prices are determined by competitive market forces and not set by
the producer. The purely internal focus of the standard cost system is no
longer valid.
➣ The local optimisation view encouraged by the cost accounting model did
not often result in global improvement. Quite to the contrary, actions taken
to improve the cost performance at one stage actually created more costs
elsewhere (but this was of no concern to the manager of that stage), creating
higher inventories and other inefficiencies.
y
➣ Factors such as reliability of deliveries, quality of products, speed with which
nl
orders can be filled and variety of products offered are the dominant factors
O
determining a company’s competitive position. None of these are considered
in the standard cost accounting model. Many of the actions required to
improve performance against these elements appear to be bad decisions from
n
a standard cost perspective.
io
➣ Direct labour costs, on which most allocation decisions were made, was no
longer the major cost component. Starting with the end of World War II, the
at
typical manufacturing company has added far more indirect labour
employees than direct labour workers. It has added layers of personnel to
ul
handle management information, to design product and support the
production process, and to provide the marketing, sales and services to
irc
customers.
We cannot deny the accomplishments of the producer-centric era, and batch
C
production in particular, in making a very wide variety of products available to a very

large segment of the population. However, the core management system developed to
d
support batch production was out of sync with the business environment that
ite
prevailed in the customer-centric era. The local task improvement view at the heart of
the management style in a producer-centric era must be replaced by a new approach
that takes a systems view of the organisation and strives to deliver on the three factors
im
of speed, variability and flexibility. As we will observe in subsequent chapters, a

number of these problems are addressed with new concepts and techniques embodied
rL
in the management of supply chains in the customer-centric era. We will close this
chapter by presenting examples of supply chains that are functioning very effectively
in the 21st century.
Fo
1.6 EXAMPLES OF WELL-RUN SUPPLY CHAINS
We start with a number of examples involving organisations, primarily drawn from

India and the U.S., that have achieved sustained success managing their supply
chains. As we present these stories, we identify the key concepts that underscore these
successes, and the chapters where these concepts are presented and discussed in more
detail.
Even as we present these examples, the reader should note that there is no universal
concept or approach that guarantees success. Indeed, some of these examples might
y
even seem, at first sight, to present contradictory approaches to supply chain
management. We can, however, assert that achieving sustained success requires
nl
significant effort. It requires a basic understanding of the macro-environment, the
O
industry, the technology, and a variety of other factors.
n
1.1
io
The Dabbawallas
at
Even as the producer-centric era tended to drive organisations to look for large
distribution networks with large warehouses that tended to store goods for long
ul
periods of time, a quiet revolution in distribution activity was taking place in
Mumbai, India, around the turn of the 20th century. In the 1890s, Mumbai was
irc
growing rapidly due to migrants from various parts of India coming to Mumbai
seeking work. These migrants found that there were no restaurants they could go to
C
for their lunch, resulting in a real need for a method to deliver freshly prepared food
to the workplace. The clichéd phrase, ‘necessity is the mother of invention,’ aptly
describes what ensued: the dabbawallas.
d
The dabbawalla is a person whose job is to carry and deliver food prepared fresh,
ite
that day at home, to an office worker in a lunch box. They use an intricate delivery
system that assigns each lunch box (‘tiffin box’) a code. The dabbawalla places the box
im
in a large transport container (‘pallet’). The coding system is now used to move these
tiffin boxes through the metropolitan railway and bus transport network, using
possibly handoff points ‘stations’. At a given station, the tiffin box could be taken out
rL
of its pallet and placed on a different pallet to be transported by a different

dabbawalla, depending on its destination, to possibly another station. The tiffin box
Fo
could thus conceivably change hands three or four times before reaching its final
destination. This intricate set of activities involving multiple splits and merges is
essentially the idea behind cross-docking – a technique currently used by modern
logistics providers to transport goods across large distances using sophisticated
transport and communication systems, to track the progress of shipments in the
supply chain.
What is remarkable about the dabbawallas is that this system of operation has
performed so well over the past 100 years, even as the world transitioned from a
producer-centric era to a customer-centric era. The explanation is that the goal of the
dabbawallas was always focused on delighting the end-customer. Remarkably, over
the years the dabbawallas have been able to transport the tiffin box from origin to
y
destination without the use of any communication device. Furthermore, their
nl
delivery accuracy is remarkable as well. Their error rate is reportedly just one in 16
million deliveries, resulting in the Forbes Global magazine awarding them its Six
O
Sigma certification in 2001. According to Forbes the dabbawalas have a 99.999999
per cent accuracy.
n
io
1.2
at
The National Dairy Development Board
ul
A comprehensive example of supply chain integration with a focus on quality is the
case of the National Dairy Development Board (NDDB) in India. NDDB began
irc
operations in 1965 initially focusing on milk procurement and processing, but over
fifty years, it has forward-integrated the chain to include distribution, value added
C
Figure 1.2: Cash Flow Chart for NDDB

d
ite
im
rL
Fo
products, and retailing. It has backward-integrated into animal husbandry, animal

feed, and packaging.
Large dairy operations of this size are usually run with organised cattle rearing and
milk production. In contrast, the operation of the NDDB supply chain is based on
y
the voluntary supply of milk by hundreds of individual cattle owners, up to twice a
day, on a regular basis. The NDDB operates in a co-operative mode, with cattle
nl
owners having a stake in the organisation.
O
Figure 1.3: Milk (Product Flow) at NDDB
n
io
at
ul
irc
C
d
A key element in NDDB’s supply chain relates to cash flow management. The cash
ite
flow logistics in this supply chain is unique in the sense that payment is made to cattle
owners the day after they deliver the milk, based on a quality assessment of the milk
im
supplied. This quality test (based on fat content) is quick, transparent and acceptable
to the cattle owner and to NDDB. Paying individual cattle owners the very next day
is crucial to the cattle owners’ working capital management. Such a payment
rL
mechanism has been developed by NDDB over time. NDDB has also played a vital
role in the development of inputs to the cattle sector to increase yields. Chapters 6, 7
and 8 discuss some key concepts driving procurement, distribution and logistics.
Fo
1.3
Asian Paints
Asian Paints (AP) is India’s largest paint company and the third largest paint
company in Asia today, with a turnover of Rs. 36.7 billion (around US$ 940
million). AP tries to gain a competitive advantage by providing quick and reliable

delivery for their fast-moving paints products through an integrated supply chain. AP
owns three layers of the supply chain – the paint factories, the regional distribution
centres and the sales depots. A significant competitive edge is provided by AP’s
‘postponement strategy,’ which allows it to meet varying customer demands with
y
minimal inventory in its supply chain. Such a strategy mixes the colours and base
nl
paints at the retail site itself to get the exact shade of colour as per customer
specifications, resulting in lower retail inventory costs while, at the same time,
O
significantly increasing product availability.
n
Figure 1.4: Flow of Product from Retailer to Customer (Asian Paints)
io
at
ul
irc
Customised Option (after):
C
d
ite
im
rL
Fo
To understand how the postponement strategy works, consider a product portfolio

consisting of 10 base paints with 50 shades in each, resulting in about 500 full form
Stock Keeping Units (SKUs). If the retailer had to stock each SKU, it would result in
a very high inventory cost. On the other hand, if the retailer tried to minimise
inventory costs by only stocking the SKUs that were in high demand, there would be
a likelihood of losing some potential sales. By adopting the postponement strategy,
the retailer only needs to stock the 10 base paints along with about 10 chemicals,
providing the same variety with appropriate blending, but by stocking only 20 SKUs.
No doubt, the retailer has to spend about five minutes more to process each order.
However, since paint is not an impulse buy, people are willing to wait or even come
back later in the day since they are assured of getting their exact choice of colour. The
y
blending equipment has to be available at retailers for this strategy. Another level of
nl
customisation allows the customer to specify the colour with the help of a sample.
Through an optical reader, the menu for appropriate blending is developed and the
O
desired colour is provided to the customer. Chapter 2 discusses the principle of
postponement strategy in more detail.
n
io
1.4
at
Hindustan Unilever Limited
ul
Hindustan Unilever Limited (HUL) is India’s largest consumer products company.
Founded in 1933 as Levers Brothers (India) Ltd. it is a key player in the fast moving
irc
consumer goods (FMCG) industry in India. HUL has generally sought competitive
advantage through a continuous re-engineering of the supply chain.
C
HUL is using different strategies to address its three major market segments: the
‘modern trade’ segment (the organised retail sector), the general trade segment and
rural markets, creating dedicated teams to address the three segments. For example,
d
in the rural market segment, a big opportunity lies in creating consumers out of non-
ite
users. The modern trade segment is driving HUL to focus on adding value by cutting
down lead times. This segment is of large retailers and a forward urban market.
im
Customers in this segment are usually serviced through a HUL warehouse that
supplies to a cluster of retail locations. This approach minimises direct delivery to
retail stores (except for perishables), allowing the retailers to concentrate on front-
rL
office and customer-focused operations. Back office, warehouse-related operations

such as receipts, sorting and repackaging are done at the warehouse. If the volumes
are large enough, HUL sends items directly from their branch, which acts as a
Fo
Carrying and Forwarding Agent (CFA) to the retailers’ warehouse (saving a 3–5 per
cent margin that the distributor would have, otherwise, charged). In fact, where
volumes are large enough, the large retailer is appointed as a ‘CFA’ that holds the
stocks in a designated area at the warehouse, with ownership remaining with HUL.
Once or twice a day, the stocks are transferred and money transfers are made and
accounted for, accordingly. The typical margin for the CFA of half to one per cent
would then be saved.
Figure 1.5: Product Flow of FMCGs (HUL)

Before:
y
nl
O
After:
For Large Sales:
n
io
at
ul
(Saving of about 1.5 per cent over ‘Before’)
irc
For Very Large Sales:
C
d
ite
(Saving of about 3.5 per cent over ‘Before’)

im
Initiatives such as these facilitate the operation of Vendor Managed Inventory

systems at customer warehouses. The general principle of vertical integration
rL
operates here in the sense that, for the particular market segment affected, the supply
chain is shortened (in this case, by eliminating the company branch/CFA and the
distributor stage). Another principle of horizontal differentiation operates, in the
Fo
sense that there are different modes of supply for organised retail and for smaller
independent retail segments.
A related and significant change is that the role of the sales personnel is now
different in this configuration. Sales personnel associated with large customers are
more relationship managers, dedicated to that customer and would do much more
than traditional order placement and normal commercial functions. Procurement
managers of the large customer can interact with the salesperson, allowing for
initiatives such as test marketing, new product placements and co-ordination with
brand managers of the manufacturer.
At a different level, HUL operates with a ‘systems thinking’ philosophy. It has a
y
management cadre that is not departmentalised. Since people move across functions,
nl
there is not much of an ‘us versus them’ mind set within the organisation. Instead,
local departmental objectives are treated in the context of the broader supply chain
O
objective. In Chapter 2 we discuss systems thinking in the context of supply chain
dynamics and in Chapter 3 we discuss basic concepts underlying supply chain
design.
n
io
1.5
at
ABC Bicycle Company
ul
The concept of the supply chain has changed significantly for ABC bicycle company,
an organisation in south India. This company began operations as a factory that
irc
sourced raw materials, made components, assembled bicycles, with warehousing, and
delivery to dealer. The focus has changed to sourcing, kitting, warehousing, delivery
C
to dealer, and assembly. Although the factory has lost some prominence as a result,
the supply chain is leaner and more ‘linear’. ABC Bicycle Company now focuses
d
more on marketing, design, and quality control.

Originally, ABC operated a typical factory. It received raw materials, performed
ite
component manufacturing and assembly, and shipped finished products. It adopted

a systematic vendor development strategy, starting with standard items such as
im
components (e.g. carriers) that were outsourced. While the main driver was lower
overhead costs for the vendors, ABC also realised that vendors had different expertise,
rL
for example, in working with different materials (steel, rubber and plastic).
Although ABC’s plant was in Southern India, the largest cycle manufacturing
centre in India was in north India, in Punjab. Even with the distance, it became
Fo
profitable to procure components from the industry cluster in Punjab. One of the
larger outsourcees was also developed for kitting (for a particular model, it would
collect items such as ball bearings, toothed wheel, pedals, chains, which would be
issued as a kit to the shop floor). The factory in the south became a second level
kitting centre. Over a period of time, the high precision components and the large
components earlier which were outsourced locally (handle bar, main frame, mud
Figure 1.6: ABC Bicycle Company (Usage of Supply Chain Concepts)
Before:
y
nl
O
n
io
After:
at
ul
Sourcing
irc
Dealer
Kitting Warehousing
(Assembling)
C
d
guard, wheel rim) were eventually also outsourced to Punjab, as per designs of ABC
ite
Company. Only painting and stickering was done at the factory.

Another motivating factor driving the move towards kitting was the observation
im
that a kit occupied less volume in a truck compared to a fully assembled bike.
Furthermore, retailers were getting more sophisticated in assembly. So it was decided
to send bicycles to the retailers Semi Knocked-Down (SKD) kitted condition.
rL
Retailers moved to having SKD packs on display and only demonstration models on
the floor.
Fo
Then, it was found that 60 per cent of value and weight of the bicycle was coming
from north India, and the company did operate in northern India markets.
Therefore, second level kitting centres were developed in West India and North
India, that would receive kits from vendors, prepare the SKD kits and then dispatch
to retailers in that region. Lean operations are discussed in Chapter 5. Quantitative
models for locating facilities are discussed in Chapter 10.
1.6
Dell Computers
It is probably widely known that Dell Computers is a pioneer in lean supply chain
y
management practices. Dell recognised that the market segment is (i) computer
nl
savvy, (ii) customisation sensitive, and (iii) price sensitive, and has designed its supply
chain accordingly. It offers customers who buy desktops a two-day delivery at a low
O
price by leveraging online (remote) ordering, assembling to order, and express
delivery. The strategy that has sustained Dell so far can be summarised as mass
customisation and a relentless focus on costs. Although Dell has come under
n
increased scrutiny, especially as its competitors such as Hewlett Packard have
io
emulated the famous ‘Direct’ model, the reader must keep in mind that Dell is a
pioneer in mass customisation and cost reduction.
at
Figure 1.7: Dell Computers (Mass Customisation)
ul
irc
C
d
ite
im
rL
Dell is able to provide mass customisation because it maintains very low

Fo
inventories, with suppliers delivering components to the assembly operations on a

just-in-time basis. This practice has led many to speculate that Dell would be hard hit
by unexpected disruptions, for instance, the 2002 port strike in the western coastal
region of the United States. Dell’s ability to provide mass customisation, however, has
provided it with the same ethos of speed and flexibility to deal with such disruptions.
Through its constant communication with its U.S.-based shipping partners and its
parts makers in Taiwan, China and Malaysia, it was aware of the possibility of a
lockout some six months before it occurred. When the strikes became a near
certainty, Dell chartered eighteen 747 aircraft from UPS, Northwest Airlines, China
Airlines, and other carriers, maintaining its costs to about US$500,000 per plane,
and was able to keep its supply chain moving. See Chapters 3, 5, 7 and 8 for a more
in-depth discussion on concepts in lean supply chain practice, lean operations,
y
logistics, warehousing and distribution.
nl
O
1.7
Tata Motors
n
Tata Motors Ltd. is India‘s largest passenger automobile and commercial vehicle
io
manufacturing company, with about a 60 per cent market share in the commercial
at
vehicle segment. It is also the world’s fifth largest commercial vehicle manufacturer. A
key initiative in Tata Motors’ supply chain management strategy is the ability to
ul
Figure 1.8: Tata Motors Ltd. (Third Party Logistics Provider)
irc
Before:
C
TATA Motors Ltd.

Sourcing
Kitting
ite
TATA—Inventory
im
After:
rL
Fo
involve supplier participation at an early stage of its new product development cycle,
an initiative that resulted in its winning a supply chain excellence award in 2005. This
was a huge change in mindset for a vertically integrated company that believed in
developing everything in-house. It is urging the suppliers to adopt a holistic
approach, encouraging them to think about the overall vehicle and how their product
y
fits into the overall plan.
nl
Another key element in Tata Motor’s strategy is the development of third party
O
logistics providers to source, kit and deliver components. This initiative has
significantly improved the inventory levels and inbound logistics costs for Tata
Motors. See Chapter 8 on logistics for a brief discussion on third party logistics.
n
io
1.8
at
Barilla SpA
ul
Barilla SpA is a large Italian food company, founded in 1877 in Parma, Italy. A major
food product for Barilla is pasta, a staple food for Italians. Barilla is the world’s
irc
leading pasta maker, accounting for 40–45 per cent of the Italian pasta market and
for 25 per cent of the United States pasta market. It has manufacturing plants in Italy,
C
Greece and the United States.

Many people are familiar with the JIT philosophy for manufacturing that we
d
discussed in section 1.5. Barilla is credited with the introduction of this philosophy
to its distribution operations in the 1990s—the concept of just-in-time distribution
ite
(JITD). The JITD concept was introduced by Barilla to combat a number of

operational inefficiencies and cost penalties that Barilla was facing during the late
im
1980s, as a result of large week-to-week variation in demands placed by its

distributors on its manufacturing facility. Much of this demand fluctuation was due
rL
to the distributors artificially inflating the true customer demand (which had a
relatively flat demand pattern) for a variety of reasons. We will discuss these reasons
in more depth in Chapter 2 when we discuss a phenomenon often referred to as the
Fo
bullwhip effect. For the moment, we will provide some insight into why such demand
fluctuations take place. For example, the distributors were prone to stockpile
inventory in anticipation of a price hike. Alternatively, whenever Barilla had a
promotion for its products, (perhaps in response to its competitors offering a
promotion) the distributors would stockpile inventory to avail of the price discount.
To cope with these problems, Barilla introduced the JITD system to determine
what products it would ship to its distributors. Under the JITD system, rather than
simply filling orders specified by the distributor, Barilla monitors the flow of its
product through the distributor’s warehouse, and then decides what to ship to the
distributor and when to ship it. The JITD implementation allowed Barilla to better
y
control its supply chain, providing it with a significant competitive edge.
nl
O
1.9
Benetton
n
Benetton is an Italian clothing manufacturer that serves a global market. The textile
io
and apparel industry has to deal with short product cycle for fashion articles, long
at
production lead-time and forecasting errors for fashion items. Every season, Benetton
used to follow the traditional way of making sweaters and hosiery: dye the yarn (fix
ul
the garment colour) and then knit the fabric (fix the style). However, it was never
clear until well into the season about which colours would be the best sellers. That led
irc
to a buildup of unwanted inventory that had to be cleared through huge markdown
sales—a chronic problem that plagues many organisations in the textile and apparel
industry. Such markdowns also promoted the bullwhip effect that we discussed in the
C
Barilla example.
d
Figure 1.9: Benetton (Technology Development)

ite
Before:
im
rL
After:
Fo
Benetton decided on a novel way to mitigate markdowns by understanding

customer value. A systematic study of customer behaviour showed that it was easier to
predict style choices than colour choices. Consequently, Benetton evolved a new
process that changed the production sequence for their single colour fabrics to do the
knitting first followed by the dyeing operation. Such a simple application of the
principle of postponement resulted in a huge return for Benetton. By delaying the
point of product differentiation until better demand patterns could be established,
y
Benetton was better positioned to align its supply chain with true customer demand.
nl
As remarked earlier, the bullwhip effect is discussed in Chapter 2 and the principle of
postponement is discussed in Chapter 3.
O
n
1.10
io
Zara
at
Zara is a Spanish clothing manufacturer/retailer whose supply chain strategy is to set
the industry standards for time to market, costs, order fulfillment and customer
ul
satisfaction. At the heart of this organisation’s success is a vertically integrated
business model that spans design, just-in-time production, marketing and sales. This
irc
model gives Zara more flexibility than its rivals to respond to fickle fashion trends.
Unlike other international clothing chains, Zara makes more than half of its clothes
in-house, instead of relying on a network of suppliers. Zara has adopted a number of
C
lean supply chain techniques. For instance, it acquires fabrics in only four colors and
delays committing these fabrics to the dyeing and printing operations until the last
d
stage of production. By delaying commitment of the fabric to special colours, Zara

ite
substantially reduces the markdowns plaguing the textile and apparel industry as we
discussed for Benetton above. Zara keeps designers attuned to changing customer
preferences. Its sales managers send timely customer feedback from its 450 retail
im
stores to in-house designers. As a result of better-managed inventories, reduced

obsolescence, and tight linkages between demand and supply, Zara is well positioned
rL
to gain market share.

Fo
1.11
Cement Industry in India

The cement industry has changed its production structure to enable greater flexibility
in transportation (use of open wagons instead of only covered wagons, coastal
transportation), by making cement in two stages, first as clinker near the raw material
Figure 1.10: Cement Industry (Transport of Cement)
Before:
y
nl
O
n
io
at
ul
irc
After:
C
d
ite
im
rL
Fo
source, followed by grinding, blending and bagging near the market. The key
elements of this strategy are flexibility and postponement (see Chapters 2, 3, 5).
1.12
y
nl
Amazon
Amazon.com has developed a customised supply chain to manage its demand
O
planning, inventory planning, warehousing, transportation, inbound, and outbound
shipping operations. Its business model has facilitated the automation of every step
n
of the supply chain process, with an ability to update the information real-time. Such
a strategic advantage allows Amazon.com a unique advantage of locating the optimal
io
distribution centre from which to pick and ship inventory within minutes of a
at
customer’s order. The customer is notified, almost immediately, on how long it will
take before the order is delivered to the customer. Amazon uses its information
ul
system to offer the customer recommendations on purchases based on a profiling of
the customer. Some key elements in developing a sustainable supply chain strategy
irc
for Amazon were customer profiling through CRM and value addition.
Figure 1.11: Amazon (Value Addition)

C
d
ite
im
rL
Fo
1.13
Pizza Hut and Domino’s

Pizza Hut traditionally had eat-in and take-away channels. Domino’s entered the
y
market with delivery and take-away channels, providing more convenience at a lower
price. Pizza Hut was forced to respond with a delivery channel. Corres-pondingly, to
nl
stay in the mind space of their customers, Domino’s also opened eat-in restaurants at
O
many locations. The main elements in this supply chain competition are to do with
an appropriate combination of product plus service.
n
Figure 1.12: Pizza Hut (Addition of Service)
io
Before:
at
ul
irc
C
d
ite
After:
im
rL
Fo
1.14
IT Hardware Manufacture
IT Hardware manufacturers are continuously restructuring their distribution
y
network to enable a responsive supply chain for products, spare parts, and repair and
return. Third party logistics service providers have played a significant role here. The
nl
IT hardware industry relied on spare parts logistics, reverse logistics and third party
O
logistics to sustain their supply chain effectiveness.
n
io
Supply chains have existed for centuries in various forms, starting with the trading supply
at
chains that were predominant through circa 1750. These supply chains primarily involved
producing goods at one location, and then moving them to other locations where they were
ul
sold or exchanged. Supply chains in the producer-centric era evolved following the
Industrial Revolution and were propelled by advances in power generation methods and
irc
advances in transportation. The next hundred years witnessed an accelerating trend in the
evolution of supply chains. A number of catalysts assisted this trend: electric power,
C
airplanes, automobiles, and telecommunication. This period also brought forth a number
of illustrious thinkers and doers who elevated supply chain management to new heights.
These individuals include Henry Ford, Sloan, Kiichiro Toyoda, and Taaichi Ohno. It was
d
an era in which manufacturers were able to charge what the market would bear and industry
flourished during the initial years of this period, especially in the United States. Towards
ite
the end of this period, however, global competition threatened their virtual monopoly in
manufacturing. With increased foreign competition, capacity also expanded and the world
im
economy began to move towards a customer-centric economy.

It was probably a coincidence, but the enablers of efficient supply chain management
appeared to take root right around this time. Indeed, by the end of the 1980s, there were a
rL
number of enablers of supply chain management in place. These include but are not
restricted to the following: the internet, lean thinking, six sigma, mass customisation,
organisation resources planning systems, disintermediation: the ‘direct’ model that works
Fo
around the method of selling via the traditional distribution channels, and software tools
for better information management.
Copacino, W. C., (1997), Supply Chain Management: The Basics and Beyond, Boca Raton:
St. Lucie Press.
y
Chopra, S. and P. Meindl, (2006), Supply Chain Management, Third Edition, Prentice-Hall.
nl
D’Avanzo, R. H. von Lewinski and L. N. Van Wassenhove, (2003), The Link between
Supply Chain and Financial Performance, Supply Chain Management Review,
O
November/December.
Lepore, D. and O. Cohen, (1999), Deming and Goldratt: The Theory of Constraints and the
n
System of Profound Knowledge, Great Barrington, MA: The North River Press.
Raghuram, G. and N. Rangaraj, (2000), Logistics and Supply Chain Management: Cases and
io
Concepts, Macmillan.
at
Simchi-Levi, D. P. Kaminsky and E. Simchi-Levi, (2003), Managing the Supply Chain,
McGraw-Hill.
ul
Simchi-Levi, D. X. Chen and J. Bramel, (2004), Logic Of Logistics: Theory, Algorithms, and
Applications for Logistics and Supply Chain Management, 2nd Edition, Springer-Verlag.
irc
Case references : The case Seth Dhaniram (C and FA) deals with a possible re-structuring of
the supply chain. Chilli in Soup (A) refers to the implications of supply chain structure
C
on quality issues and product responsibility.

d
ite
1. Explain how Technology is an important driver of supply chain management

im
strategies. Use examples from real-world organisations to support your answer.

2. What are the significant differences between supply chains in the producers-centric
rL
and the customer-centric eras? Give at least two such differences.

3. The chapter briefly describes some problems with batch productions, especially with
regard to the resulting high levels of inventory. What are some of the other problems
Fo
that result from producing and transporting large batches?

4. Give at least one example of how the local optimisation mindset could be detrimental
to the overall functioning of the organisation.
5. Discuss how effective operations strategies, for instance a lead time reduction strategy,
can help leverage an organisation’s return on assets.
6. Why do you think the traditional cost accounting systems do not work too well in a
customer-centric era.
7. How did the Japanese auto manufactures manage to significantly erode the
dominance of the US auto manufactures?
y
8. This chapter presented examples of organisations with well-run supply chains. Can
nl
you think of other organisations that have well-run supply chains? If so, identify these
organisations, providing specific aspects of the functioning of these organisations to
O
support your answer.
n
io
at
ul
irc
C
d
ite
im
rL
Fo
Fo
rL
im
ite
d
C
irc
ul
at
io
n
O
nl
y
y
CHAPTER
nl
2
Understanding Supply Chain
Dynamics
O
n
CHAPTER OUTLINE
io
Introduction
at
2.1 Supply Chain Dynamics in Action
2.1.1 Coping with the Dynamics of the Supply Chain
ul
2.2 The Bullwhip Effect
2.2.1 The Beer Game
irc
2.2.2 Analysis of the Bullwhip Effect
2.2.3 Improved Forecasting and POS Data do not Eliminate the Bullwhip Effect
2.3 The Impact of Lead Times
C
2.3.1 Littles Law

2.4 Inventory Management and Supply Chain Dynamics
d
2.4.1 Risk Pooling

2.5 Offshoring and Outsourcing: Effect on Supply Chain Dynamics and Costs
ite
2.5.1 The Hidden Costs in Offshoring Math

2.6 Lessons Learned
im
Conclusions
References
Exercises
rL
Fo
Understanding Supply Chain Dynamics 37
INTRODUCTION*
I
n his path-breaking work on sys lier, and to be discussed in more detail
tems dynamics, Jay Forrester [J. in this chapter. The simulation has
y
W. Forrester, 1958] stated that since been refined and is now played
‘there will come general recognition of as the popular ‘Beer Game’ simulation
nl
the advantage enjoyed by the pioneer- [J. W. Forrester, 1961]. The game un-
ing management who … improve their derscores the importance of under-
O
understanding of the interrelationships standing supply chain dynamics, ap-
between separate company functions plying systems thinking to coordinate
and between the company and its mar- activities within and between organisa-
kets, its industry, and the national tions, the crucial role lead times play in
n
economy.’ Forrester, in effect, was enhancing or inhibiting competitive-
io
stressing the need for managers to un- ness, and the role of information sys-
derstand supply chain dynamics and tems in supply chain management.
adopt a holistic view, or a systems per-
at
This chapter is devoted to under-
spective, to run their business. standing supply chain dynamics. We
Forrester had essentially identified the discuss a phenomenon known as the
key management issues associated
with supply chain management.
ulbullwhip effect, a term we introduced
in Chapter 1, in depth since this phe-
irc
Forrester illustrated supply chain nomenon is widespread and creates
dynamics and its effect on supply considerable difficulties in supply
chain performance using a computer chain management. Understanding
C
simulation. This simulation was devel- the root causes of the bullwhip effect
oped in the early 1960s as part of his will allow us to mitigate the deleterious
research on industrial dynamics and it consequences of this phenomenon.
illustrated the challenges faced in We start the chapter with a case study
d
managing supply chains, including the that demonstrates how this problem
ite
bullwhip effect that we referred to ear- manifests itself in the supply chain.
im
rL
Fo
* A number of sections in this chapter are drawn from the book, Streamlined, by Mandyam
M. Srinivasan. The authors thank Cengage (formerly Thompson-Taxere) for giving us
permission to use this material.
2.1 SUPPLY CHAIN DYNAMICS IN ACTION

In the fall of 2000, Solectron, the world’s biggest electronics contract manufacturer,
knew something was going awry. The major manufacturers of telecommunication
equipment, including Cisco, Ericsson, and Lucent were predicting an explosive
y
growth in demand for networking gear and wireless equipment. These telecom giants
nl
were asking Solectron and other contractors to supply components and raw materials
for their operations as fast as they could, assuring them they would pay for excess
O
materials. Solectron, which supplied each major player, knew that the supplies
demanded by its customers added up to a demand for telecommunications
n
equipment that was unreasonably high, even under a best-case scenario. However, it
was forced to produce at maximum output in order to meet customer demand.
io
Relying on forecasting software that projected continued growth, the top
at
management at Cisco was seemingly oblivious to the problems that Solectron was
observing. In 2001, ‘irrational exuberance’ clashed with reality. The explosive growth
ul
in demand forecast by the software did not materialise. Instead, Cisco’s sales plunged
30 per cent in the third fiscal quarter of 2001. It was forced to write off US$2.2
irc
billion in inventory and lay off 8,500 people. Cisco’s stock sank to less than US$14
from a high of US$82 just 13 months earlier [S. Berinato, 2001]. When demand for
C
their equipment shrank dramatically in 2001, the telecom giants scaled down their
operations sharply and this, in turn, impacted suppliers in a big way. Many suppliers
were left with excessive inventory that had been built in response to their customers’
d
demand forecasts. Specifically, Solectron was left holding US$4.7 billion in

ite
inventory [P. Engardio, 2001].

im
2.1.1 Coping with the Dynamics of the Supply Chain

One could argue that the problems faced by Cisco and Solectron were precipitated by
rL
the dot com implosion. However, the fact is similar experiences are mirrored by
enterprises in almost every industry although arguably not in such a dramatic
manner. These enterprises experience huge variations in inventory levels, orders and
Fo
shipments at each step in the chain, with the variations typically more pronounced as
one moves further upstream. Such demand and inventory variations result in large
inventory holding costs, lost sales from stock-outs, and most importantly a lack of
responsiveness to customer demand. It turns out that much of the demand variation
is caused by the supply chain itself, not by the end customer.
The fast moving consumer goods industry displays similar behaviour. Consider
the production and distribution of diapers. Given the consistency in diaper demand,
it would be natural to expect the diaper supply chain to operate efficiently. Indeed,
when logistics executives at Procter and Gamble examined the demand for its diapers
at retail stores, it found a relatively flat demand. However, the orders Procter and
y
Gamble placed on its suppliers for this product showed considerable variation.
nl
Understanding how supply chains and their constituents behave over a period of
O
time is important because it affects all supply chain partners. Understanding this
behaviour also forms the basis for planning capacities, strategies and operational
decisions for transporters, traders and commercial entities that affect the end to end
n
performance of a supply chain.
io
The pasta maker Barilla SpA that we discussed in Chapter 1 significantly mitigated
the bullwhip effect by implementing a just-in-time distribution (JITD) system that
at
provided Barilla information about the demands that customers made on its
distributors and retailers. That information allowed Barilla to specify where
ul
inventory was to be held in the downstream distribution facilities [J. Hammond,
1994]. Barilla thus controls the flow of physical goods through the supply chain
irc
based on accurate demand information that is not biased by the distribution/retail
centres perception of customer demand and Barilla’s ability to deliver on these biased
C
demand estimates. In effect, Barilla has taken on the vital task of matching the
production and distribution network supply with the end-customer demand. The
d
reader should note that this high degree of control of the supply chain is not always
possible. Rather, Barilla, by virtue of its clout in the supply chain, is one of a relatively
ite
few organisations—such as Dell, WalMart, Hindustan Unilever Limited, etc., that

are able to dictate how the supply chain operates. Even with this level of clout, it was
im
not easy for Barilla to put through its just-in-time distribution system.
The JITD distribution system used by Barilla is a classic example of a pull-based
rL
demand-driven system. A dichotomy of terms used to describe how a supply chain

functions is that of push and pull. The term push is used to denote the triggering of
material flows downstream through planned dispatches, based on forecasts. In
Fo
contrast, the term pull is used to denote the triggering of material flow through
replenishment orders. Generally, cost control and capacity utilisation are primary
concerns in parts of the supply chains where the push philosophy dominates, while
customer satisfaction, inventory problems surrounding shortage and obsolescence,
and effectiveness, are major concerns in the pull part of supply chains. The
understanding of why and how push and pull systems, or a combination of them,
work is important in the design of supply chains. Such understanding helps

determine, for instance, where decoupling points – points at which inventory is
usually held – should be located. This is discussed in the next chapter. In subsequent
chapters, we also see how various entities can plan their operations from a supply
chain perspective. An understanding of the dynamics discussed in this chapter would
y
be crucial for this.
nl
The examples presented above illustrate how minor changes in demand at the end-
O
user or the retail level results in huge variation in demand at upstream enterprises in
the supply chain. This phenomenon, termed the bullwhip effect, owes its origin to the
fact that a slight motion of the handle of a bullwhip can make the tip of the whip
n
thrash wildly at speeds up to nine hundred miles per hour, about 20 per cent faster
io
than the speed of sound, creating a sonic boom. In the context of a supply chain, the
bullwhip effect manifests itself through increasing demand variability as you move
at
upstream in the supply chain. That is, small shifts in the level of customer demand
experienced by the retailer are magnified as the demand information is passed up the
ul
supply chain creating increasingly higher variation in the orders received by upstream
suppliers.
irc
2.2 THE BULLWHIP EFFECT

C
The bullwhip effect results in tremendous inefficiencies of the supply chain. It results
in excessive inventory investment, poor customer service, lost revenues, misguided
d
capacity plans and ineffective transportation and production schedules. Many

ite
enterprises have gained a significant competitive advantage by understanding the

underlying causes of the bullwhip effect and working with their supply chain partners
im
to reduce it. This, in turn, enables them to reduce inventories and become more
responsive to customer demand. A very effective way to demonstrate the underlying
causes of the bullwhip effect is through the Beer Game simulation.
rL
2.2.1 The Beer Game

Fo
The beer game simulates the behaviour of a serial supply chain consisting of four
enterprises engaged in the production and delivery of a single blend of beer: a factory,
a distributor, a wholesaler, and a retailer. Figure 2.1, illustrates this linear
arrangement. The goal of each enterprise is to manage the demand imposed by its
customer.
Each week, an enterprise in the supply chain receives orders from its downstream
customer and places orders with its upstream supplier. At each stage there is a time lag
between having an order and compliance, besides costs for storage and rush orders.
Initially, there is no information sharing beyond what is conveyed by orders and
shipments. All four enterprises in the supply chain have to decide what to order from
y
their upstream supplier based on the orders they receive from their downstream
nl
customer and their inventory on hand. There is a two-week lead time before an order
placed by an enterprise reaches its upstream supplier. Similarly, there is a two-week
O
manufacturing lead time, from the time an enterprise receives an order until a
shipment against this order reaches the downstream customer. In effect, there is at
n
least a four-week lead time from the time an order is placed by an enterprise on its
io
upstream supplier until the time it receives a shipment against this order, as depicted
in Figure 2.1.
at
ul
Figure 2.1: The Bullwhip Effect in the Beer Game
irc
C
d
ite
im
rL
Fo
At the start of the simulation, the system is in steady state with the consumer (end-
user) buying four cases of beer each week, while each enterprise is ordering and
receiving four cases of beer each week. Each enterprise is holding an inventory of
twelve cases of beer. The retailer’s demand is revealed at the start of each week—for
the first few weeks this demand is steady at four cases per week. The demand on the
y
other enterprises is determined by the orders working their way upstream—initially
nl
four cases per week. At the end of each week, each position in the supply chain
decides the number of cases it wishes to order from its upstream supplier.
O
The steady state is now disrupted. For week five, the demand by the consumer
increases to eight cases per week and is held steady thereafter. Even this one-time step
n
change is enough to cause significant problems upstream. As the change in demand
io
propagates upstream, shortages or surpluses accumulate at each stage in the supply
chain. As indicated in Figure 2.1, orders and inventories spike wildly. These spikes
at
become magnified as you move upstream.
ul
2.2.2 Analysis of the Bullwhip Effect
irc
The discussion in this section is based on a paper by K. Gilbert [2003], which
explains how the beer game is able to demonstrate the importance of systems
C
thinking and its impact on the supply chain. The game shows that a locally managed
supply chain is inherently unstable. The systems perspective recognises that if each
element in the supply chain tries to optimise its own operations in isolation, everyone
d
suffers in the long run. In the context of the beer game, each enterprise in the supply
ite
chain makes decision in isolation without input from its immediate upstream and
downstream supply chain partner. Moving from a local optimisation framework to a
im
global optimisation framework poses a tremendous challenge for enterprises, as it is a

radical shift from the traditional approach towards managing an enterprise. System
thinking has been pointed out many times as strategically important, but is still not
rL
sufficiently understood.
Because firms work with limited information, it is thought that a major cause for
Fo
the chaos in the supply chain is the lack of visibility. In the absence of
communication, each firm in the supply chain acts in its own self-interest and on the
basis of their own forecasts. The one thing you know for sure about a forecast is that
it’s wrong; the one thing you never know is just how wrong. If firms have visibility on
the entire supply chain, chances are they would do much better. Therefore, a
commonly held belief is that the bullwhip effect is mainly due to lack of point-of-sale
(POS) data and/or good forecasts. In fact, obtaining POS data and good forecasts are
often cited as the primary reasons why the enterprises in the supply chain should
collaborate.
On the contrary, it turns out that the primary culprit for the bullwhip effect is lead
time. Even when there are no breakdowns in communication, you will still feel the
y
bullwhip effect due to procurement and manufacturing delays. This is not to say that
POS information and improved forecasting have little impact. In fact, reducing lead
nl
time, in combination with improved visibility along the supply chain, can
O
significantly and positively impact the bullwhip effect, as we demonstrate below.
So, what exactly is the impact each of these variables has on the bullwhip effect?
We find it is instructive to analyse the beer game using a quantitative approach, to
n
identify the impact of these variables a little more precisely. The beer game, it may be
io
recalled, is started with each enterprise carrying twelve cases of beer, and experiencing
a demand of four cases of beer each week. The lead-time for each enterprise to receive
at
a shipment against an order is four weeks.
ul
At the start of the simulation, the system is in steady state. The equilibrium is then
disrupted and the end-user demand increases to eight cases from week 5. Consider
irc
the impact of the increased demand on the retailer. The retailer begins this week with
twelve cases, receives four cases, but sells eight cases, and ends the week with only
eight cases in inventory. The retailer must now decide how many cases to order.
C
Suppose that each firm’s ordering policy is based on two very simple, but logical
rules; one rule to provide the forecast and another rule to determine the order
d
quantity:
ite
The forecast rule: The forecast of the weekly demand for each of the next four
weeks is just the average of the weekly demand over the four immediately
im
proceeding weeks.
The order quantity rule: Based on this forecast, the amount ordered is just
rL
enough to replenish the ending inventory (four weeks from now when the
order arrives) to a target of twelve cases.
Based on rule 1, the retailer forecasts his weekly demand to be five cases per week
Fo
for each of the next four weeks: (4 + 4 + 4 + 8)/4 = 5.

Rule 2 requires that the retailer’s inventory on hand plus the inventory on order to
be sufficient to cover the forecasted demand for the next four-weeks and have twelve
cases left in inventory. Therefore, the retailer must order the sum of the inventory
target (twelve) plus the forecasted demand for the next four weeks minus the
inventory that he already has on hand or on order:
Order = Inventory Target (12) + Forecasted demand for next four weeks (5 each) –
Current Inventory (8) – Orders already placed for the next three weeks (4 each) = 12
+ (5 + 5 + 5 + 5) – 8 – (4 + 4 + 4) = 12 + 20 – 8 – 12 = 12.
The retailer will thus place an order for twelve cases on the wholesaler.
y
A fundamental insight: The consumer demand increased by 100 per cent (from
nl
four cases per week to eight cases per week) but the retailer’s order to the wholesaler
increased by 200 per cent (from four cases per week to twelve cases for the following
O
week). The retailer thus doubled the variation in demand. This increase in variation is
due to the four-week lead time required to react to the forecasted increase in demand.
n
Next consider the wholesaler. Assume that the wholesaler behaves in an identical
manner as the retailer, except that the wholesaler’s demand is created by the retailer’s
io
orders. Initially the wholesaler receives four cases per week, sells four cases per week
at
and ends each week with twelve cases. Then the wholesaler unexpectedly receives an
order from the retailer for twelve cases. The wholesaler will begin the week with
ul
twelve cases, receive four cases, sell twelve cases and end with an inventory of four.
The wholesaler uses the four week average forecasting rule and the inventory target of
irc
twelve cases to arrive at the following demand forecast for each of the next four weeks:
The wholesaler’s forecast is: (4 + 4 + 4 + 12)/4 = 6 cases per week.
C
Hence, the wholesaler’s order will be:

Order = Inventory Target + Forecasted demand for the next four weeks – Current
d
Inventory – Orders already placed for the next three weeks = 12 + (6 + 6 + 6 + 6) – 4

– (4 + 4 + 4) = 20.
ite
The wholesaler’s order on the distributor has thus increased from four cases per
week to twenty cases the following week, an increase of 400 per cent.
im
Following the forecasting rule and order quantity rule, the distributor reacts to the
wholesaler’s order of twenty cases by ordering thirty-six cases, an increase of 800 per
rL
cent. The factory responds to this order by ordering enough raw material from its
supplier to make sixty-eight cases, an increase of 1600 per cent.
Fo
The variation is thus doubled at each stage. Of the sixty-four case increase in the
factory’s orders, only four cases were directly attributable to a change in consumer
demand. The lead times present in this value stream created 94 per cent of the variation
observed in the factory’s orders.
We summarise the results of this analysis as follows: Lead times significantly
exacerbate the bullwhip effect.
For the beer game analysis with a four-week lead-time and orders placed using the
forecasting rule described in this section, the variation in orders grows by a factor of
2 at each stage. In other words, the increase in variation at each stage is multiplicative.
Using the computations above, we find that the variation in orders grows by a factor
of 1.25 at each stage.
y
We thus observe that a moving average forecast does not reduce the bullwhip
nl
effect. We now show that the bullwhip effect is present even when there is perfect
O
information about the present and the future, instantaneously available to all
enterprises in the supply chain.
n
2.2.3 Improved Forecasting and POS Data do not Eliminate
io
the Bullwhip Effect
at
Let us use the same beer game scenario as before except that each stage is instantly
made aware of the consumer’s orders. Let us assume too, that the consumer orders
ul
four cases for weeks one through four, and eight cases for week 5, as before. However,
to enable a proper comparison with the analysis in the previous section, we will
irc
assume that the perfect information scenario reveals that the consumer demand for
the following weeks (week six onwards) is five cases of beer. This assumption allows a
C
fair comparison, because the forecasting method used in the preceding section
predicts a steady demand on the retailer for five cases of beer for the following weeks.
We also assume that this demand information is conveyed instantaneously upstream.
d
In order to keep the comparison fair, we assume that the lead-time to react to an order
ite
is four weeks at each stage.

Following exactly the same approach as before, the retailer will order twelve cases
im
of beer from the wholesaler for week 5 in order to bring the retailer’s inventory back
to the target level of twelve cases of beer. That is, the 100 per cent increase in demand
on the retailer translates to a 200 per cent increase in demand on the wholesaler over
rL
the previous week as before. Since the retailer sees the demand for the following
weeks to be five cases of beer, the retailer tells the wholesaler to expect a demand of
Fo
five cases for each of the following weeks. The wholesaler who sees an order of twelve
cases this week, is also aware that the retailer will order five cases each week thereafter,
and is currently receiving four cases of beer from the distributor for the next four
weeks. Thus, to bring the target inventory to twelve cases, the wholesaler orders
sixteen cases of beer from the distributor. In other words, a 100 per cent increase in
the demand on the retailer translates to a 300 per cent increase in demand on the
distributor over the previous week. Similarly, the factory will receive an order for
twenty cases, a 400 per cent increase in demand over the previous week, while the raw
material supplier will receive an order for twenty-four cases, a 500 per cent increase in
demand over the previous week.
y
This example, which assumes perfect information about the present and the future
would require POS data to be available at all stages in the supply chain and a perfect
nl
forecasting mechanism. Admittedly, this is not a very likely scenario. The point of the
O
example is to show that the bullwhip effect is still present even with such perfect
information, albeit to a smaller extent. With such perfect information the variation
in the orders generated at successive stages does not grow multiplicatively; rather it
n
grows additively (by 100 per cent at each successive stage).
io
There is a variant of the beer game, termed the near-beer game, which demonstrates
that POS data and good forecasting tools do not eliminate the bullwhip effect. In the
at
near-beer game, the supply chain consists of three enterprises; a supplier, a brewery,
and a customer. There is a single type of beer brewed. There is a delay of one week to
ul
receive raw material from the supplier. It takes one week to brew the beer and one
week to deliver to the customer. The system is initially in steady state with the
irc
customer ordering ten cases of beer each week. The brewery has ten cases in
inventory, ten cases of beer brewing and ten cases of raw materials arriving from the
C
supplier. For week two, demand increases to fifteen cases per week and remains at
fifteen cases thereafter. The simulation ends when the supply chain is back in
d
equilibrium with fifteen cases of beer. In the near-beer game, the brewery has perfect
information about the demand for beer, but the bullwhip effect does not go away.
ite
The near-beer game imparts all the lessons conveyed by the beer game. It also
teaches one additional lesson that the original game does not: the bullwhip effect is
im
present even if there is perfect information about the future that is shared among all
channel partners. Note that having perfect information about the future is even
rL
better than having POS data and excellent forecasting tools. The near-beer game thus
demonstrates that the bullwhip effect is best addressed by reductions in the
manufacturing and order lead times.
Fo
2.3 THE IMPACT OF LEAD TIMES

The analysis provided above demonstrates that lead time affects the variation in
demand in a multiplicative manner. That suggests we can minimise the bullwhip
effect (demand fluctuations in the supply chain) by reducing lead times. Reducing
lead times result in reduced operating costs as less capacity is needed to handle
demand fluctuations.
Reducing lead times provides a number of other benefits as well. Lets look at how
lead times affect forecast accuracy. Suppose the manufacturer in a supply chain
y
operates in a build-to-stock environment. If the manufacturer requires, say, four
weeks to build products, then the implication is that the manufacturer should build
nl
to a forecast four weeks further into the future. However, if the lead time is two
O
weeks, then the manufacturer only needs to build to a forecast two weeks further into
the future. Consider the implications. As Figure 2.2, indicates, the longer the time
horizon for the forecast, the less reliable it is. Thus, a forecast made for a demand that
n
is two weeks further into the future is clearly more reliable than a forecast made for a
io
demand that is four weeks further into the future.
at
Figure 2.2: Forecast Accuracy Decreases as the Forecast Horizon Increases
ul
irc
C
d
ite
im
Preparing for a forecast always carries with it an element of speculation. The

rL
enterprise that builds products based on forecasts typically pads the forecasted
demand in order to buffer for the uncertainty in the forecast. Such padding further
distances the amount produced from the true customer demand, adding to the
Fo
variation in the supply chain; and the longer the lead time, the greater the amount of
padding. Hence, as lead times decrease, the demands by the elements in the supply chain
converge more closely to a pure pull strategy in which there is no variation added by the
supply chain.
2.3.1 Little’s Law

Reducing lead times has another important benefit. Lead time reduction results in
lower inventory costs, as quantified by Little’s Law. This important law states that the
lead time in the system is directly proportional to the inventory in the system. In
y
particular, Little’s Law states that the average system lead time is equal to the average
nl
inventory in the system divided by the system throughput.
Little’s Law is an exact formula; it also permits an intuitive explanation. To provide
O
some intuition into Little’s Law, suppose each job takes t time units to process. So the
throughput rate, TH, which is the number of jobs processed per unit time is TH =
n
1/t. Suppose the work-in-process is W units. If a new order is now placed, then the
lead time, LT, for this order will be the time it takes to clear the W units of WIP,
io
namely, LT = W t. This gives W = TH ´ LT.
at
Little’s Law is useful when we consider inventory stocking plans. For instance, if
the lead-time for supplying material is high, then we need to carry more inventory
ul
somewhere in the pipeline to make sure we do not run out of inventory. An example
will better clarify the last statement.
irc
2.1
C
CSN Inc., is an organisation located in Chennai that makes fabricated parts. It

d
procures sheet metal from a supplier in Faridabad. CSN Inc., consumes 10 Kg of this
ite
sheet metal every day. The supplier has a 20 day lead time. What is the sheet metal
inventory in the system? To make matters simple, assume that the supplier is
extremely reliable, and that the lead time is exactly 20 days for every supply.
im
In this example, TH = 10 Kg/day, and LT = 20 days. So, the amount of inventory

in the system (either at CSN Inc., or in transit) is W = 10 ´ 20 = 200 Kg.
rL
The implication of Little’s Law is that when lead times are high, it results in
increased inventory in the pipeline. Conversely, when inventory in the supply chain
Fo
increases, lead times will correspondingly increase as well. This problematic and
cyclical relationship between lead times and inventory provides a powerful reason for
reducing lead times.
Ideally, if lead times were small and every enterprise in the supply chain could react
to a pure pull signal it would be possible to run the supply chain with near zero
inventory. Each enterprise in the supply chain would wait for its customer to place an
order before it ordered parts from its suppliers and would only begin production on
the order when the parts arrive. Thus, they would not need to carry any raw material,
work-in-process, or finished goods inventory. In turn, the lower inventories in the
pipeline usually results in lower lead times, generating a virtuous cycle. Needless to
say, the ideal is unlikely to be realised in practice for most supply chains unless lead
times are reduced.
y
nl
2.4 INVENTORY MANAGEMENT AND SUPPLY CHAIN DYNAMICS
O
The preceding discussion clearly points out the need for minimising inventories,
emphasising the need for organisations to continue looking for opportunities to
n
reduce inventories (and, therefore, lead times). However, there appear to be
io
limitations on the extent to which inventory can be reduced, simply because of the
nature of the production and distribution processes in the supply chain.
at
Furthermore, quite often there is pressure from marketing, sales, and even the finance
department to increase inventories. For instance, marketing and sales may exert
ul
pressure on the organisation to have high levels of finished goods inventory to
provide better customer service. Economic/cost considerations may also dictate the
irc
need for higher inventories in the system.
We now present some guiding principles for good inventory management. To this
C
end, note that from a supply chain perspective, inventory can be present in the form
of raw material, work-in-process (WIP) or finished goods inventory. Note too that,
from a supply chain perspective, for any organisation that is not at the downstream
d
end of the supply chain, its finished goods inventory is simply WIP in the system.
ite
The pressure to have increased inventories comes from the notion that a) customer
demand is unpredictable, b) supplier and manufacturing lead times are long, c) the
im
organisations in the supply chain have to deal with different products, not all of
which can be produced at the same time within the organisation, and there are setup/
changeover costs, d) customers need the product immediately, and will either buy a
rL
competitor’s product or shop elsewhere if the product is not available on the shelf.
While there are other reasons for driving up inventories, these four are fairly
Fo
representative of the reasons for having higher inventories.

Consider first customer service. The pressure to have high finished goods
inventory comes from the basic assumption that if you have finished goods inventory,
then you will be better positioned to satisfy customer demand without incurring any
delays due to manufacturing lead times. From a systems perspective, though, this
finished goods inventory probably exists because it was built in anticipation of a
future customer demand—one that may probably never materialise. (Presumably

that is why the organisation is carrying finished goods inventory instead of having
shipped it to its downstream customer.) In fact, Taiichi Ohno, widely acknowledged
as the person responsible for the Toyota Production System, is reputed to have said,
‘the more inventory you have on hand the less likely you are to have the one item your
y
customer actually wants.’ Also, in the process of creating finished goods inventory
nl
(which may not be consumed in the near future), the organisation has misallocated
its capacity to produce the wrong product. Furthermore, it has used raw material that
O
could have probably been used to produce an alternate product that the customer
really wanted.
n
From an economic angle, the pressure to have higher inventories stems from the
io
notion that if you build large batches, it will result in reduced setup/changeover costs;
also, if products are produced and transported in large batches, it will reduce
at
transportation costs. And it can be argued that higher (finished goods) inventories
reduce stock-out costs. On the other hand, from an economic perspective, the
ul
pressure to reduce inventories comes from the fact that high inventory levels result in
higher inventory carrying costs, storage and obsolescence costs, handling costs, etc.
irc
The classic economic order quantity (EOQ) or economic lot size models attempt to
balance these costs to arrive at the optimum batch to produce or procure, using
C
quantitative analysis and reasoning. Chapter 10 discusses these models in more

detail. In this chapter, we simply note that the notion of trade-offs between these
d
opposing costs is very relevant. However, we note that these models often have some
conceptual flaws. Consider, for instance, the classical EOQ model which provides a
ite
formula for the optimal order quantity. Rather than getting into mathematical details
in this chapter, we simply present the intuition behind this formula.
im
The EOQ model assumes that there are two costs that have to be traded off. One is
the inventory carrying cost, which increases linearly as the order quantity increases.
rL
The other cost is the ordering cost, which decreases in a non-linear manner as the
order quantity increases (the rate of decrease becomes smaller as order quantities
increase). Since one cost increases with increasing order quantity while the other cost
Fo
decreases, it seems intuitively clear that there is an optimal order quantity value that
minimises the sum of these two costs. Aside from the obvious over-simplification this
model makes of the real world, namely merging all the costs into just two cost
numbers, what is a possible flaw with this approach?
A flaw with the EOQ model is that it assumes the inventory carrying cost and the
ordering cost are fixed. As a result, many organisations naively attempt to estimate
these costs as accurately as possible using any available tools, determine the EOQ for
their products or raw material components, and use these numbers to drive their
procurement and manufacturing activity. When organisations operate in this
manner, they typically neglect to drive down these costs. For instance, in the 1980s,
even as the U.S. automakers were using the EOQ formula to drive purchasing or
y
manufacturing decisions, the Japanese manufacturers refused to accept the costs as
nl
fixed, but instead focused on driving down setup times (and, thereby, the setup
costs), which in turn resulted in smaller batch sizes being produced or ordered.
O
In sum, while there is often a pressure to hold more inventory for economical
reasons or for improved customer service, such pressures often stem from a lack of a
n
systems perspective. Organisations need to continue their efforts to drive down
io
inventories as far as possible simply because the presence of inventories is a symptom
or manifestation of long lead times; and long lead times only serve to increase the
at
bullwhip effect.
2.4.1 Risk Pooling

ul
irc
Inventory reduction can be realised through a simple concept—risk pooling. This
concept is typically explained using some statistical reasoning to explain how the
C
aggregate of a number of individual demands will have less variation than the sum of
the variation on the individual demands. Rather than getting into mathematical
details, we resort to intuition and consider a specific example of a distributor
d
supplying to multiple retail outlets. If each of these retail outlets serves their
ite
constituent customers, it has to carry inventory. The inventory carried by each outlet
should be adequate to meet the average demand placed on it, plus an additional
im
safety stock to satisfy the variation in the demand at the retail outlet. Thus, the total
inventory in the system is the sum of the inventory required to satisfy the average
demand at each retail outlet plus the sum of safety stocks maintained at each retail
rL
outlet.
On the other hand, if the entire demand was met directly from the distributor’s
Fo
facility, then the inventory the distributor must carry is the sum of the inventory
needed to meet the average demand at each outlet plus a safety stock that meets the
variation in demand across all outlets. As observed in the previous paragraph, the sum
of the variation in demand at each outlet will be greater than the combined variation
in demand across all outlets. To help the reader think this through, note that if the
demands on the different retail outlets were independent of each other, then if one
retailer encounters a demand in a given time period that is more than its average, it is
quite likely that during this same time period another retailer would encounter a
demand less than its average demand.
Next, we discuss offshoring and outsourcing in the context of their effect on
y
supply chain dynamics. This discussion is pertinent because such activities, especially
offshoring, typically result in higher inventories. For the purposes of our discussion,
nl
we define offshoring as the act of procuring, from a distant overseas location, goods
O
(not services) that were formerly sourced from a nearby location.
n
2.5 OFFSHORING AND OUTSOURCING: EFFECT ON SUPPLY CHAIN
DYNAMICS AND COSTS
io
The topic of offshoring/outsourcing is usually contentious and one can present very
at
compelling reasons for and against offshoring or outsourcing. It is not our intent to
discuss this topic in detail in this book. Rather, we restrict discussion to offshoring
ul
and briefly present its pros and cons, when viewed in the context of supply chain
dynamics.
irc
Organisations resort to offshoring for a variety of reasons. One reason is to locate
manufacturing activity in a country where the organisation wants a global presence,
C
usually to sell its products there. Another reason for offshoring is to use alternate
sources for products or materials that local suppliers are finding difficult to supply. A
d
third reason is to reduce costs; the intent is to source products or raw materials from
the most cost-effective source. Let’s look at these three cases, one by one.
ite
In an era of global competition, many organisations are trying to establish a global

presence for their products, attempting to penetrate markets in offshore locations. A
im
direct consequence of this endeavour is that these organisations may offshore their
procurement of raw material or components in order to manufacture their products
rL
from these remote locations. This is certainly a very good reason to offshore, since it
can facilitate the organisation’s goal to grow its business, profitably. Having local
suppliers will very likely make the products more appealing in the public’s eye, and
Fo
can lead to generous tax breaks and other governmental incentives. Furthermore, a
key benefit of having suppliers located close to the manufacturing site is that the lead
time for supply, and therefore the bullwhip effect, is significantly reduced.
Organisations also identify offshore sources for products because they indicate
difficulty sourcing these same products from nearby suppliers. While this might be a
valid reason for offshoring, one has to question whether the organisation has
systematically examined all possible alternative sources within its own borders. Quite
often, organisations are trapped in a cost-world mentality and turn a blind eye to
potential suppliers that might be willing to provide the product albeit at a higher
price. That leads to the third reason for offshoring—the desire to reduce costs.
y
Offshoring, when used as a means of combating higher costs, has proven to be a
mixed bag. Many organisations, especially in the U.S. and in Western Europe, have
nl
resorted to offshoring production to low-cost countries and the phrase, LCC
O
offshoring (to represent offshoring to low-cost countries) has become a common
buzzword in the 21 st century. More often than not, these decisions are a
manifestation of local thinking. The purchase department is often rewarded on a
n
metric known as purchase price variance (PPV) and that drives them to seek low cost
io
sources without considering the ramifications.
at
2.5.1 The Hidden Costs in Offshoring Math
ul
Consider a component required by a manufacturer to make its product. The
manufacturer currently produces 1,00,000 units of this product every month.
irc
Suppose the product requires a component that is currently purchased by the
manufacturer from a supplier located in close proximity for Rs 250 per unit, inclusive
C
of all taxes and duties. Suppose a supplier from a low-cost country is offering the
same product for Rs 125 per unit, FOB (‘Free on Board’) inclusive of all taxes and
duties. On first glance, it appears that the manufacturer can potentially save Rs 1.25
d
crores by offshoring the procurement of the component. What are the additional
ite
costs the manufacturer may not have considered? As we outline additional costs, the
reader should keep in mind that some or all of these costs may exist even with the
im
local supplier—if they already exist, then the reader should also note that they will
usually get magnified when dealing with offshore suppliers.
The first set of costs to be considered is possible additional transportation and
rL
handling costs incurred to move the product from the port of entry to the
manufacturing site. This will include any custom duties and tariffs, brokerage fees,
Fo
etc.
A second set of costs relates to risk in the form of currency risk, country risk and
competition risk (intellectual property risk). The last risk is if the offshore supplier of
the component decides to manufacture the product. This risk is also present with
local suppliers, but there is even less control over offshore suppliers. Another risk is
the so-called ‘job-switching’ risk. In a number of developing countries, many
promising employees are lured away, often on a day’s notice or even less, by a lucrative
offer from another employer.
The third set of costs relates to warranty and obsolescence costs. The following
questions need to be addressed: scrap cost per unit, inspection costs (do we need to
y
have our inspectors on site), disposal costs (costs incurred when the project is
complete), and warranty claims cost (who will pay—the offshore supplier or the
nl
manufacturer?).
O
A fourth set of costs relates to schedule of non-compliance. What is the expediting
cost? How many expedited shipments are needed each year? What is the cost of
stockouts and lost sales? This cost includes the cost of airplane trips made by high-
n
paid executives to the offshore site for troubleshooting and/or root-cause analysis.
io
The fifth set of costs is probably one of the more insidious costs. It relates to the
at
cost of additional inventory that now needs to be carried. This inventory manifests
itself in the form of additional inventory in the pipeline since procurement lead times
ul
are now much longer (remember Little’s Law), the additional safety stocks that need
to be maintained since there is usually more uncertainty in supply, and the batch
irc
processing costs that may arise at the manufacturer’s site since components now arrive
in large batches with the attendant overtime costs. More importantly, additional
inventory needs to be held to manage the bullwhip effect resulting from increased
C
lead times—this is typically a very large amount that is extremely difficult to

estimate.
d
There are other hidden costs that we may mention in passing—costs that are very
ite
difficult to estimate. They include the loss of innovative ability, brand and customer
relationships, loss of productive time/recovery time for the busy executive to fly
overseas for troubleshooting and root cause analysis, the lack of accountability, data
im
security/patent issues, etc.

rL
2.6 LESSONS LEARNED

The analysis we carried out shows that long lead times and the lack of POS data
Fo
exacerbate variation in the supply chain. A direct consequence of this observation is

that since lead times increase with more stages and handoffs in the supply chain,
reducing the number of stages reduces variation in the supply chain. In the analysis
we also observed how the variation at each stage is either additive or multiplicative,
depending on whether or not the system had perfect information about the future.
The use of POS data can help reduce variation from a behavioural perspective as
well. Managers of lean supply chains realise that end-user demand is more predictable
than the demand experienced by factories. Hence, they are more likely to ignore
signal distortions sent through the supply chain and instead focus on the end-user
demand. Such a practice does not react to day-to-day fluctuations but instead favours
y
running a level production schedule each day, helping mitigate variation in the
nl
supply chain.
O
It must be noted, though, that without perfect information about the future,
sharing POS information does not give much leverage to the enterprises when the
lead times are high. They still need to anticipate future consumer orders. Unless these
n
consumer orders are steady, the bullwhip effect will not be additive. This leads to the
io
following observation:
Point-of-sale data can reduce the bullwhip effect. However, without perfect
at
information about the future, POS data does not eliminate the bullwhip effect. Lead-
time reduction is also necessary. ul
Here are other ways to mitigate the bullwhip effect:
irc
➣ Smaller order batches result in smaller fluctuations. This highlights the need
to work with suppliers to enable more frequent deliveries in smaller order
C
increments.
➣ Maintaining stable prices for products reduces the customer’s temptation to
d
over-purchase when prices are low and cut back on orders when prices are
ite
high. This leads to more steady demands from customers.

➣ Allocating products among customers based on past orders rather than based
im
solely on their present orders will reduce hoarding behaviour when shortages
occur. Unrestricted ordering capability can be addressed by reducing the
maximum order size and implementing capacity reservations. For example,
rL
one can reserve a fixed quantity for a given year and specify the quantity of
each order shortly before it is needed, as long as the sum of the order
quantities equals to the reserved quantity. Leading enterprises like Barilla
Fo
SpA adopt this approach in their distribution strategy.

A key learning point is that structure drives behaviour. The way in which the supply
chain is designed largely determines how it will perform. The structural framework
for the beer game has the following components, all of which play a role in enhancing
the bullwhip effect:
➣ Lack of visibility along the supply chain—no POS data and a lack of
coordination or communication up and down the supply chain
➣ Long lead times for material and information flow
➣ Many stages in the supply chain
y
➣ Lack of pull signals
nl
➣ Order batching
O
➣ Price discounts and promotions.
The observation that structure determines behaviour is not a novel concept. Deming
n
alluded to this phenomenon when he said that management must take the
responsibility for poor performance and take steps to reduce process variation instead
io
of blaming the workers for poor quality. However, there are behavioural phenomena,
not necessarily driven by the structure, which also contribute to the bullwhip effect:
at
➣ Over-reaction to backlogs ul
➣ Withholding orders in an attempt to reduce inventory
irc
➣ Hoarding—where customers order more than they need because they are
anticipating a price increase or because the supplier has a promotional sale
➣ Shortage gaming where customers order more than they need because they do
C
not have faith in the supplier’s ability to deliver quality products and/or
because they do not expect the supplier to supply the entire order
d
➣ Demand forecast inaccuracies where a customer adds a certain per centage to

ite
the demand estimates, resulting in reduced visibility of true customer

demand
im
➣ Attempting to meet end-of-month, quarter, or year metrics.

In the beer game simulation, the bullwhip effect is observed even though the supply
rL
chain deals with a single product and there is just a one-time spike in demand.
Enterprises in the real world usually deal with multiple products with demands that
vary from period to period. Managing multiple products at a common facility leads
Fo
to batching of production or processing of any one product, in an attempt to reduce

the capacity lost while changing from one product to another. This adds to the
bullwhip effect.
Furthermore, in the real world there are many other factors that play a role.
Quality problems and unplanned events such as strikes and accidents induce
additional variation in the supply chain, and exacerbate the bullwhip effect. With all
these factors, it is easy to see why the bullwhip effect is present in almost any industry.
One of the major bullwhip related costs is the cost of capacity which is required to
meet some demand by customers, over a period of time. A simple analysis will show
that meeting demand that is stable and known ahead of time is far less expensive than
meeting the same demand in aggregate quantitative terms, but one which varies over
y
time. If this variation is not known, and is unpredictable and has to be forecasted and
nl
protected with inventory, the costs are even higher.
O
As we noted earlier, one of the lessons is that structure drives behaviour. Many
enterprises have gained a significant competitive advantage by understanding the
underlying causes of the bullwhip effect and building the necessary structural
n
framework within their own enterprise walls. These enterprises are also working with
io
their upstream and downstream partners to mitigate the bullwhip effect.
at
CONCLUSIONS
ul
irc
The bullwhip effect underscores the need for enterprises to understand the dynamics
of the supply chain and the primary causes of the bullwhip effect. The primary causes
for the bullwhip effect are:
C
➣ Lack of visibility along the supply chain

d
➣ Long lead times for material and information flow

ite
➣ Actions undertaken within the enterprise, such as order batching, price

discounts and promotions, etc., exacerbate the bullwhip effect.
im
In particular, lead times significantly affect the performance of the supply chain:
➣ Longer lead times lead to increased inventory in the system. Conversely,
rL
increased inventory in the system causes lead times to increase, resulting in a

vicious cycle
➣ Similarly, flow is enhanced by reducing lead times. Reduced lead times and
Fo
improved flow go hand in hand, creating a virtuous cycle

➣ Lead times are also influenced considerably by the variation in the system,
and therefore the manager of the enterprise should focus on reducing
variation
➣ In particular, therefore, it is important to design products and processes so
that they mitigate demand volatility.
Berinato, S., (2001), What Went Wrong at Cisco, CIO Magazine, August.
Engardio, P., (2001), Why the Supply Chain Broke Down, Business Week, 19 March.
y
Forrester, J. W., (1958), ‘Industrial Dynamics: A Major Breakthrough for Decision
nl
Makers,’ Harvard Business Review, 36 (4).
———, (1961), Industrial Dynamics, M.I.T. Press.
O
Gilbert, K., (2003), The Lean Enterprise, in E. R. Cadotte and H. J. Bruce (eds.), The
Management of Strategy in the Marketplace, Mason, OH: Thomson-Southwestern.
n
Hammond, J. H., (1994), Barilla SpA (A), Harvard Business School Case 9–694–046.
io
Senge, P. M., (1994), The Fifth Discipline: The Art and Practice of the Learning Organisation,
New York: Doubleday.
at
Case references : The issue of inventories is actually pervasive in supply chain management,
and is present explicitly in a number of cases, such as Laxmi Transformers, FarmAid
ul
Tractors Limited, Western Oil Limited (A) and others. Perishable inventories are
relevant in Bayer Crop Science and Food World (B).
irc
C
d
1. What is the major reason for the bullwhip effect?

ite
2. How does better information improve the bullwhip effect?

3. Consider the beer game with four elements, the retailer, wholesaler, distribution and
factory. Suppose that the demand on the retailer increases from four cases of beer to
im
eight cases of beer in week 5 and remains steady thereafter, just as before. However,
the lead time for an organisation to receive a shipment against an order is now two
rL
weeks, not four weeks. What is the resulting increase in the order placed by the
factory on its raw material supplier?
4. What are the ways by which an organisation can minimise the bullwhip effect? Be as
Fo
comprehensive as you can in your answer.

5. What is the significance of Little’s Law? What implications does Little’s Law have for
effective supply chain management?
6. Suppose the average monthly demand for a certain product is 12,000 units. The
organisation manufacturing these products works 20 days a month, and carries and
average inventory of 3,000 units of this product. What is the expected time the
organisation will take to respond to a new order?
7. Give the benefits and drawbacks of carring inventory.

8. What are some of the drawbacks of using the Economic Order Quantity formula to
determine how much inventory to carry?
9. What are the benefits of offshoring? What are some of the drawbacks? Compare and
y
contrast the benefits and drawbacks.
nl
O
n
io
at
ul
irc
C
d
ite
im
rL
Fo
y
CHAPTER
nl
Designing the Supply Chain
3
O
n
CHAPTER OUTLINE
io
Introduction
at
3.1 Some Issues in Supply Chain Design
3.2 Steps for Designing and Managing Lean Supply Chains
ul
3.2.1 Develop a Systems Perspective
3.2.2 Understand the Customers and Their Expectations
irc
3.2.3 Map the Value Stream
3.2.4 Benchmark Best Practices
3.2.5 Design Products and Processes to Manage Demand Volatility
C
3.2.6 Create Flow Across the Supply Chain

3.2.7 Categorisation Based on Work Flow: The V, A, and T Configurations
d
3.2.8 Characterisation Based on Fulfillment Strategy: BTS, ATO, BTO, ETO

3.2.9 Develop Metrics Using a Systems Perspective
ite
Conclusions
References
im
Exercises
rL
Fo
Designing the Supply Chain 61
INTRODUCTION
I
t has often been remarked that sured in months. The suppliers to
these manufacturing organisations, in
y
the design of any product usu-
ally locks in place as much as 80 turn, produced and delivered supplies
nl
per cent of the cost of producing and in large lots. Manufacturers and the
delivering the product over its lifetime. suppliers were able to co-exist, bliss-
fully unaware of any perceived threats
O
The same could be said of supply
chain design. The manner in which the to their operation, lumbering along in
supply chain is designed plays a very true behemoth-like fashion.
significant role in the cost of operation The 21st century supply chain oper-
n
of the supply chain. Ultimately, this ates in a vastly different environment.
io
cost of operation will affect the profit- Capacity now outstrips demand for al-
ability and survival of the supply chain most any product or service de-
in the customer-centric era. manded by an end-user, while the ex-
at
Supply chain design did not receive ecution time is measured in days,
much attention for the greater part of hours, and sometimes, minutes. As we
the 20th century because most of the
organisations operated in a produc-
ul
noted in Chapter 1, today’s supply
chain operates in a demand-driven,
customer-centric world. It must re-
irc
tion-centric mindset. Scale of econo-
mies drove the conceptual foundations spond quickly to rapidly changing cus-
of the supply chain. This worked well tomer demands in an agile, gazelle-like
as long as demand outstripped supply manner. To remain competitive in the
C
and customers were willing to compro- 21 st century, organisations must

mise their needs. However, such a recognise that the competitive arena
mode of operation resulted in unwieldy has shifted from competition between
d
behemothsorganisations that pro- organisations to competition between

supply chains. It is no longer a case of
ite
duced their products in large lots.

These large batches of products were Hindustan Unilever Limited competing
transported in full truckloads to re- against Procter and Gamble. It is
gional warehouses and distribution Hindustan Unilever Limited’s supply
im
centres from which they were delivered chain competing against Procter and
to retail stores or to other manufactur- Gamble’s supply chain.
ing facilities. A customer unwilling to This chapter presents some funda-
rL
buy the product had to place a special mental principles that organisations
order and wait for a long time for the should undertake to design a supply
product to be manufactured and deliv- chain. Specifically, it presents seven
ered. The lead time, namely the steps to design and build the necessary
Fo
elapsed time between order placement structural framework to enhance the

and product delivery was usually mea- performance of the lean supply chain.
3.1 SOME ISSUES IN SUPPLY CHAIN DESIGN

Which organisation in the supply chain should drive the supply chain design? Let us
first consider situations where this is a relatively straightforward question to answer.
Erecting the structural framework is, no doubt, an easier task for supply chains that
y
have a dominant organisation in the supply chain. For instance, in the previous
nl
chapters, we discussed some organisations such as Barilla and Toyota that took on the
responsibility for designing and managing the supply chain. In general, the
O
organisation driving the supply chain is usually the one primarily responsible for the
brand image. The brand owner could be:
n
➣ an upstream manufacturer such as Intel, who would have a significant
io
influence on the supply chain because of the technology driven inputs to the
supply chain, for example, through new product introduction
at
➣ a downstream manufacturer such as Hindustan Lever, which through a
network of own and contract manufacturing could position brands and
ul
provide market inputs to the rest of the supply chain
irc
➣ a retailer such as Walmart, which provides a large part of the final exposure to
the market place
➣ a trading company in commodities, which is usually largely anonymous to
C
the general business world, but which wields considerable influence through
volumes of purchases
d
Firms that provide transportation, warehousing or logistics services tend to

ite
diversify their business across many different products and services. Consequently,
these firms are generally unlikely to be brand owners. However, with international
im
outsourcing, a number of these firms have vertically integrated into a number of

businesses associated with logistics, such as global sourcing, quality certification, and
transportation through different modes. Some of these firms also provide other key
rL
logistics services such as tracking and insurance, customs and commercial

documentation. Consequently, they play an important role in the success of global
Fo
supply chains.
In general, there are relatively very few supply chains that have a dominant player.
However, even when there is no clearly dominant authority in the supply chain, it
will benefit the members in the supply chain if they are at least aware of the steps they
can follow to make their supply chain more competitive. The true competitive edge is
realised only when all key members in the supply chain jointly agree to work with
these steps. This chapter illustrates the principles of effective supply chain design,
focusing on manufacturing and related activities, since that is where the primary
value addition and technology inputs are high and where there is a lot of scope for
improvement. Once that is achieved, those principles can perhaps be applied to other
types of organisations, as elaborated in later chapters.
y
nl
3.2 STEPS FOR DESIGNING AND MANAGING LEAN SUPPLY CHAINS
O
Figure 3.1 entitled ‘The Lean Supply Chain Roadmap,’ presents seven steps that
organisations could adopt in their journey to develop lean supply chains. These steps
n
are discussed in more detail below.
io
at
Figure 3.1: The Lean Supply Chain Road Map
ul
irc
C
d
ite
im
rL
Fo
3.2.1 Develop a Systems Perspective

The beer game discussed in Chapter 2 underscores the importance of systems
thinking and its impact on the supply chain. The systems perspective recognises that
if each element in the supply chain tries to optimise its own operations in isolation,
y
everyone suffers in the long run. Thus, if each organisation in the supply chain makes
decisions in isolation without input from its immediate upstream and downstream
nl
supply chain partner, that only serves to exacerbate the bullwhip effect. Moving from
O
a local optimisation framework to a global optimisation framework poses a tremen-
dous challenge for organisations, as it is a radical shift from the traditional approach
towards managing an organisation.
n
For instance, an important element of supply chain management is long-term
io
partnerships with key suppliers. Suppose management institutes a measurement
system that rewards the Purchase department for obtaining products from its
at
suppliers at low cost. No doubt, reduced material costs directly affect the profitability
of the organisation. However, such a measurement system could drive the Purchase
ul
department into an adversarial position with its suppliers, encouraging Purchase to
play off potential suppliers against each other in an attempt to drive them to lower
irc
their prices. The lack of a systems perspective has resulted in a situation where the
organisation will now find it very difficult to establish long-term partnerships with its
C
suppliers.
As another example, consider a supply chain consisting of just two organisations,
A and B, where A is a supplier to B. Suppose A receives a mandate to reduce finished
d
goods inventory and responds accordingly. From a local perspective, the performance
ite
of A is enhanced. However, if the inventory reduction at A is done in isolation,

without any corresponding improvement in the pattern of B’s demands on its
im
suppliers, then A might be putting itself in jeopardy because it will be unable to react
quickly to unforeseen changes in demand from B. In fact, from B’s perspective, A will
be perceived as less flexible, and so B may either decide to carry some inventory of its
rL
own or find another supplier. The lack of supply chain metrics has led the manager of
A to make local improvements that did not lead to improved overall performance of
the supply chain.
Fo
In hindsight, systems thinking is so intuitive that one may wonder why it was not
applied to manage supply chains earlier. One reason is advocated by Senge [1994]
who claims that organisations do not practice systems thinking because they are more
absorbed with ‘detail complexity,’ as opposed to ‘dynamic complexity.’ A manager
who deals only with detail complexity is obstructed from seeing how different types
of interactions reach beyond his/her organisation, and change over time.
Another reason why systems thinking was not applied in the past was that until a
few years ago, requiring the members in the supply chain to work towards a unified
supply chain plan would have, at best, seemed a dream. However, the Internet and
availability of technology that provides visibility on end-user demand to all supply
chain partners, has led to a perceptible change. Supply chains that provide visibility
y
on customer demand are in a better position to ensure that small fluctuations in end-
nl
user demand do not amplify into huge swings in the demands placed on the
manufacturer. Recognising this fact, the members in many supply chains are now
O
more willing to set aside their traditional arms-length relationships to build long-
term partnering arrangements to achieve the competitive benefits derived from an
n
integrated supply chain. The increased visibility on end-user demand also allows the
io
managers of supply chains to better understand customer expectations.
The reader at this stage might be wondering how he or she could apply systems
at
thinking in his/her organisation. We provided a specific example in Chapter 1, when
we discussed how Hindustan Unilever adopts this philosophy to cut across
ul
departmental boundaries in its decision-making process. To answer this question in a
more general framework, we will use a conversation that takes place in The Goal
irc
[Goldratt and Cox, 2004], which depicts the story of a plant manager, Alex Rogo,
who is advised by his old physics professor, Jonah. Alex meets Jonah, by chance at an
C
airport between flights and tells his professor that robots installed in his plant have
increased productivity by 36 per cent. Jonah questions Alex whether his organisation
is really making 36 per cent more money from your plant, just by installing some
d
robots. Alex reflects on this question and says no. Jonah next asks Alex whether his
ite
plant was able to ship more products every day after installing robots. Alex again
answers no. Jonah then asks Alex whether he fired anybody (answer: ‘No’) or whether
im
inventories went down (again, answer: ‘No’). Jonah now questions Alex decision on
installing the robots in the first place.
The above dialogue between Alex Rogo and Jonah provides an operational
rL
perspective for applying systems thinking. For any decision under consideration, ask
whether the decision will either:
Fo
➣ increase Throughput (T), namely, the money making potential of the

organisation,
➣ reduce ‘Inventory’ (I) in the form of working capital and/or fixed assets, or
➣ reduce ‘Operating Expenses’ (OE) long-term, namely, result in any reduction
in fixed expenses over the long haul.
If the answer to these three questions is ‘no,’ then the decision under consideration
is very likely to be questionable. Thus, the three measures provided above (T, I, and
OE) provide the levers for systems thinking and supply chain coordination.
The last item, reducing OE long-term, merits further elaboration. Quite often,
y
managers are forced into some cost-cutting initiatives as a knee-jerk reaction to some
external pressures, for instance, a desire to affect the stock price. However, unless the
nl
long-term ramifications of such cost-cutting moves are carefully considered, such
O
moves will typically backfire, resulting in a sub-optimal conclusion.
n
3.2.2 Understand the Customers and Their Expectations
io
What is customer value? This is a crucial question addressed next. In the supply chain
design, we must always address the question on whether the supply chain is designed
at
to be responsive to customer needs and values. As we observed in Chapter 1,
Benetton was able to gain a significant competitive edge when it decided to adopt a
ul
different process for delivering customer value by postponing the differentiation of
its product. That process change allowed Benetton to significantly reduce the
irc
problem of markdowns that plague the textile and apparel industry. Similarly,
Hindustan Unilever has the strategy of continuously re-engineering its supply chain,
C
introducing vendor managed inventory concepts, adopting systems thinking, etc., as

discussed in Chapter 1.
d
In sum, there is an opportunity for the supply chain partners to challenge the way
in which product is traditionally being delivered. Understanding customer value
ite
means that you should, at the minimum, identify the attributes your product must
have. You must identify the order qualifiers and order winners for your product [Hill,
im
2000]. Order qualifiers are attributes that the product must have for the customer to
even consider purchasing it from you. Order winners are the attributes that will get
you the customer’s order. Order qualifiers and order winners determine the
rL
competitive priorities your supply chain should focus on. Should it be speed to
market, product design, product quality, on-time delivery, or a combination of these?
Fo
The answer to this question depends on what is important to the customer base.
Order qualifiers and order winners will, no doubt, be different for different
customer segments. For instance, if the airline industry wishes to attract family
vacationers travelling relatively short distances, then low price tickets could be order
winners. On the other hand, for business travellers travelling coast to coast in the
U.S., travel on wide-body jets may be the most preferred alternative. Thus, comfort
and on-time travel would be order winners. Similarly, to determine order winners
and order qualifiers for computer buyers, you need to segment the users into desktop
and laptop user, and for, say, laptop users, you need to further segment them into
business travellers, home users that primarily use laptops for word processing and
email, and users that primarily use laptops for interactive games. Each market
y
segment will have different order winners and order qualifiers that have to be
nl
determined by surveying a large number of users within each market segment. In
addition to understanding what the order qualifiers and order winners are, you must
O
also determine how you are currently performing relative to these attributes both in
the eyes of your customers and relative to the competition.
n
Order qualifiers and order winners are important to the designer of the supply
io
chain because they facilitate the dialogue between marketing, which identifies the
voice of the customer, and operations, which is responsible for delivering on these
at
attributes. Note that order qualifiers and order winners are dynamic attributes.
Changes in the market place and changes in technology can change the order
ul
qualifiers and order winners. Over time, many perennial order winners may become
mere order qualifiers as a result of tremendous advances made in the area of
irc
manufacturing technology, planning systems, and information technology. For
instance, a combination of quality and low cost used to pack a winning punch only a
C
short time ago, but it is no longer a differentiator. A key function of marketing is to

stay abreast of these changes so the organisation can remain proactive in meeting
customer expectations and continuing to increase revenues.
d
Once order winners and order qualifiers are identified, operations must evolve
ite
different product delivery strategies for the different market segments. Supply chains
can be characterised as build-to-stock (BTS), assemble-to-order (ATO), build-to-order
im
(BTO), and engineer-to-order (ETO). The BTS supply chain provides the fastest
response to the customer, but is accomplished with pre-build end item speculation.
In contrast, BTO and ETO provide a long response time to the initial customer
rL
order, but this is accomplished with little pre-build speculation. The customer must
wait for most of the parts and components to work through the supply chain as
Fo
custom parts. The ATO configuration provides a middle ground where the response
time to the customer is confined to the assembly time. This is accomplished with pre-
build speculation of components and modules, rather than end items.
The delivery strategy could thus be a combination of several models: BTS, BTO, ETO,
and so on. At a macro level, the supply chain structure should delineate which segments
are going to operate in a BTS mode and which segments will operate in a BTO mode.
As far as possible, we would like to wait for the customer’s order and build the
product per customer’s specifications. However, customers have differing delivery
expectations. Some customers would be prepared to wait for a product that is built
exactly according to their specifications while others would prefer to compromise on
some product features if they can purchase the product immediately. For the latter
y
type of customer, the unavailability of a product could result in a lost sale, and the
nl
organisation may necessarily have to resort to a BTS mode of operation for such
customers and maintain some finished goods inventory. However, finished goods
O
inventories do not always solve the problem. As noted in Chapter 2, Taiichi Ohno is
reputed to have said that ‘the more inventory you have on hand the less likely you are
n
to have the one item your customer actually wants.’ In other words, the fact that you
io
have a lot of inventory of a certain product is symptomatic of the organisation having
built the wrong kind of finished goods inventory. Deciding on the right kind of
at
finished goods inventory is always a challenge, and organisations would therefore like
to maintain as little finished goods inventory as possible. For instance, in the
ul
electronic industry where some products depreciate almost as fast as groceries do, it is
advisable to maintain as little finished goods inventory as possible.
irc
It is therefore expedient to develop a customer time-based demand profile that
identifies customer expectations in terms of lead time, and develop finished goods
C
inventory strategies accordingly. To understand how the customer time-based

demand profile could be used, consider, for instance, a product such as under-the-
counter dishwashers. Customers for this dishwasher are individual homeowners,
d
retail stores, or building contractors. The individual homeowner is either one who is
ite
in need of a dishwasher now, because his dishwasher is broken and not repairable, or
is shopping around to replace a functioning dishwasher. For the homeowner whose
im
dishwasher is broken, the lead time expectation on the order could be zero or, at
most, one day. For the retail store owner, the lead time expectation could be one or
two days, whereas for the building contractor, the lead time expectation could be one
rL
week. For the individual homeowner shopping around for a dishwasher, the lead time
could even be as high as a month; this customer may wait until she gets a real good
deal. Clearly, there are different types of customers, each with their own lead time
Fo
expectations. Clearly, too, the manufacturer of dishwashers does not have to carry
finished goods inventory for all customer types. If the manufacturer is able to build
dishwashers to order in three days, then it only needs to carry finished goods
inventory for retail store owners and customers seeking to replace a broken
dishwasher. Figure 3.2 presents a customer time-based demand profile.
Figure 3.2: The Customer Time-Based Demand Profile
y
nl
O
n
io
at
ul
This profile identifies the percentage of customers who demand products with a
irc
lead time expectation of one day (they want their order filled right away), are
prepared to wait for two days, three days, and so on. To arrive at this profile, the
C
organisation should gather data on customer orders received (not customer orders
filled), and the delivery time expectations that were associated with each order.
d
Ideally, the organisation should also gather data on potential customers who decided
not to place an order simply because the lead time was too high. Ideally, too, it is
ite
advisable to develop one such time-based demand profile for each product, rather
than a generic profile for each market segment.
im
The time-based demand profile is a very useful tool to match customer

expectations with operations capabilities. For instance, suppose the lead time for the
organisation to fulfill demand is between two to four weeks. All orders requiring
rL
delivery in less than two weeks would have to be met with finished goods inventory.
Customers that place such orders can be classified as ‘At-Once’ customers. Customers
Fo
placing orders that fall within the lead time window are classified as ‘At-Lead-Time’
Customers, and those that are prepared to wait for more than four weeks would be
‘Beyond-Lead-Time’ customers. Ideally, the organisation would like to have more At-
Lead-Time and Beyond-Lead-Time customers and no At-Once customers, because in
that situation, all orders could be built to demand; that is, the organisation can
operate in a pure BTO mode. The At-Once customers, on the other hand, require a
BTS delivery strategy [Reeve and Srinivasan, 2005].
The time-based demand profile highlights the importance of lead time reduction,
because the At-Once customer base becomes smaller as the lead time to process an
order decreases. To enable lead time reduction, the organisation could pursue a
postponement strategy as we discuss in a subsequent section of this chapter. The
time-based demand profile could also be used to develop pricing strategies. For
y
instance, At-Once customers could be charged a premium, whereas Beyond-Lead-
nl
Time customer could be given a discount. In this manner, it is possible to influence
the demand profile itself so that more customers become At-Lead-Time or Beyond-
O
Lead-Time customers.
n
3.2.3 Map the Value Stream
io
A value stream map illustrates the structure of the physical flow of goods and
at
information flow, and highlights areas in the value stream (supply chain) that require
more attention. It is not the intent to provide a detailed discussion on the topic in this
ul
book. We will simply discuss the vital importance of value stream mapping in supply
chain design and identify a number of sources to the reader interested in further
irc
pursuing this topic [M. Rother and J. Shook, 1999; M. Rother and R. Harris, 2001;
J. P. Womack and D. T. Jones, 2002].
C
A typical approach to value stream mapping is to first select the product family for
which the value stream should be mapped. The product family is often chosen simply
based on the biggest customer for the organisation. The next step is to draw an
d
existing value stream map (an ‘As-Is’ map). The non-value added work and
ite
constraints are then minimised to design a map of the desired end-state value stream
(a ‘To-Be’ map).
im
As shown in Figure 3.1, the intent of a value stream map is to provide

opportunities for removing muda, mura, or muri. These are three Japanese words that
respectively mean wastefulness, unevenness and overburdening. Muda exists in every
rL
organisation, in the form of unnecessary delays, inventory, motion, transportation,

and so on, none of which add any value to the customer. Mura exists in a variety of
Fo
forms—unevenness in quality, unevenness in sales and production (which could be

caused by poor use of incentives and promotions, or simply the desire to meet
quotas), or unevenness in supplier delivery performance. All of these can result in
Muri. For instance, when there is unevenness in the demand, that could well translate
to overburdening some of the resources, albeit temporarily. However, muri could also
exist due to the presence of either physical constraints or simply due to some policies
that create artificial constraints. In summary, muda, mura and muri result in wasteful
use of resources in an organisation. The focus of the value stream map is to highlight
the sources of these problems and to initiate the steps that would remove these
problems.
y
It is usually hard to determine the degree of detail for a value stream map. The
manager of an organisation is mainly interested in the immediate linkages the
nl
organisation has with its upstream and downstream partners. In that regard,
O
broadening the map to include organisations beyond the immediate upstream and
downstream organisations can result in loss of detail. However, if one can map the
flow in this manner to second tier suppliers, and/or to the organisation’s customer’s
n
customer, it may be easier to visualise the kind of systems that must be put in place to
io
achieve better customer service.
To get started on the value stream mapping exercise, Rother and Shook [1999]
at
recommend starting with a ‘door-to-door’ value stream map, and working on a
process that can typically be improved within around 90 to 120 days. Such an
ul
approach has the advantage of providing momentum to lean efforts. A potential
disadvantage is that it could lead to missed opportunities since a door-to-door map
irc
might not capture operational inefficiencies outside the organisation, where
improvements could provide more impact.
C
Advances in IT have significantly facilitated value stream mapping. They now

allow the capture of a much higher level of detail than was possible in the past.
d
Moreover, computer systems make it possible for a manager to look at the degree of
ite
detail necessary to effectively manage his area. Once the As-Is map is complete, key
structural elements should be examined in detail to identify how it should be
managed so as to best impact the performance of the entire value stream. From an
im
operational point of view, there are several key areas of concern in any product/work
flow. These include:
rL
➣ Segments of the value stream where processing times are large. Time is a key
element in gaining a competitive edge in the consumer era. Segments of the
value stream that have difficulty in responding quickly to changes should be
Fo
scrutinised and managed carefully.

➣ Any segment or path that contains physical constraints—either in the form
of material constraints or capacity constraints. Effective management of the
value stream is impossible without knowing where the constraints are and
how they influence the performance of the value stream as a whole.
➣ Points where there is a high degree of resource sharing. When the same
resource is required to process a variety of products, contention for priorities
will arise and any mismanagement of these priorities can result in delivery issues.
➣ Points where common materials are transformed into different product
y
streams. An example of this is the steel making process in an integrated steel
mill where the basic pig iron can be converted into a variety of different alloy
nl
steels. Once this transformation is complete the processes cannot be reversed.
O
If the wrong product is produced, we end up with inventory (the product just
produced) and an urgent shortage (the product that should have been
produced).
n
➣ Points where multiple materials must come together. These are assembly
io
points. Since the assembly process requires all required materials to be
available, the logistical challenge of making sure that all the different
at
materials required arrive on time is significant.
➣ Points of excessive variation. ul
A comprehensive value stream map can thus highlight the weakest linkages,
irc
indicate points in the value stream which have high lead times, and so on. You can
identify where there are insufficient resources available to manage supplier
C
relationships, whether there is a proliferation of suppliers at some of the

organisations, or whether the size of the supply chain allows you to develop sound
supplier development strategies.
d
ite
3.2.4 Benchmark Best Practices

im
Benchmarking is a continuous, systematic procedure aimed at measuring the

organisation’s products, services and processes against the best-in-class practices.
Benchmarking is not aimed at imitation. Rather, it is focused on studying and
rL
learning from others and adapting practices that best suit the organisation.
Organisations can choose to perform competitive benchmarking or functional
benchmarking. In the former case, the focus is on comparisons with leading
Fo
organisations in the same industry. Competitive benchmarking usually involves the

organisation benchmarking itself with the leader in that industry. Such a form of
benchmarking allows the organisation to avoid making some of the mistakes that the
leader may have made in its journey to become the best in the industry. The obvious
disadvantage of such a form of benchmarking is that the organisation becomes a
follower. Typically, the leader will be one step ahead of the competition.
Functional benchmarking, on the other hand, compares processes and activities

such as customer service, the design process, and the product delivery system against
exemplary practices by best-in-class organisations in any industry. This is the more
ambitious form of benchmarking since it selects organisations recognised as world
leaders in some processes and attempts to match up to them. For instance,
y
organisations may choose to benchmark Disney for customer service, Dell for rapid
nl
customisation, Toyota for process execution, and American Express for its ability to
get customers to pay quickly.
O
The advantage of functional benchmarking is twofold. Since it is not comparing
itself with the leader within an industry, the organisation carrying out the
n
benchmarking does not need to worry about following the leader. Second, and more
io
important, functional benchmarking provides opportunities for the organisation to
identify new and innovative ways of fulfilling customer demand. For instance, Taiichi
at
Ohno credits his contributions to this system to two concepts: the moving assembly
line pioneered by Henry Ford, and the supermarket [Ohno, 1988]. In a visit to the
ul
United States in 1956, Ohno observed how supermarkets used a continuous
replenishment of merchandise. This concept gave Ohno the idea to set up a pull
irc
system in which each production process became a ‘supermarket’ for the succeeding
process. Each process would produce to replenish only the items that the downstream
C
process selected. When Ohno adopted the supermarket concept for Toyota, he had a
competitive edge since this concept was not used by any of the other automobile
manufacturing organisation at that time. The following lean supply chain principle
d
[Srinivasan, 2004] summarises this discussion.

ite
im
LEAN SUPPLY CHAIN PRINCIPLE

rL
Focus on customer needs and process considerations when designing the product.
Organisations can gain tremendous competitive advantage through best-in-class
practices that cut across industries.
Fo
3.2.5 Design Products and Processes to Manage Demand

Volatility
The beer game showed us how even a small change in demand can lead to large
inventories upstream, especially when lead times are large. Hence, there is a clear case
that can be made for the supply chain to design products and processes that can
mitigate and/or cope with demand volatility. Demand volatility is an accepted fact of
life in business that has become even more challenging as consumers grow more
demanding. The first challenge is to design the structure of the supply chain and
develop policies that will reduce demand volatility as far as possible. Having
y
mitigated demand volatility, the next challenge is to manage demand volatility with
nl
existing processes and equipment and yet achieve high levels of customer satisfaction
and operational effectiveness.
O
It is a remarkable fact that much of the demand volatility is self-induced. A classic
example of self-induced demand volatility is the instability created by sales
n
promotions or rebates, which usually generates a sharp surge in end-user demand and
io
the inevitable bullwhip effect that accompanies such a surge. Managers of supply
chains fail to recognise that quite often the volatility of demand is significantly
at
influenced by an organisation’s sales or marketing activities. Conversely, the sales and
marketing group often fail to use appropriate tools and techniques that can help
ul
mitigate demand volatility. For instance, one simple approach to eliminate such self-
induced volatility is to follow the every day low price concept, as offered by some
irc
organisations such as Wal-Mart. The end-of-quarter or end-of-year ‘channel-stuffing’
actions undertaken to show higher operating efficiencies and margins also increases
C
demand volatility. Such actions merely serve to create inventory elsewhere in the
pipeline that will then have to be disposed through a sale, further exacerbating the
bullwhip effect.
d
Another example of self-induced volatility arises as a result of batching. While the

ite
end-user demand for a product may be fairly level organisations often deliver the
product in large lots to achieve economies of scale. The same practice applies all the
im
way up to the manufacturer who resorts to a large-lot production strategy, choosing

to wait until the cumulative demand reaches an ‘economic lot size’ before beginning
production. From the beer game we observed that a single spike in demand created
rL
large ripple effects upstream. By batching a steady end-user demand, we immediately

generate the bullwhip effect as far as the upstream suppliers are concerned. The
Fo
solution here is to work with small batches and level production schedules. Working
with a level production schedule, ignoring noisy data resulting from promotional
activity or end-of-the-quarter channel-stuffing activities, will help prevent the
bullwhip effect from propagating upstream.
Improving the responsiveness and reliability of the supply chain is yet another way
to reduce demand volatility. It is important to understand how the responsiveness of
the supply chain is itself a major determinant of demand volatility. First, as we

observed earlier, the further out into the future one has to forecast, the more
unreliable the forecast is. Hence, a supply chain that is less responsive (requires more
time to deliver products to the consumer) has to contend with a demand forecast that
is further out into the future and is correspondingly less reliable. Consequently, the
y
variation in the actual demand observed is higher. Second, if customers sense that the
nl
supplier is unable to deliver what they want, and on time, then they will
understandably hedge their requirements in possibly two ways. They will demand
O
more product than their actual requirements and/or will ask for it to be delivered
sooner than it is really needed. In either case, demand volatility is increased. On the
n
flip side, if you respond quickly to customer demands, the customers will have more
io
faith in your ability to deliver and are therefore less likely to pad their actual
requirements or their desired due dates.
at
A very useful approach to manage demand volatility is to ‘maximise external
variety with minimal internal variety.’ This phrase succinctly captures the basic
ul
principle that should be followed, especially when designing supply chains that deal
irc
with high product variety and demand volatility. This principle can be accomplished
by structuring your product offerings so that commitment of material and resources
can be postponed for as long as possible. In other words, we want to work with a
C
relatively small number of standard products (‘modules’) internally in semi-finished

or finished form to configure a large variety of end products. It will be convenient for
d
us to refer to the above principle as the RAP (keep the in-process inventory as ‘raw as
ite
possible’) principle. Figure 3.3 illustrates the RAP principle. This principle is often
referred to as the principle of postponement, and we saw some applications of this
important principle in Chapter 1.
im
The RAP principle should drive the design of new products and services. It
provides a very convenient way to meet customer demand quickly, without storing a
rL
lot of finished goods inventory. At the same time, it delays committing raw material,
labour, and fixed assets to make products based on forecasts, in anticipation of future
demand. As shown in the ‘After’ portion of Figure 3.3 differentiation of the product
Fo
is postponed to the final assembly stage. By postponing differentiation to the final

assembly stage, it is possible to maintain an inventory of the products in an
undifferentiated ‘Sub-Assembly’ form, ready for conversion into one of three possible
finished products relatively quickly.
Figure 3.3: The RAP Principle
y
nl
O
n
io
at
ul
The following lean supply chain principle [Srinivasan, 2004] presents this idea.
irc
C
Maximise external variety with minimal internal variety. It is desirable to maintain

d
inventories in an undifferentiated form for as long as it is economically feasible to do

so.
ite
The RAP principle is particularly valuable for managing short-life products.

Postponement increases service levels while reducing costs and order fulfillment risk
im
and is especially valuable when there are many derivative products and forecast error
is high. The RAP principle has an added benefit. It is a well-known fact that when we
rL
aggregate a number of independent demands, then the aggregate demand has

significantly less variation than the individual demands. When the RAP principle is
adopted, the same standard modules go into a wide variety of end products. If we
Fo
store modules, rather than end products, then we can manage demand volatility with
much less inventory since the variation in the demand for modules is much less than
the variation in the demand for end products. Applying the RAP principle thus helps
reduce demand volatility.
Despite our best efforts to manage demand volatility using one or a combination
of the approaches suggested above, suppose we still find demand variation present in
the supply chain. What would you do to cope with this variation? The classical (and
traditional) approach is to resort to finished goods inventory to buffer the variation.
However, we have seen that using this approach has its drawbacks since inventory
results in slower responsiveness and longer lead times. Taiichi Ohno is reputed to
have said that ‘Inventory is the root of all evil.’ (We should state, instead, that
y
‘Variation is the root of all evil,’ because variation is the root cause for such inventory
nl
build-up.) Little’s Law shows us that lead times and inventory go hand in hand,
creating a problematic and cyclical relationship; and the beer game showed us the
O
effect of lead times on supply chain performance. Moreover, when organisations are
forced to commit their scarce and inflexible resources to produce finished goods
n
inventory in the face of demand volatility, they quite often end up mis-allocating the
io
resources to produce the wrong kind of products. Thus, there is a powerful incentive to
reduce inventories in the supply chain. So, we come back to the question: how can
at
organisations cope with demand variation in such a situation?
Instead of using inventory to buffer variation, a better approach is to buffer
ul
demand variation with a small amount of reserve capacity. This approach is likely to
be met with some resistance from managers who do not view the supply chain from a
irc
systems perspective, but are driven by metrics that force them to cut costs and run
their operations at full utilisation. Such metrics penalise managers if they do not use
C
their flex capacity even though the situation may not warrant it. However, as we
repeatedly emphasise throughout the book, a key factor in building lean supply
chains is to maintain flexibility at every stage; and that is best achieved with capacity,
d
not inventory. We can summarise this discussion with the following lean supply chain
ite
principle [Srinivasan, 2004]:

im

rL
Buffer the variation in demand with capacity, not inventory.

While the ideal situation is to buffer variation with capacity rather than with
Fo
inventory, there are a number of situations where you are compelled to use inventory
to accommodate variation. For instance in the consumer products industry, if the
end-user does not find the goods available on the shelf, he/she may go elsewhere. If
you are forced to buffer variation with inventory, then a related issue is how inventory
could be used more strategically. Should you have finished goods inventory or should
you apply the RAP principle and have partially processed items, ready to be
assembled to customer order, so that you preserve some flexibility? The answer, of
course, depends on the lead time expectations of the customer. The point is that you
have to decide where to strategically locate the inventory to meet these expectations.
You must determine how far upstream in the supply chain you can afford to locate
inventory and still meet customer lead time expectations. We summarise the
y
discussion on the different ways to mitigate and cope with demand volatility as
nl
follows:
O
➣ Avoid using sales promotions/rebates and metrics that promote the end-of-
the-quarter syndrome to the extent possible. If possible, use an ‘Every-Day-
Low-Price’ approach.
n
➣ Avoid batching. Try to work with small batches and a level production
io
schedule.
at
➣ Improving the responsiveness and reliability of the delivery system will
significantly reduce demand volatility.
ul
➣ Maximise external variety with minimal internal variety. Be aware of the
power of the RAP principle.
irc
➣ Buffer variation in demand with capacity, not inventory.
➣ Where inventory is necessary, it should be placed at strategic locations in the
C
supply chain to meet customer lead time expectations.

d
3.2.6 Create Flow Across the Supply Chain

ite
The ability to react quickly to customer demand without carrying large amounts of
inventory at various stages in the supply chain is better achieved if every organisation
im
in the supply chain works in harmony to build products at the rate demanded by the
end-user. This concept of flow balance essentially means that all the organisations are
rL
‘rowing the boat’ at the same pace. Clearly, if some organisations in the supply chain
work faster than some others, the imbalance in flow will result in inventory piling up
in front of the weaker links, namely the organisations that work at a slower pace.
Fo
Balancing flow across the supply chain requires a systems perspective. The idea is
to focus on the product and identify all the steps it goes through in the process of
moving from the raw material stage until it is delivered to the end-user. Are there
process steps that introduce unnecessary delays? Are there any unnecessary non-
value-added activities that the product goes through? Where are the potential
bottlenecks that delay the smooth flow of the product? Are some of these bottlenecks
due to unnecessary processing steps? Have information-processing delays been

eliminated? Such questions will bring forth problems with existing work practices
that may hinder the smooth flow of the product.
A major impediment to flow is caused when organisations produce products in
y
anticipation of future demand. When organisations use forecasts to drive their
production schedules, the forecasting methods used can clearly impact the proper
nl
execution of schedules. A poor forecast may result in raw material not being available
O
when the customer asks for a product to be delivered. Even when the organisation
uses good forecasting techniques, the forecasting process could generate problems.
Forecasting has historically been applied with a silo mentality, with multiple
n
departments within the same organisation independently creating forecasts for the
io
same products, using their own assumptions, measures, and level of detail. The
impact of these localised forecasts on the supply chain is often considered only
at
informally, if at all. The functional silos in the organisation worsen the situation. For
instance, it allows the sales function to envision growing demand while the
ul
operations function is left guessing how much the customer really wants.
Although it is desirable to use pull systems wherever possible, organisations in a
irc
supply chain still have to anticipate customer’s demands and must therefore rely on
forecasts. The key is to resist the temptation to execute the production schedule based
C
on a forecast. Rather the idea is to use demand forecasts for planning production and
use pull signals based on true customer demand to schedule production. In addition,
d
a system that responds to pull signals inherently has less variation than a system that
pushes products through the supply chain. The following lean supply chain principle
ite
[Srinivasan, 2004] summarises the preceding discussion:

im

rL
Use forecasts to plan and pull to execute. A system that reacts to pull signals will have
less variation than a comparable system that adopts a push mode of operation.
Fo
No doubt it is desirable for these organisations to use pull signals to trigger

supplies from their upstream organisations as well. Carrying this idea one step
further, the intent is to start with the customer, deliver what is demanded, build what
is sold, supply what is consumed, and above all, balance the flow. The essence of the
lean supply chain is to create flow—across the supply chain. Figure 3.4 conveys this
idea. This figure shows a simple relationship between the manufacturer (‘Factory’),
its customer and its supplier.
Figure 3.4: Use Pull Signals to Create Flow Along the Supply Chain
y
nl
O
n
io
at
In essence, the lean supply chain is designed to just deliver what is demanded,
ul
build what is sold, and supply what is consumed, all the time maintaining a flow
balance. These flow objectives are accomplished with high frequency pull signals at
irc
each level of the supply chain. Thus the vision of the lean supply chain is to have
every upstream process react to a pull signal from its downstream customer and
C
produce a product only when the customer demands it. Such a vision allows the
members of the supply chain to delay commitment of their valuable resources and
their raw material until there is a definite demand for their product or service. This
d
vision is consistent with the familiar postponement design attribute that is key to
ite
most successful supply chain designs.

However, this pure conceptual vision must be adjusted for real-world imperatives.
im
For example, in a retail environment, the consumer typically requires spontaneous

fulfillment from stocked items. This requirement does not negate a lean approach to
managing the supply chain. This is because the lean supply chain is not about zero
rL
inventories; it is about the strategic use of inventories. The question is: how do we
ensure that the supply chain operates with minimal inventories at the right location?
Fo
To answer this question, let us use the model in Figure 3.4 as a canonical model
replicated throughout the supply chain. That is, unless the organisation is a producer
of the basic raw material used in the process, practically every organisation has a set of
customers and a set of suppliers. We refer to this as a supply chain triad. Within this
triad we will maintain the terms, customer, factory, and supplier.
To decide on how much inventory to hold, we categorise the supply chain along
two dimensions—one based on order fulfillment strategy and the other based on work
flow. The former was discussed in the context of understanding and responding to
customers and their expectations. We now discuss the catagorisation based on work
flow [M. M. Srinivasan and J. M. Reeve, 2006].
y
nl
3.2.7 Categorisation Based on Work Flow: The V, A, and T
O
Configurations
The flow of work through a production process has a direct impact on how the
n
supply chain is designed, built, and managed. Practically every process can be
io
categorised as belonging to one of three types of work flow, V, A, and T, or some
combination of these three types. They are described below.
at
The V-Type Flow
ul
A V-type flow occurs when a few basic raw materials are processed into a variety of
end items. As shown in Figure 3.5, the product flow diagram resembles the letter V,
irc
hence the name. V-type supply chains are characteristic of food processing, metals,
chemicals, paper, and other continuous or batch processing supply chains.
C
In a V-type supply chain the upstream elements are fairly uniform and simplified.
As the product moves downstream it is split into different specifications, product
codes, or sku’s. In addition, V-type supply chains will also split the product into
d
geographic locations in the downstream. For example, in papermaking, a common

ite
chip stream can be split into either bleached or unbleached pulp. These two streams
can be produced into a wide number of grades and colours on a paper machine. After
im
the paper machine, the paper can be converted into an even wider variety of cut sizes
and package counts. Thus, the downstream end of the supply chain presents the
greatest opportunity for mis-allocation in a V-type supply chain. In fact, it is safe to
rL
say that each individual divergence point represents an opportunity for material mis-
allocation.
Fo
The mis-allocations are exacerbated by the desire to enhance efficiency, resulting in

material being released earlier than needed. These actions result in excess inventory of
some products and a shortage of others as highlighted earlier. As observed by Umble
and Srikanth [1997], managers of plants that have a V-type flow are often puzzled
when, despite holding large finished-goods inventory, they have to scramble to meet
market requirements. They level the blame on the customer, pointing to the
constantly changing demand pattern. They do not recognise that while demand
changes do occur, most of the problems are self-inflicted.
Figure 3.5: The V-Type Supply Chain
y
nl
O
n
io
at
ul
irc
C
d
ite
The A-Type Flow

im
A-type supply chains are characteristic of aerospace products, capital equipment, and
consumer electronics. The A-type structure is the opposite of the V-type structure as
shown in Figure 3.6. A-type supply chains have their greatest complexity on the
rL
upstream end. The upstream is characterised by hundreds and maybe even thousands
of individual parts and components, that move through a variety of supply chain
Fo
elements to a point of final assembly. After the product is assembled, often the
distribution is fairly straight-forward. For example, a commercial airliner is very
complex on the upstream end of the supply chain. Individual parts, such as flat
pattern wing parts, are fabricated and assembled on to wings, which are then
assembled to the fuselage. Complexity converges to final assembly. After assembly the
downstream distribution end of the A-type supply chain is very simple.
Figure 3.6: The A-Type Supply Chain
y
nl
O
n
io
at
ul
irc
C
Thus, the A-type supply chain manager’s focus is upon upstream mis-allocation.
Such mis-allocation can occur when upstream parts are being produced for forecasted
d
product that is eventually not sold. The finished goods inventory, in this case,
ite
represents the mis-allocated capacity and materials for every sub-component and
component in that finished item. These mis-allocations, of course, have domino
im
impacts on other components that were not made (but should have been). A-type
supply chains attempt to protect against assembly mis-allocation by committing to
excess inventory of component parts. This, however, is an unattractive solution
rL
because it can lead to obsolescing parts, and hence forever losing the materials and
capacity represented by those parts.
Fo
The A-type supply chain will have the same problems experienced by V-type
supply chains when products are processed in large batches. This causes a wave-like
flow of material, further aggravated by the bullwhip effect. The use of large batches at
one level in the supply chain causes downstream customers to receive material in an
erratic fashion, while causing serious stress on upstream suppliers. Some of the
solutions we will discuss for the V-type supply chain apply equally well to the A-type.
The T-Type Flow

As the name implies, the T-type flow takes place when the flow across the supply
chain resembles the letter T and it reflects a situation where a limited number of
components are assembled into a wide variety of end items. T-type supply chains are
y
commonly found in assemble to order (ATO) environments, where customer lead
times are relatively short. The T-type supply chain has elements of both A and V,
nl
together. On the one hand, the T-type and the V-type flows share the common
O
characteristic of divergence, although the divergence is concentrated in the final
stages for the T-type supply chain. With a T-type flow, it is thus relatively much easier
to avoid mis-allocation by delaying commitment of resources and materials until
n
receipt of a firm order. The T-type and A-type flows are similar in the sense that they
io
are both dominated by many interactions that occur at the assembly stage.
at
3.2.8 Characterisation Based on Fulfillment Strategy: BTS,
ATO, BTO, ETO ul
While the nature of the product will dictate the V-A-T workflow configuration,
irc
customer requirements will influence the demand fulfillment strategy. In the ideal
world, there is a marriage: as observed above, a supply chain with the T-type flow is
C
readily amenable to an ATO mode of operation. Similarly, it is tempting to

conjecture that a supply chain with a V-type flow is amenable to a BTO operation
while an A-type flow is more amenable to a BTS mode of operation because of the
d
relatively fewer products being processed. Indeed, as noted by Schragenheim and

ite
Dettmer [2001], some authors try to link A-type flows with BTS and V-type flows
with BTO. However, aside from the close relationship between the T-type flow and
im
the ATO mode of operation, in practice there appears to be little correlation between
the type of flow and the demand fulfillment strategy. For example, an A-type supply
chain may build to stock (automobiles) or build to order (construction equipment).
rL
Likewise, a V-type supply chain is used with BTO strategy (specialty chemicals) as
well as with a BTS strategy (paper).
Fo
Applying Lean Principles to a BTS V-Type Process

Consider a beverage bottling facility. The customer demands bottled beverages
instantaneously, and so the supply chain must be configured as BTS. In addition, the
product and process streams are such that the process has a V configuration. That is,
the scope of product is least on the upstream (flavours and containers), and becomes
greater by moving downstream to sku’s (stockeeping unit flavour and package
combinations) and locations. Under this design, the supply chain is most vulnerable
to mis-allocating capacity and materials to downstream sku’s and locations, since
product must be speculatively placed. Given this constraint, the supply chain
responds best when the execution time interval is the least.
y
The objective of the beverage company supply chain is to build and deliver a
product mix within a time interval that matches the demand mix. Thus, if all sku’s are
nl
sold every day from every location, then the optimal supply chain design attempts to
O
produce and deliver every sku to every location every day. This means the can
supplier must produce every can specification every day, the plant must bottle every
product every day, and distribution must deliver every product every day to every
n
customer. This is the basics of achieving flow. Without this rapid execution cycle, the
io
planning process must necessarily resort to more speculative placements. The design
follows the basic principle of minimising the impact of variation to very short time
at
intervals. Alternatively stated, the longer the time intervals, the greater the forecast
errors. ul
Once the upstream executes within the time interval of demand, the supply chain
will execute production and delivery from replenishment signals. That is, the actual
irc
sales of sku’s from the shelves will provide the signals for upstream replenishment.
These signals will be generated either by the route truck drivers, account
C
representatives, or POS information. Thus, the complete supply chain is responding

to demand, rather than anticipating demand with production.
d
This design has significant implications for the bottling plant. The plant is no
ite
longer concerned with plant efficiency. Indeed, the bottling plant may be inefficient
in the classic sense, because its only responsibility will be to replenish what was sold
during the previous day, regardless of the available capacity. Rather, the plant’s
im
improvement focus is to maintain capacity for immediate activation from

replenishment demand signals. In addition, the facility must achieve fast flavour and
rL
container changeovers. These changeovers must be drastically reduced in order for

the plant to economically produce every sku (or the Pareto 80 per cent) every day.
Flow and lead time are closely related. The lack of flow, or any imbalance in flow,
Fo
results in long lead times; conversely, long lead times are symptomatic of poor flow.
As we have observed before, long lead times negatively affect supply chain
performance in a number of ways. If the supply chain is not very responsive due to
long lead times, then it increases the volatility in demand. If one can respond to
changes in customer demand quickly, then the fluctuations in production and
inventory levels are considerably reduced. The ability to respond quickly to market
changes allows small adjustments in production levels to meet market demand and
promotes flow. On the other hand, if intervention must be delayed to accommodate
an inflexible system, management is forced to make much larger changes in
production levels, and that impedes flow. Therefore, one way to create flow is to focus
on reducing lead times.
y
To reiterate, lowering lead times leads to lower inventories in the pipeline that in
nl
turn results in lower lead times, generating a virtuous cycle. Similarly, lowering lead
O
times creates more flow in the supply chain which in turn reduces lead times, again
generating a virtuous cycle. When lead times are reduced the organisation is also in a
better position to implement pull systems that produce to actual customer demand
n
rather than production based on forecasts. There are a number of ways in which the
io
supply chain can reduce lead times and thereby increase flow. One way to reduce
customer delivery lead times is to operate in a Build To Stock (BTS) mode. This way,
at
the customer may pull from finished goods for an immediate response. However, as
we indicated earlier, this often mis-allocates capacity and resources, because the
ul
organisation could end up producing, say, or light beer when customers are
demanding premium lager beer. Adopting a mixed-model production schedule will
irc
reduce lead times. If every product is produced during every production period, then
the product batch sizes are reduced. As a consequence, the organisation does not have
C
to store as much inventory.

As noted in Figure 3.1, to create flow in the supply chain, you first need to identify
d
the products that should be produced in the supply chain before attempting to create
flow within the four walls of the organisation and flow across the value stream. The
ite
Theory of Constraints provides a way to identify the products that will help the
organisation benefit the most. Discussion of the Theory of Constraints is beyond the
im
scope of this book and the reader is referred to [Srinivasan, 2004] for this purpose.
rL
3.2.9 Develop Metrics Using a Systems Perspective

When the CEO of an organisation is posed the question, ‘How are we doing?’ by the
Fo
board of directors, what would the typical response be? The CEO would probably
present the performance of the organisation in terms of measures such as return on
investment, profitability, inventory turns and material costs. These are, no doubt,
important measures to gauge the financial health of the organisation. However, all of
these are internal measures, and may not adequately address the performance of the
organisation in the future. We indicated earlier that the battle has shifted: from
rivalry between organisations to rivalry between supply chains. Outstanding

performance at one location in the supply chain is no longer adequate if the rest of
the supply chain members are not up to par. The financial health of an organisation
in the supply chain is directly affected by the overall performance of the chain. In the
future, organisations must not only focus on performance within the organisation
y
but on performance across the supply chain as well. That requires a radical shift from
nl
a local optimisation mindset to a global optimisation mindset.
O
The establishment of metrics is of utmost importance since they will drive the
behaviour of the organisations in the supply chain. These metrics must be developed
with a systems perspective: as the beer game demonstrates, a locally managed supply
n
chain is inherently unstable. The next chapter discusses performance metrics in more
io
detail. At this point we will just note that one approach to developing these metrics is
provided by the Theory of Constraints(TOC). That is, when developing the metrics,
at
it is worth considering the following questions (these are simple restatements of the
questions that Jonah, the TOC guru poses to Alex Rogo in the book by E. M.
ul
Goldratt and J. Cox [2004]):
irc
➣ Does the metric help sell more products, profitably—that is, will the metric
help increase T ?
➣ Does the metric help reduce investments in resources—that is, will the metric
C
help reduce I?
➣ Does the metric help reduce payments/expenses, long-term—that is, will the
d
metric help reduce OE (long-term)?

ite
im
CONCLUSIONS
rL
Ultimately, the true competitive edge is realised only when the members in the
supply chain understand and agree to work with the following basic principles:
➣ The idea is not to simply benchmark your organisation against the
Fo
competition. That will result in the organisation adopting the same practices
(good and bad) of the competitor (the ‘leader’) and promote a ‘follower’
mentality. Instead, organisations can gain a tremendous competitive
advantage by adopting best-in-class practices that cut across industries.
➣ The phrase, ‘maximise external variety with minimal internal variety,’
succinctly captures the approach towards designing the process used to
produce the products. The RAP principle (‘keep the material as Raw As
Possible’) is a very convenient way to meet customer demand quickly without
storing a lot of finished goods inventory. Adopting the RAP principle also
results in reduced variation in the system because aggregated demand has
significantly less variation than the individual demands.
y
➣ Buffer the variation in demand with capacity, not inventory. As far as
nl
possible, the demand variation should be met with additional capacity. While
O
finished goods may allow the organisation to service customers faster, very
often the organisation ends up carrying the wrong kinds of products.
➣ If you have to hold inventory, it should be located strategically in the supply
n
chain. Use the RAP principle to determine how far upstream in the supply
io
chain it can be located so that you can still meet customer lead time
expectations.
at
➣ Use forecasts to plan and pull to execute. A system that reacts to pull signals
ul
will have less variation than a comparable system that adopts a push mode of
operation.
irc
➣ Develop metrics using a systems perspective.
C
d
Goldratt, E. M. and J. Cox, (2004), The Goal: A Process of Ongoing Improvement, North
ite
River Press Publishing Company (3rd Edition).

Hill, T., (2000), Manufacturing Strategy, Boston, MA: Irwin McGraw-Hill.
im
Ohno, T., (1988), Toyota Production System: Beyond Large-Scale Production, Cambridge,
MA: Productivity Press.
Reeve, J. M. and M. M. Srinivasan, (2005), Which Supply Chain Design is Right for You?
rL
Supply Chain Management Review, May/June 2005, pp. 50–57.

Rother, M. and R. Harris, (2001), Creating Continuous Flow, Lean Organisation Institute
Inc.
Fo
Rother, M. and J. Shook, (1999), Learning to See, Lean Organisation Institute Inc.
Schragenheim, E. and H. W. Dettmer, (2001), Manufacturing at Warp Speed, Boca Raton,
FL: St. Lucie Press/APICS Series on Constraints Management.
Senge, P. M., (1994), The Fifth Discipline: The Art and Practice of the Learning Organisation,
New York: Doubleday.
Srinivasan, M. M., (2004), Streamlined: 14 Principles for Building and Managing the Lean
Supply Chain, Mason, Ohio: Thomson Publishers.
Srinivasan, M. M. and J. M. Reeve, (2006), The Lean Supply Chain: The Path to
Excellence, Handbook of Global Supply Chain Management, Sage Publications.
y
Umble, M. and M. L. Srikanth, (1997), Synchronous Management, Volume Two, Guilford,
CT: The Spectrum Publishing Company.
nl
Womack, J. P. and D. T. Jones, (2002), Seeing the Whole: Mapping the Extended Value
O
Stream, Lean Organisation Institute Inc.
Case references: The idea of a value stream is useful in several cases, including those that
n
have a marketing emphasis, such as CONCOR.
io
at
ul
1. Discuss some of the salient issues in supply chain design .
2. What are order winners and order qualifiers? Why are they important?
irc
3. Give examples of organisations or supply chains that operate using a) a build-to-stock
strategy, b) a build-to-order strategy.
C
4. What is the primary benefit of using a customer time-based-demand profile?

5. What are the benefits of competitive benchmarking? What are some drawbacks of
d
competitive benchmarking?
ite
6. What are some benefits of functional benchmarking? Can you think of any problems
with functional benchmarking?
7. Can you think of specific examples where an organisation (or a set of organisations)
im
has gained tremendous competitive advantage by benchmarking a process (or a set of

processes) completely outside their industry?
rL
8. What are the ways by which an organisations can mitigate the volatility of demand it
experiences?
9. Give specific industry examples of the RAP principle and how it benefits the supply
Fo
chain operations of the participating industry.

10. Give two examples each of industries that have an A-type supply chain, a V-type
supply chain.
y
CHAPTER
nl
Performance Measurement
4
O
n
CHAPTER OUTLINE
io
Introduction
at
4.1 Measuring Supply Chain Performance
4.2 The Prisoners Dilemma ul
4.3 The Integrity Motors Case
4.3.1 Current Mode of Operation
irc
4.3.2 Integrity Motors Embarks on a Dell Strategy
4.4 Evolving Supply Chain Metrics
4.5 Performance Monitoring
C
4.5.1 Benchmarking for Continuous Improvement

4.5.2 Balanced Scorecard
d
4.5.3 The SCOR Model

4.5.4 Standard Costing and Activity-Based Costing
ite
4.6 Competences and Performance

Conclusions
im
References
Exercises
rL
Fo
Performance Measurement 91
INTRODUCTION
T
he performance of a supply often uses the quote, ‘Tell me how you
chain is the result of policies will measure me and I will tell you how
y
and procedures that drive I’ll behave,’ to emphasise the impor-
various critical segments of the supply tance of an effective measurement
nl
chain. In all probability the current system. In a related vein, it is often said
strategy and its resulting structure are that what cannot be measured cannot
O
a consequence of the individual seg- be improved, and so there is a persis-
ments of the supply chain making iso- tent desire to be able to measure ap-
lated decisions that make sense to propriate indices that deal with supply
them from their local perspective. chain performance.
n
These individual decisions result in an A question we need to address
io
unsynchronised supply chain that is carefully is, ‘How can we design a pro-
characterised by long lead times and cess for managing organisations con-
many pockets of inventory. Traditional,
at
sistent with the fact that these
locally focused, measures of cost and organisations are components of com-
other performance measures are in- plex and highly interconnected sys-
complete and lead to sub-optimal de-
cisions. Metrics that are carefully
ul
tems?’ And a related question is, ‘How
can we devise metrics that can help
evolved, with the goal of improving
irc
these organisations better manage
supply chain performance and com- their processes?’ Based on these two
petitiveness, can have a positive effect questions, we conclude that a key ob-
because they set in place these poli- jective of supply chain metrics is to
C
cies and procedures. Furthermore, provide a basis for evaluating the per-
over time, people and systems react formance of the whole supply chain as
according to how they are monitored one system.
d
and measured. In fact, Goldratt [1988]

ite
im
rL
Fo
4.1 MEASURING SUPPLY CHAIN PERFORMANCE

Ideally, the supply chain should meet a number of different objectives, some of which
may seem to be in direct conflict at least at first glance. Some common objectives are to:
1. Optimise customer service
y
nl
2. Reduce time to market
3. Meet delivery performance parameters
O
4. Improve quality
5. Minimise costs and operating capital
n
6. Reduce the fixed assets employed
io
7. Minimise inventory
at
8. Maximise flexibility/agility
Numerous measurement services and benchmarks for meeting these objectives are
ul
provided by industry forums, consulting firms and academic institutions. Sector
specific norms may be used for comparison although, as discussed in Chapter 3,
irc
significant insights can be gained by cross-industry comparisons. The following
metrics have been suggested, to meet some of these objectives:
C
Customer Service Level

d
Fill rate, Perfect order percentage, Lead time, On-time delivery, Stock outs, Delivery
consistency, Response time to Enquiries, Customer complaints, Sales force
ite
complaints, Overall satisfaction, Price competitiveness.

im
Cost Measures
Total cost, Cost per unit at each stage, Cost/sales, Inbound freight, Outbound
rL
freight, Inventory carrying cost, Inventory turn over ratio, Cost of returned goods,
Cost of damage, Cost of back orders etc.
Fo
Quality Measures
Defect rates, Order entry accuracy, Billing errors, Number of customer returns,
Shipping accuracy.
Productivity Measures
Partial factor and total productivities.
Asset Management Measures

Return on Investment (ROI), Return on Assets (ROA), Profitability of partners in
the chain etc.
NPPD (New Product and Process Development)
y
Developmental lead time, Innovativeness, Frequency of new product introduction
nl
etc.
O
Besides meeting desired objectives, these metrics should ideally also facilitate
business processes involved in supply chain management. These business processes
include:
n
io
1. Customer relationship management
2. Order/Demand management
at
3. Production management
4. Supply management
5. Distribution management
ul
irc
6. New product and process development (NPPD)
7. Reverse logistics
C
Clearly there are some challenges. Developing metrics to accommodate potentially

conflicting objectives that, at the same time, also address the different business
d
processes is certainly a difficult task. In addition, these metrics should mesh with
ite
those developed for individual firms within the supply chain; and evolving “local”
metrics for individual firms within the supply chain, in and of itself, presents a
challenge. So, is the notion of “supply chain” metrics really practical or are we just
im
trying to pursue an unrealistic, unattainable goal?

To answer this question, let us simply focus on organisations that are in business to
rL
make money for their stockholders, and identify some financial metrics that would
be, more or less, universally acceptable to the individual firms in the supply chain.
Clearly, Net Profit would be a universally acceptable financial metric. Another
Fo
financial metric would be R. In the previous chapters, we introduced three metrics: T,

I, and OE, within a Theory of Constraints (TOC) context. Note that Net Profit = T
– OE, and ROA is, simply, (T – OE) / I. Thus, TOC presents a good framework to
begin discussing metrics from a systemic or a supply chain perspective. Indeed, in
Section 4.3, we present a case where the members in a supply chain evolve a strategy
to jointly improve their individual net profits and ROAs.
Although these financial metrics can be acceptable to all supply chain partners, the
reader may observe that unless the individual firms in a supply chain cooperate, it
would be unreasonable to expect these firms to evolve organisational metrics that
completely align with supply chain metrics. Co-operation between participating
members in the supply chain is often a complex issue due to a lack of trust among the
y
members. However, there are some clearly discernible mutual benefits to the supply
nl
chain members that should provide a powerful incentive for these members to jointly
evolve metrics that are synergistic. That requires us to first identify where
O
opportunities for such synergies lie, and then to evolve metrics that exploit these
synergies. We will show how this is possible for multiple organisations in the supply
n
chain. To that end, we introduce and discuss a game popularly known as the
io
Prisoner’s Dilemma.
at
4.2 THE PRISONER’S DILEMMA
ul
The Prisoner’s Dilemma requires us to introduce two terms: the zero-sum game and
the non-zero-sum game. In game theory, the zero-sum game describes a situation
irc
where one player’s gain (or loss) is exactly offset by the loss (or gain) of the other
player. The term, zero-sum, thus describes a situation where a gain for one set of
players is offset by a corresponding loss for another set of players. Chess is a zero-sum
C
game because there is exactly one winner and one loser—it is impossible for both
players to win (or for both to lose). Poker is another example of a zero-sum game, if
d
the house’s cut is ignored. In a more general setting, zero-sum games can also be
ite
thought of as constant-sum games where the benefits and losses to all players add up
to the same value. In this regard, cutting a cake is a zero-sum (more precisely a
constant-sum) situation because a larger piece given to one person implies there is less
im
cake available to share among the others.

In contrast, non-zero-sum games describe situations where the sum of the
rL
aggregate gains and losses of the players is either less than, or more than zero. For
example, Monopoly (or Trade) can be viewed as non-zero-sum games if they are not
played with the intention of having just one winner. That is because all participants
Fo
can win property from the bank. In principle, it is possible for two players in this
game to reach an agreement and help each other in gathering a maximum amount
from the bank although that is not really the intent of the game.
The Prisoner’s Dilemma is, arguably, the most famous among all non-zero-sum
games. In this game, two men fleeing from the scene of an armed robbery, firearms in
hand, are apprehended and put in jail. We will refer to these two men as Prisoner A
and Prisoner B. The prosecutor strongly believes these two men jointly attempted the
armed robbery. However, since the crime was not captured on hidden camera, she can
obtain a conviction only if at least one of them confesses. To obtain a conviction, the
prosecutor first orders the prisoners to be held in separate cells. She then visits each
prisoner individually with the following offer. ‘You can confess or choose not to
y
confess. If you confess and your partner remains silent, then you will walk free while
nl
your partner will undergo a jail sentence of 10 years. In the same manner, if your
partner confesses but you do not, then he walks while you do 10 years in jail. If you
O
both confess, both will go to jail but I will make sure you both get out of jail in six
years. If neither of you confess, then you will only have to face a token imprisonment
n
for one year for possession of firearms.’ Table 4.1 presents this situation graphically.
io
What do you think the prisoners will do in this situation?
at
Table 4.1: The Prisoner’s Dilemma
ul
irc
C
d
ite
im
The surprising truth about the game is that no matter what the other prisoner
does, it appears to benefit each prisoner to confess. To see this, consider the
rL
perspective of a prisoner, say, Prisoner A. Reflecting on the prosecutor’s offer, he

realises that if Prisoner B confesses, then he does better for himself by confessing for
otherwise he will spend 10 years in jail if he chooses not to confess. If Prisoner B does
Fo
not confess, then again he realises he will do even better for himself by confessing as
he will be set free. In other words, Prisoner A realises quickly that confessing
dominates not confessing. The same thought process would lead Prisoner B also to
confess. But this outcome, where both confess, is arguably the worst outcome,
resulting in a combined jail term of 12 person-prison years. If both of them had trust
in each other and not confessed, they would have ended up with one year each in
prison (2 person-prison years). In fact, even if one of them had confessed while the
other did not, the combined jail term would have been 10 person-prison years.
The Prisoner’s Dilemma shows that for many practical situations it is not a zero-
y
sum game and that there are benefits to evolving joint strategies that mutually benefit
multiple players in a supply chain. More to the point, it is possible for multiple
nl
organisations to jointly arrive at metrics that positively affect their individual
O
performances rather than coming up with metrics derived in isolation that could, in
fact, hurt their performance in the long run. That is, in the context of supply chains,
it is possible to evolve strategies that result in a non-zero-sum game, benefiting
n
multiple parties. We now present a case to demonstrate how a systems perspective
io
facilitates the evolution of such a strategy.
at
4.3 THE INTEGRITY MOTORS CASE
ul
The Integrity Motors case (Reeve, 2002) demonstrates how the Build-To-Order
(BTO) strategy we discussed in Chapter 3 helps improve the financial performance of
irc
multiple firms simultaneously. The supply chain in the case has three firms involved
in the manufacture and delivery of automobiles. We will first present the current
C
mode of operation for the three firms and then present the scenario after the three
firms work together to help support a BTO strategy.
d
Integrity Motors manufactures and sells automobiles. The supply chain involves
over 1,200 different vendors who supply materials and parts to Integrity Motor’s
ite
assembly plants. The assembly plants produce automobiles sold through a dealer
network. Integrity Motors uses rail and truck transportation for their inbound and
im
outbound transportation needs. In this case we focus on outbound transportation.

Southern Railroad is a private enterprise that moves vehicles to a distribution and
mixing centre, whereupon vehicles are delivered to local dealers (via truck). Integrity’s
rL
transportation department interfaces with the railroad and motor carriers with orders
for equipment. For simplicity, in this case we ignore the motor carriers and assume
Fo
that the supply chain involves three major parties: Integrity Motors, Southern
Railroad and the Dealers.
4.3.1 Current Mode of Operation

With the current mode of operation, each dealer receives a monthly allotment of
automobiles that are sold to consumers at the dealer’s premises. Most customers
purchase cars from dealer’s stocks. Integrity is never certain of its daily needs, and is
therefore unable to provide the railroad (Southern Railroad) a good assessment of its
needs. As a result, the railroad reserves railcar capacity to handle demand surges.
These railcars are not constructively placed and thus are not subject to demurrage.
Vehicles in-transit are owned by Integrity Motors. Figure 4.1 shows the supply chain.
y
Lead-times for the current mode of operation are presented in the figure.
nl
The following information is provided for each of the supply chain partners. This
O
information can be used to determine the income from operations (net profit), and
expected pre-tax return on invested capital or ROA, for the members in the supply
chain.
n
Average Dealer:
io
1. Provides an average discount from manufacturer’s suggested retail price of
at
US$2,000.
2. Operating costs equal 10 per cent of Cost of Goods Sold (COGS).
ul
3. Dealer’s cost: US$25,000.
irc
Additional Statistics:
4· Average dealer sold 480 vehicles per year
C
5. Days in Consumer Sales

d
Inventory (see Figure 4.1) 60 days

ite
Accounts Payable (given) 45 days

Accounts Receivable (all cash sales) 0
im
6. Average total fixed plant: US$4,000,000 of which 60 per cent supports

vehicle sales (the remainder supports vehicle service).
rL
Integrity Motors
Integrity sells 27,50,000 vehicles per year.
Fo
1. Total dealer sales: US$68,75,00,00,000 or 27,50,000 cars, or US$25,000 per

vehicle
2. Rebate average: US$1,100 per car
3. COGS: 85 per cent of dealer price
4. Selling, General and Administrative (SGA) expenses: 8 per cent of dealer
price
5. Freight of US$350 per vehicle includes US$200 to motor carrier and

US$150 to railroad. Integrity is responsible for freight and freight on board
(F.O.B. dealer).
y
Figure 4.1: Supply Chain for Integrity Motors—Current Mode of Operations
nl
O
Demand
Forecasting
n
Market Production Raw Factory Shipping Dealer The
Material Production Sales and
Research Planning Inventory by Rail Inventory rebares
Logistics and Truck Customer
io
Design
Research
at
Macro Cycle-time Demand Schedule Demand Customer
Lead-time Micro Satisfaction Level?
Demand Uncertainty Uncertainty Uncertainty Uncertainty Demand
Uncertainty Uncertainty and and and and Uncertainty
Variation Variation ul Variation Variation Likelihood of Repeat
Purchase?
Selection and choice?

irc
Raw WIP F.G. Transport F.G.
Inventory material Inventory Inventory Time Inventory
Inventory ~3 weeks ~3 weeks ~14 days ~60 days
~2 weeks
76 per cent
customers not
C
Average 19 Week Lead-Time fully satisfied with

the purchase
Working Transport Real Estate experience
Capital Investment Investment
Vehicle
d
Working Factory Eqpt. Investment

Fixed Assets
Capital Work Capital
ite
Space Space
im
Additional financial statistics:

rL
Days in Dealer Sales

Inventory (see Figure 4.1) 70 days
Accounts Payable 30 days
Fo
Accounts Receivable (see dealer) 45 days

Factory Property Plant and Equipment: US$16,50,00,00,000
Southern Railroad
Southern Railroad’s costs, calculated in revenue ton mile (rtm) terms are as follows:
Cost Category Cost Per Revenue Ton Mile

Wage US$ 0.0150
y
nl
Fuel US$ 0.0120
Equipment costs US$ 0.0080
O
Roadway maintenance/depreciation US$ 0.0040
S, G, and A US$ 0.0225
n
Terminal related costs are expressed in railcar terms:
io
Spotting and loading costs: US$128 per railcar
at
Switching and classifying cost: US$64 per railcar switch
Additional statistics: ul
Average weight of a vehicle 1.25 tons
irc
Average distance moved (to mixing facility) 1,600 miles
Average number of switches per car movement 3 switches
C
Number of vehicles per railcar 16 vehicles

Total revenue ton miles 182 billion
d
Revenue per VIN number (vehicle) is same as Integrity’s freight cost, US$150 per
ite
vehicle.
im
Additional financial statistics:
Revenue (current year) US$6,17,00,00,000

rL
Selected balance sheet items End of current year

Accounts receivable US$47,50,00,000
Fo
Materials and supplies US$9,00,00,000

Accounts payable US$84,80,00,000
Property and Equipment US$18,30,00,00,000
The following table, provided for a typical dealer, uses the above data to obtain
income from operations and the ROA for the dealer. Use a similar approach to
determine these numbers for Integrity Motors and for Southern Railroad. To help
you fill out the rest of the numbers in the latter two tables, the answers for Income
from operations per vehicle and the ROA are provided for Integrity Motors and for
Southern Railroad.
y
Dealer: Current Mode of Operation Calculation Explanation
nl
Manufacturer’s suggested price US$30,000 Given
Less: dealer discount US$2,000
O
Customer price US$28,000
Less: Dealer COGS US$25,000
n
Less: Operating Costs US$2,500 10 per cent x $25,000
io
Income from operations per vehicle US$500
at
Investment per vehicle
Inventory US$4,109 60/365 *($25,000)
Accounts Payable
Accounts Receivable
ul US$(3,082)
0
45/365 *($25,000)
irc
Dealer Property and Equipment (sales only) US$5,000 ($4MM*60 per cent)/480 vehicles
Total investment per vehicle US$6,027
C
Pre-tax ROIC (ROA) 8.29 per cent $500/$6,027
Integrity Motors: Current Mode of Operation Calculation Explanation

d
Dealer price US$25,000 Given

ite
Less: Rebate
Net revenue
im
Cost of goods sold

Gross profit
S, G, and A
rL
Freight
Total operating costs and expenses
Fo

Inventory
Less: Accounts payable
Accounts receivable
Factory Property, Plant, and Equipment

Pre-tax ROIC (ROA) 2.63 per cent
y
Southern Railroad: Current Mode Operation Calculation Explanation
nl
Freight revenue US$150 From Integrity (350–200)
Wages
O
Fuel
Equipment costs
n
Spotting and loading
io
Roadway maintenance/depreciation
Switching and classifying costs
at
S, G, and A

ul
irc
Materials and Supplies
C
Accounts receivable
Property and equipment (rtm allocation)
Total investment per vehicle US$194
d

ite
4.3.2 Integrity Motors Embarks on a ‘Dell Strategy’

im
Integrity Motors is planning a significant change in its supply chain strategy, moving
from a BTS to a BTO car manufacturer. In this strategy, customers will order cars
rL
from a kiosk (an internet web site). The order will be electronically transmitted to
Integrity, and an automobile built and delivered to the customer’s specification. With
this strategy customers can purchase cars ‘their way.’ Vehicles will need to be designed
Fo
under a modular concept and the assembly plant will need to have ‘plug-and-play’
capabilities. The vehicle will need to be designed to transform nearly 3,000 parts to
around 30 modules and options. Under this plan the following operating statistics
are expected:
Integrity’s Build-to-Order Plan
Total days in materials and work-in-process 2 days

Total days in finished goods 2 days
Total days in accounts receivable –5 days
y
Total days in accounts payable 30 days
nl
Shipping time (premium) 3 days
Elapsed time order-to-delivery 7 days
O
The –5 days in accounts receivable reflects the fact that Integrity collects money
from the customer before it ships the product.
n
Additional Assumptions:
io
Dealer
at
1. Dealer earns a 3 per cent commission on sales price to consumer for sale
support. ul
2. Dealer property investment for sales support reduced by 75 per cent.
irc
3. Vehicle is paid for by consumer at time of BTO order. Dealer does not pay
Integrity directly, but acts as a commission dealer without taking ownership
C
of vehicles.
4. Consumer price for BTO is US$26,000 and not subject to rebates. (Price to
d
consumer is better than previous case of US$30,000 – discount US$2,000 –

rebate US$1,100 = US$26,900).
ite
5. Dealer operating costs are 70 per cent of commission revenue. The reduction
in selling expenses from reduced sales overhead costs (interest, space,
im
property tax, etc) is offset by higher selling expenses.
Integrity
rL
1. To simplify, assume all 2.75 million vehicle sales are BTO.

2. IT, engineering, and manufacturing infrastructure investment
Fo
US$1,10,00,00,000.
3. Web site investment paid by integrity, US$1,00,000 per dealer.
4. Premium shipping costs, US$500 per vehicle, US$225 to railroad.
5. Additional general expenses for new BTO ordering system (including
depreciation), US$55,00,00,000. Add appropriate amount to conventional
SGA.
6. Additional annual manufacturing costs for new BTO system (including

depreciation): US$82,50,00,000. Add appropriate amount to conventional
COGS.
Southern Railroad
y
1. Railroad earns US$225 per vehicle shipment for premium shipment service.
nl
2. Additional Railroad investment to support ‘just-in-time’ (JIT) rail service:
O
US$5,00,00,000.
3. Additional railroad operating costs to support JIT rail service US$.02 per rtm.
n
4. Assume 12 per cent of railroad rtm traffic (182 billion rtm’s) is related to
automobiles. Assume 25 per cent of this traffic is Integrity’s fast service JIT
io
rail delivery.
at
5. With conventional rail transport a vehicle was on the railroad for 10 days.
With the JIT plan a vehicle is on the railroad for 2 days. US$3 billion of the
ul
total US$18.3 billion property, plant, and equipment investment is due to
railcars. Due to the higher velocity, railcar assets are used more efficiently
irc
than with a conventional approach. That is, Southern Railroad could make
the same revenue shipments with less railcar assets (or ship more volume with
C
the same assets). Adjust the assets for this improvement.
Dealer: BTO Scenario Calculation Explanation

d
Dealer’s commission US$780 3 per cent x US$26,000

ite
Less: Operating Costs

im

rL
Inventory —
Accounts Payable
Accounts Receivable
Fo
Dealer Property and Equipment (sales only)

Integrity Motors: BTO Scenario Calculation Explanation
Revenue per vehicle US$26,000 Given

Cost of goods sold
Gross profit
S, G, and A
y
Commission to dealer
nl
Freight
Total operating costs and expenses
O
n
io
Inventory
at
Accounts receivable
Kiosk investment
Additional IT, Engineering, and Mftng.
Factory Property, Plant, and Equipment
ul
irc
C
Southern Railroad: BTO Scenario Calculation Explanation

d
Freight revenue US$225 Given

Wages
ite
Fuel
Equipment costs
im
Spotting and loading

Roadway maintenance/depreciation
Switching and classifying costs
rL
S, G, and A
Additional BTO operating costs
Fo

Materials and Supplies
Accounts receivable
Property and equipment (rtm allocation)

Additional B-T-O investment
Benefit from decreased railcar travel time
Total investment per vehicle US$186
Pre tax ROIC (ROA) 22.50 per cent
y
nl
Supporting calculations—Integrity revenue ton miles (rtm):
O
Total revenue ton miles on railroad 182,00,00,00,000
Per cent related to automobiles 0.12
n
Automobile revenue ton miles 21,84,00,00,000
Per cent related to Integrity 0.25
io
Integrity RTM’s 546,00,00,000
at
Supporting calculations—Rail Throughput Improvement
ul
Conventional vehicle time on rails (days) 10
irc
B-T-O vehicle time on the rails (days) 2
Improvement ratio 80 per cent
The calculations show that the ROA goes up significantly for all three supply chain
C
partners. In addition, the income from operations per vehicle also goes up for
Integrity Motors and Southern Railroad. No doubt, income from operations per
d
vehicle drops for the Dealer. However, the reader should note that the dealers now do
ite
not hold any inventory. It is now easily verified that the dealers will still come out
ahead (generate more profit) even if the released cash (from not having to carry
inventory) is invested at a rate of return greater than 8.29 per cent (the pre-tax ROA
im
under the current mode of operation).

What should be the driving metrics for the parties involved in the above exercise?
rL
Certainly, an important metric for Integrity Motors will be fill rate. For Southern
Railroad, on-time delivery will be an important metric. You can identify other
important metrics for the supply chain members that will help evolve their BTO
Fo
strategy.
4.4 EVOLVING SUPPLY CHAIN METRICS

The preceding case clearly demonstrates the value of having shared vision and shared
supply chain metrics. Indeed, many organisations have attempted to evolve supply
chain metrics. However, the lack of a systems perspective has often hindered their
efforts at identifying good metrics that would find ready support from other member
firms in the supply chain. As a consequence, these organisations often end up with
metrics that are very narrowly focused. Lambert and Pohlen [2001] indicate that
many measures currently identified as supply chain metrics are, in fact, simply
y
measures of internal logistics operations. In fact, typical metrics used to measure
nl
delivery performance, such as fill rate and on-time performance will be applied in a
different context if a systems perspective is adopted. To give a simple example, an
O
improvement in inventory turns by a retailer is likely to have a more significant
impact on overall supply chain performance than a corresponding improvement in
n
inventory turns by a supplier. The reasoning behind this observation is that as
io
inventory moves closer to the point of consumption, it increases in value.
It warrants repeating: optimising the separate links of the supply chain
at
independently does not optimise the supply chain. For example, consider an
organisation that produces and ships products in large batches. No doubt this
ul
organisation has minimised its production and transportation costs. However, it has
increased the inventory for the buyer. Moreover, viewed from the perspective of the
irc
supply chain, the long lead times created by big batches and shipping quantities are
very costly. They force large amounts of WIP in the supply chain, further reducing
C
the flexibility and responsiveness of the chain. So a question one can ask is, ‘Is there a
way to optimise inventory costs across the entire supply chain?’
d
Such a question would have been dismissed in the past as a theoretical exercise.
However, this question is rapidly changing from a theoretical to a practical one as
ite
managers of supply chains face increasing pressures on customer service and asset
performance. Sony, for instance, is acutely aware of the fact that any inventory of its
im
products at Best Buy and Circuit City ultimately affects its profitability if it remains
on the shelf for more than a few days. Sony has changed its delivery metric from ‘sell-
in’ to ‘sell-through.’ The difference is that the former metric allowed its Sales
rL
department to chalk up a sale when the product was shipped to the customer (Best
Buy, Circuit City, etc.) whereas the latter metric chalks up a sale only when the
Fo
product is sold and paid for. To give another example, Procter and Gamble uses its
VMI process to routinely measure both its own inventory and the downstream
inventory of its products.
An important set of supply chain metrics relate to speed—timeliness,
responsiveness and flexibility. About a decade ago, there was a major emphasis on
time-based competition. This emphasis was, and still is, very important. As we have
stated before, lead times play a crucial role in enhancing the competitiveness of the
supply chain. The metric of interest here is the supply chain lead time, not the lead
time for an organisation. It is measured simply by adding up the lead times at each
stage in the supply chain. Hausman [2002] reports on one high-tech organisation
that began to measure the supply chain lead time, once it was made aware of the
y
benefits of such a metric. Soon the organisation was able to reduce the supply chain
nl
lead time from 250 days to less than 190 days by some obvious, simple
improvements.
O
Another important metric that relates to speed is adherence to quoted delivery
time. Some organisations such as Dell Inc., for instance, go to extremes to meet this
n
metric. In fact, Dell has the motto of ‘Under-Promise, Over-Deliver.’ It quotes a five-
io
day delivery for a BTO product and typically ships the product in four days or less. Its
suppliers are well aware of Dell’s emphasis on on-time delivery performance and are
at
prepared to act accordingly.
From an individual firm’s point of view, supply chain metrics can move in the
ul
following directions:
irc
➣ Move from those focusing on one actor to those encompassing two actors,
capturing their interface, e.g. fill rates or order processing times, rather than
production volumes
C
➣ Measure outputs rather than inputs, e.g. number of customer complaints or

response time for a service call, rather than number of employees trained
d
➣ Focus on distributions rather than averages, e.g. measure age-wise profile of

ite
inventory rather than average, monitor service times above a certain limit, or
accounts receivables above a certain amount
im
➣ Focus on service measures in addition to product measures, e.g. delivery

time, in addition to production quality aspects
rL
A well-designed measurement system can significantly help align processes across

the supply chain, targeting the most profitable market segments. Some of the key
dimensions used to measure supply chain performance are speed, quality, service, and
Fo
asset management. Speed relates to timeliness, responsiveness and flexibility as noted

earlier. Quality can relate to a variety of measures including defect rate, reliability,
and appearance. Service relates to the ability to anticipate, capture, and fulfill
customer demand on time. Asset management relates to the efficiency with which the
supply chain is able to provide return on its assets.
Finally, properly designed supply chain metrics help clarify the inter-relationship
between organisation and supply chain performance. These metrics should
encourage co-operative behaviour of functions within the organisation (functional
integration) and across organisations (partnering in the supply chain). Supply chain
metrics help align activities and share in joint performance measurement information
y
to implement strategies that achieve supply chain objectives.
nl
O
4.5 PERFORMANCE MONITORING
To help organisations improve their overall supply chain performance, it is useful to
n
have a focused procedure based on a set of performance measures that will guide the
io
process of improvement. One set of suggested guidelines for such an activity is
presented below:
at
➣ Measure performance at every step of the supply chain, end-to-end—this
guards against part optimisation, which is sometimes ‘easy’ to achieve and
ul
can sometimes lead to misleading, short-term benefits.
irc
➣ Ask customers to control performance—this focuses on the overall
effectiveness of supply chain initiatives. Contrast this global suggestion with
the local one of measuring outputs (relevant to internal customers, perhaps),
C
rather than inputs.

➣ Focus on Key Performance Indicators (KPIs) for each organisational unit.
d
➣ Use both financial and operational performance measures—attempts to

ite
balance the short term goals of cash flow, profitability and viability, together
with the longer term strengths of process capability, flexibility and low waste.
im
➣ Monitor full supply chain costs—look at impacts in different parts of the

chain, including one’s partners
rL
➣ Take action based on KPIs—so that there is a purposefulness and momentum

to the measurement activity and people see the benefits of such activity.
Fo
4.5.1 Benchmarking for Continuous Improvement

Benchmarking is a process of explicitly comparing one’s performance to that of
others, with a view of identifying best practices leading to superior performance.
While comparisons within a certain sector or industry segment may be most direct
and meaningful, it is said that cross industry comparisons and looking for best
practices across different industries can push organisations to achieve significant

benefits. For example, one of the motivations for the Just-in-time manufacturing
revolution in manufacturing in the automobile sector has been the retailing model in
supermarkets, which is replenishment based.
y
Some steps suggested for a systematic benchmarking exercise are:
nl
➣ Define and classify processes
➣ Identify core processes ‘the vital few’
O
➣ Develop Metrics
➣ Monitor
n
➣ Global Benchmark
io
➣ Analyse performance gaps
at
➣ Pursue process improvement
4.5.2 Balanced Scorecard

ul
irc
The goal of the Balanced Scorecard is to align business activities to the vision and
strategy of the organisation, to improve internal and external communications, and
C
to monitor organisational performance against strategic goals. It was developed in the

early 1990s by Kaplan and Norton, 1992, and originally used to provide a framework
for measuring performance. Since then, this tool has evolved considerably. Now
d
many organisations both from industry and government view it as a strategic

ite
planning, management and control system for managing their business.

In its original form, balanced scorecards consisted of simple tables divided into
im
four sections to represent four different perspectives in business. These perspectives

were labeled ‘Financial,’ ‘Customer,’ ‘Internal Business Processes,’ and ‘Learning
rL
and Growth.’ The balanced scorecard approach thus intends to represent a balanced
representation of the financial, customer-related, operational, and human-resource-
related measures that are relevant to an organisation. Figure 4.2 presents a schematic
Fo
view of the balanced scorecard. As can be noted from the figure, the balanced
scorecard presents four perspectives on organisational management. Interpreting
these perspectives in a supply chain context is a useful exercise.
The Financial perspective considers the issue ‘How do we look to shareholders?’
The Customer perspective captures the key dimension of ‘How do customers see us?’
The Internal Business Process perspective asks the question ‘What must we excel at?’
The Learning and Growth perspective addresses the issue of ‘How can we continue
to create and improve value?’
Designing the Balanced Scorecard requires the selection of five or six good
measures for each perspective. Kaplan and Norton have observed that organisations
y
are using the scorecard to drive strategy execution, make strategy more operational,
align strategic initiatives, link financial budgets with strategy, and to conduct
nl
periodic strategic performance reviews to learn about and improve strategy. As
O
observed by Kaplan and Norton, the idea is to retain financial measures, but these
measures tell the story of past events and are inadequate for guiding and evaluating
the journey that organisations in the customer-centric era need to make to create
n
future value.
io
at
Figure 4.2: Balanced Scorecard Model (Schematic View)
[See Kaplan and Norton, 1992] ul
irc
C
d
ite
im
rL
Fo
4.5.3 The SCOR Model

The SCOR model was developed by the Supply-Chain Council (SCC). It integrates
the concepts of business process re-engineering, benchmarking and process
improvement into a cross functional framework. It tries to address management
y
issues at the enterprise level rather than a functional level based on five distinct
nl
management processes Plan, Source, Make, Deliver and Return. For each of these, an
increasingly holistic view of the supply chain is taken in four stages of management,
O
starting from a functional focus and ending with enterprise collaboration.
n
Figure 4.3: The SCOR Model, Schematic View
io
at
ul
irc
C
d
ite
im
rL
A process reference model has been designed for effective communication among
supply chain partners, ranging from supplier’s suppliers to customer’s customers. It is
claimed that the SCOR model can be used to describe, measure and evaluate supply
Fo
chain configurations. The model isolates key SCM processes and matches them
against industry-specific best practices and thereby provides users with a framework
for understanding where they need to make improvements
4.5.4 Standard Costing and Activity-Based Costing

Standard costing and activity-based costing (ABC) are tools for management and cost
control. In theory, they allow management by exception, and ABC has been referred
to as a powerful tool for measuring performance. To illustrate these two techniques,
y
we will use data from a fictitious organisation, CSN Inc.
nl
CSN Inc.
Three home-maintenance specialists, Cromby, Steele and Nash, have banded
O
together to form CSN Inc., based in Baton Rouge, Louisiana, where it is sunny, the
year round. They offer the following services: Plumbing, Window Cleaning,
n
Installing Gutter Guards, and Landscaping.
io
The salaried staff at CSN Inc., includes a General Manager (Cromby), the
Customer Service Representative (Steele), and the Front Office Person (Nash). The
at
total monthly wages paid to these 3 men (the salaried staff ) is US$18,000 including
benefits (administrative overhead). Other non-administrative overhead costs amount
ul
to US$9,000 per month consisting of: Rental space (US$3,500), truck fleet
maintenance to maintain three trucks (US$2,500), marketing and advertising
irc
expenses (US$1,500), and depreciation (US$1,500).
The current business climate for home maintenance services is very strong in Baton
C
Rouge and CSN Inc., is seeing ample demand for each one of its products. However, there
is also a severe limitation of qualified workers in the area. CSN Inc., has adhered to a
motto: ‘Teach Your Children Well,’ ever since their younger days. So they have employed
d
their own children, five excellent high-school graduates, to run the operations.
ite
These five employees are each paid a competitive salary of US$2,000 per month
including benefits. In return, they are each expected to work 200 hours a month
im
resulting in a total of 1,000 hours of available capacity. CSN Inc., has thus fixed its
labour rate as US$2,000/200 = $10 per labour hour. Table 4.2 presents the average
time per job and some revenue/cost data for the services offered by CSN Inc., based
rL
on data gathered over the past six months.

Fo
Table 4.2: Data for CSN Inc.
Plumbing Window Installing Landscaping

Cleaning Gutter Guards
Ave. Time Required/Job Two hours Four hours Three hours Five hours
Ave. Charge per job US$130.00 US$170.00 US$200.00 US$250.00
Cost of Material US$30.00 US$10.00 US$70.00 US$75.00
Currently, CSN Inc., is completing an average of 90 plumbing, 70 window

cleaning, 80 gutter guards, and 60 landscaping jobs a month. At this output level, we
note that CSN Inc., uses (90 ´ 2 + 70 ´ 4 + 80 ´ 3 + 60 ´ 5) = 1,000 hours, namely,
all its available capacity.
Based on this data, CSN Inc., assigns labour costs to the four products as follows.
y
Since it takes two hours for plumbing, the labour cost for a plumbing job is
nl
US$10 ´ 2 = US$20 per job. Similarly, the labour costs for the other three products
are: Window cleaning: US$40, gutter guards: US$30, and landscaping: US$50.
O
n
STANDARD COSTING
io
Standard Costing first identifies a ‘cost driver’ to spread fixed costs across the
at
products. For our example, suppose the cost driver identified is production volume.
Currently CSN Inc., is producing (90 + 70 + 80 + 60) = 300 jobs a month. Hence,
ul
administrative overhead charged to each job is US$18,000/300 = US$60.00.
Similarly, the non-administrative overhead allocation per job is US$9,000/300 =
irc
US$30.00. Table 4.3 presents the allocation of the non-administrative and
administrative overhead in two separate rows. The profit for each product, as
C
obtained by Standard Costing, is also presented in the table.

d
Table 4.3: Profit Computation for CSN Inc. Using Standard Costing
ite

im

rL
Labour Cost US$20.00 US$40.00 US$30.00 US$50.00

Admin. O/H Allocation US$60.00 US$60.00 US$60.00 US$60.00
Non-Admin. O/H
Fo
Allocation US$30.00 US$30.00 US$30.00 US$30.00

Profit US-$10.00 US$30.00 US$10.00 US$35.00
Standard Costing usually has to reconcile all the overhead and labour costs at the
end of an accounting period based on actual usages. This is not currently an issue for
CSN Inc., since it is using all its labour capacity, and is absorbing all overheads. So
there is no labour variance or overhead absorption variance to worry about.
At the current levels of output, the profit realised by CSN Inc., is 90 ´ (–US$10)
+ 70 ´ US$30 + 80 ´ US$10 + 60 ´ US$35 = US$4,100 per month
The profit data reveals that landscaping is the most lucrative operation. Indeed,
this is a product that CSN Inc., introduced into its portfolio only recently and it is a
y
product that they wish to promote to the extent possible. The Customer Service
Representative, Steele, is happy with the analysis. He is avidly promoting this
nl
product among CSN Inc.’s clients, singing praises on their capability to deliver on
O
this product.
The General Manager, Cromby, is not quite sure whether Steele’s pitch is off-key
or not. He is worried about the seemingly arbitrary way in which the overhead is
n
allocated. He has heard about a technique called Activity-Based Costing (ABC)
io
which uses different ‘cost drivers’ for different categories of overhead.
at
ACTIVITY-BASED COSTING (ABC)
ul
irc
ABC is a technique that assigns costs to products based on the activities they require.
ABC focuses on the activities of a production cycle. It is based on the premise that
C
outputs (products or services) require activities to produce them, activities consume

resources, and resources consume costs. In the context of our example, the four
products require activity in the form of administration, marketing, transportation,
d
etc. In turn, these activities consume resources in the form of administrative time,
ite
fleet usage, buildings, etc. The cost of the resources used for these activities has to be
accounted for accordingly. Thus, ABC recognises the causal relationship of cost
im
drivers to activities.
For our example, suppose that Cromby is not happy with the current choice for a
cost driver to allocate either category of overhead. He wants to allocate administrative
rL
overhead based on the actual time and effort spent by the three managers on the
various products. He also wants to allocate the non-administrative overhead using
Fo
labour hours as the cost driver, instead of production volume. So he gathers

information (expending some effort and cost) on the actual time Cromby, Steele and
Nash spend on the various products. The data reveals the following breakdown on the
time administration spends on the four products: plumbing: 30 per cent; window
cleaning: 35 per cent; gutter guards: 20 per cent; and landscaping: 15 per cent. Based
on this data, Cromby calculates the total administrative overhead dollars to be spent
on plumbing as US$18,000 ´ 0.30 = US$5,400. Since there are 90 plumbing jobs

completed each month, the administrative overhead allocated to a plumbing job is
US$5,400/90 = US$60. In a like manner, the administrative overhead allocation to
each of the other jobs is obtained as shown in Table 4.4 below.
y
nl
Table 4.4: Administrative Overhead Allocation for CSN Inc. Using ABC
O
Percentage Effort 30 per cent 35 per cent 20 per cent 15 per cent
n
Admin O/H Allocated = US$5,400 US$6,300 US$3,500 US$2,700
io
percentage effort
´ US$18,000
at
Number of Jobs 90 jobs 70 jobs 80 jobs 60 jobs
Administrative O/H US$60 US$90 US$45 US$45
Allocation per job
ul
To allocate the non-administrative overhead, CSN Inc., uses labour hours as the
irc
cost driver. Since the total non-administrative overhead is US$9,000, and since the
total available labour capacity is 1,000 hours, the non-administrative overhead is
C
charged at the rate of US$9,000/1,000 = US$9 per labour hour. Since plumbing
takes two hours, the Non-Admin. Overhead allocated to a plumbing job is =
US$9 ´ 2 = US$18. Thus the non-administrative overhead allocation per job is:
d
plumbing (two hours): US$18; window cleaning (four hours): US$36; g. guards
ite
(three hours): US$27; landscaping (five hours): US$45. Table 4.5 presents the
complete cost and profit data using ABC.
im
Table 4.5: Profit Computation for CSN Inc. Using ABC

rL

Fo

Labour Cost US$20.00 US$40.00 US$30.00 US$50.00
Admin O/H Allocation US$60.00 US$90.00 US$45.00 US$45.00
Non-Admin. O/H Allocn. US$18.00 US$36.00 US$27.00 US$45.00
Profit US$2.00 US-$6.00 US$28.00 US$35.00
With ABC, the profit data shows that while landscaping is still the most profitable
operation, plumbing is now no longer a losing proposition, whereas window cleaning
is.
Since ABC uses more meaningful cost drivers to drive the allocation of costs, it is
y
widely accepted as a better method than Standard Costing for measuring
performance. However, the improved accuracy of tracking comes at a cost. ABC
nl
requires more data to be gathered to monitor where costs are being expended, and a
O
number of organisations find this task to be onerous.
Continuing with our example, suppose that the customer service representative,
Steele, has discovered that there is ample demand for services offered by CSN Inc. He
n
has estimated the following monthly demand for each type of service in the county:
io
plumbing: 250 jobs per month; window cleaning: 160 jobs per month
at
install gutter guards: 145 jobs per month; landscaping: 120 jobs per month
Suppose, too, that CSN Inc., is in a position to choose among the various products
ul
it offers, and pick the more profitable offerings. Based on the data obtained using
irc
ABC, the company would like to do as many landscaping jobs as possible first, since
that would yield the highest profit per job. The next best offering would be gutter
guards, and so on. Completing 120 landscaping jobs will use up 120 ´ 5 = 600 hours
C
of capacity. The remaining 400 hours can be used for Gutter Guards, to complete
400/3 = 133 jobs, leaving one hour of unused capacity.
d
The resulting profit from this product mix would appear to be 120 ´ US$35 + 133
ite
´ US$28 = US$7,924. However, this is not the true profit because there are
unabsorbed overheads and unabsorbed labour, resulting in overhead and labour
variances. These variances are reconciled as follows. The 120 landscaping and 133
im
gutter guard jobs will each recover US$45 of administrative overhead, that is: US$45
´ 120 + US$45 ´ 133 = US$11,385. So, the administrative overhead Variance =
rL
US$18,000 – US$11,385 = US$6,615. The one hour of unused labour gives a labour
usage variance of US$10 and a non-administrative overhead variance of US$9. So,
the total of all the Variances is: US$6,634. Subtracting this amount from the
Fo
apparent profit of US$7,924, we end up with a net profit of US$1,290. Thus, the
‘optimal’ profit is actually much less than the profit we obtained earlier (of
US$4,100) using an apparently arbitrary product mix! Why do we obtain such a
surprising anomaly?
The answer is that when we ‘optimise’ the product portfolio using ABC (or any
other technique to allocate fixed costs), we use historical data to drive future decision
making. However, in the process of doing so, if we end up with a different product
mix, the allocated costs must be revised (or variances must be reconciled) to correctly
account for the revised product mix. That can now provide a different cost/profit
picture driving towards a new product portfolio, and so on, ad nauseam.
y
The above analysis thus allows us to make the following observation: ‘ABC is good
for tracking where costs were incurred, not for product mix decisions.’ Indeed, that is
nl
where the strength of ABC lies: in tracking where costs were incurred, so that these
O
costs can be better controlled in future.
n
4.6 COMPETENCES AND PERFORMANCE
io
Prahalad and Hamel, 1990, emphasised that an organisation should focus on devel-
oping core competences that help to create enduring customer satisfaction and loy-
at
alty. Core competences are embodiment of organisational knowledge built up over a
time and not easily imitated, that yield competitive advantage to the firm. A firm’s
ul
core competences reside in the accumulated intellectual capital of the firm and in-
clude its technologies, skills, experience, and management process. The resource
irc
based view which is gaining increasing acceptance in strategic management literature
is based on the idea that a firm performs well over time because it develops a core
C
competence which allows it to outperform its competitors. It is also argued in the litera-
ture that a firm should not be viewed as portfolio of assets (internal competences) but as
a set of mechanisms by which customer pleasing capabilities are selected and built.
d
Therefore development of competence for supply chain management requires the

ite
identification, selection and use of strategic Supply Chain Management (SCM) capa-
bilities that can be directly related to the business objectives of the firm.
im
In the last two decades, with the adoption of supply chain management
philosophy and increased outsourcing, the purchasing function got transformed
from a routine activity to a strategic one. Firms started to develop competence in
rL
purchasing to gain competitive advantage in the supply chain. Currently, managers

view purchasing, operations and marketing functions as intimately linked parts of the
Fo
supply chain, each with the ability to contribute strategically to the firm. Past
literature also suggests that supply chain competence can be conceived as consisting
of three distinct competences: purchasing competence, production competence and
marketing/logistics competence.
➣ In the fast changing supply chain environment, managers are beginning to
realise that innovation and information technology (IT) are the prime
movers of improved productivity and drivers of global competition.

Examination of the supply chain success stories reported from India reveals
strong support to this realisation. The competences in IT and innovation are
essential for firms to integrate internally and also externally with its suppliers
and customers.
y
nl
O
CONCLUSIONS
Successful management of supply chain requires cross-functional co-ordination.
n
SCM performance metrics, in this context, must be clear and accountability must be
io
unambiguous. Well defined performance metrics, valid across the supply chain, is
part of the seamless information flow to decision makers that is considered critical for
at
supply chain success. The vision of supply chain performance metrics is no longer a
distant dream. The internet can be a key enabler for monitoring supply chain
ul
performance since it facilitates the sharing of information in a collaborative and
irc
timely manner. One of the obstacles to developing supply chain metrics is that the
members in the supply chain have to set aside concerns about sharing what is deemed
to be confidential information. Many organisations are reluctant to even share their
C
key performance indicators, much less divulge the values of these indicators. One way
to work around such local thinking is to educate the supply chain partners and let
d
them recognise that their performance is measured ultimately by the end customer.
That is easier said than done; unless the partners can see tangible benefits, they may
ite
not be willing to give up their local thinking. A more practical way to address this
problem is to develop trust so that those sharing information do not have to be
im
concerned about the data being used against them. Such a trust is nourished by
working with few suppliers on long-term contracts, and partnering with one or at
rL
most a select few logistics providers. This topic is addressed in more detail in
subsequent chapters.
Fo
Goldratt, E. M., (1989), The Haystack Syndrome: Sifting Information out of the Data Ocean,
North River Press.
Hausman, W. H., (2002), Supply Chain Performance Metrics, in C. Billington,

T. Harsison, H. Lee and J. Neale (eds.), The Practice of Supply Chain Management.
Kluwer Publishers.
Kaplan, R. S. and D. P. Norton, (1992), The Balanced Scorecard: Measures that Drive
Performance, Harvard Business Review, Jan–Feb, 71–80.
y
Lambert, D. M. and T. L. Pohlen, (2001), Supply Chain Metrics, The International Journal
nl
of Logistics Management, 12, 1, 1–19.
O
Prahalad, C.K. and G. Hamel, (1990), The Core Competence of the Corporation, Harvard
Business Review, May–June, 71–91.
Reeve, J., (2002), Integrity Motors and Southern Railroad, Center for Executive Education,
n
Knoxville, TN: The University of Tennessee Press.
io
Case references : Performance measures are almost essential for any case analysis and have
to be considered carefully in any given context.
at
ul
irc
1. For the Integrity Motors case, indicate what some of the metrics would be for the
C
manufacturer, shipper, and dealer, to help them realise their BTO strategy.
2. Complete the tables for the current set-up and for the BTO strategy for the Integrity
Motors case.
d
3. Discuss the four perspectives of the balanced scorecard in a supply chain context,
ite
providing specific examples of metrics that can be applied. Be sure to adopt a systems
perspective when presenting these metrics.
im
4. For CSN Inc., is there an optimal product portfolio that maximises profits, given the
constraint of 1,000 hours of labour capacity? If so, determine that profit.
rL
Fo
y
CHAPTER
nl
5
SCM Effectiveness and Lean
Thinking
O
n
CHAPTER OUTLINE
io
Introduction
at
5.1 The Toyota Production System
5.2 Creating Flow: The Tools and Techniques of Lean Thinking
ul
5.2.1 5-S
5.2.2 Flow Charts
irc
5.2.3 Takt Time
5.2.4 Average Labour Content and Minimum Manning
5.2.5 Mixed Model Scheduling and Small Batch Production
C
5.2.6 One-Piece Flow

5.2.7 Cellular Layout
d
5.2.8 Standard Work

5.2.9 Pull Replenishment and Kanbans
ite
5.2.10 Point of Use Material Storage

5.2.11 Mistake Proofing and Method Sheets
im
5.2.12 Total Productive Maintenance

5.3 Continuous Improvement and the Pursuit of Perfection
Conclusions
rL
References
Exercises
Fo
SCM Effectiveness and Lean Thinking 121
INTRODUCTION*
I
n 1980, Japan became the million vehicles—three million more
world’s leading producer of au than what their U.S. counterparts
manufactured that same year—and re-
y
tomobiles with production of a
little over 11 million units out of a main the world’s leading automobile
nl
worldwide total of over 38.6 million manufacturer for the next 15 years?
units. That gave the Japanese Japan’s rise to this pre-eminent po-
O
automakers 28.5 per cent of the world sition was fuelled by its application of
market. Slipping to second place for lean thinking principles and concepts.
the first time since taking the lead from It is a widely accepted notion that
France in 1904, the United States pro- these principles and concepts origi-
n
duced just over 8 million automobiles nated from Japan. What is not as
io
for a market share of about 21 per widely known is the fact that a number
cent. In 1955, the Japanese auto of these principles and concepts were
manufacturers had produced about originally developed in the United
at
69,000 vehicles, a year in which the States. In fact, Henry Ford inspired a
U.S. auto industry built 9.2 million ve- number of elements in the Toyota Pro-
hicles. Who could have predicted that
just 25 years later the Japanese auto
ul
duction System such as kaizen, flow,
pull production and the focus on elimi-
industry would produce more than 11 nating waste.
irc
C
d
ite
im
rL
Fo
* A number of sections in this chapter are drawn from the book, Streamlined, by Mandyam
M. Srinivasan. The authors thank Cengage (formerly Thompson-Taxere) for giving us
permission to use this material.
5.1 THE TOYOTA PRODUCTION SYSTEM

The Japanese automobile manufacturers succeeded in producing vehicles at low cost
by adapting Henry Ford’s ideas into their just-in-time (JIT) production management
philosophy and system. The JIT management system had its roots in the Toyota
y
Motor Corporation, and owes its conception and subsequent development to two
nl
individuals, Kiichiro Toyoda and Taiichi Ohno, although the latter is acknowledged
widely as the creator of the Toyota Production System (TPS).
O
Inspired by Henry Ford’s book, Today and Tomorrow, Kiichiro Toyoda had
formulated, as early as 1936, a clear mental picture of the production system he
n
wanted. The basic idea was to only produce what was needed on a given day, namely,
io
to initiate a production run when it was needed, rather than making a production run
in anticipation of a demand. He initiated this idea in the automobile department he
at
had started within his father’s organisation, the Toyoda Automatic Loom Works.
Slips were passed around indicating the number of parts to be made or processed that
ul
day. This was the origin of the kanban method of production and it provided the
basis for the JIT system. He began convincing suppliers to cooperate with his JIT
irc
system. He also changed the traditional physical layout of the plant so that machine
tools were organised in a flow line. That made the supply line shorter so parts could
get into the assembly process sooner.
C
Taiichi Ohno, who continued and improved on the processes put in place by
Toyoda, credits his contributions to TPS to two main concepts [Ohno, 1988]. The
d
first concept, from Henry Ford’s book Today and Tomorrow, was the moving assembly
ite
line that provided the basis for the production and assembly system used in TPS. The
second concept was the supermarket operations he observed during a visit to the
im
United States in 1956, which provided stores with a continuous supply of

merchandise. The supermarket gave Ohno the idea to set up a pull system, in which
each production process became a supermarket for the succeeding process. Each
rL
process would produce to replenish only the items that the downstream process
selected.
Fo
The Toyota Production System emphasizes working with kanbans and minimal
inventories, but merely trying to reduce inventories without considering other factors
can have serious consequences. The system is now much more vulnerable to
disruptions. Figure 5.1 uses a ‘river and rocks’ analogy where the water level is
analogous to the inventory level in a facility. A higher water level hides potential
blemishes in the process such as unreliable suppliers, scrap loss, machine breakdowns,
and so on. As the water (inventory) level is lowered, these problems surface, and
forces management to work on correcting these defects. The key is to resist the
temptation to reduce the water level (inventory) too quickly. Rather, the idea is to
lower the water level a little, break apart the newly exposed rocks (obstacles), and
then lower the water level once again.
y
nl
Figure 5.1: Inventory Hides Defects
O
n
io
at
ul
irc
C
d
ite
im
Before attempts are made to reduce inventory levels, some of the major elements
rL
that should be in place are reliable processes, preventive maintenance systems, cross-
trained workers, setup reduction, and reliable suppliers. If these elements are not
already present in a factory, then putting them in place takes time; certainly it cannot
Fo
happen overnight. The Japanese spent perhaps close to 25 years perfecting the system
before it was successfully applied on a wide scale.
Figure 5.1 also carries another important message. For the boat to move faster, all
oarsmen should row at the same time. The most effective method of production
within the organisation is to have all processes work at the same rate. There is no
point in having some of the resources work faster than the others, because that will
invariably pile up inventory in front of the slower processes. If all resources respond
to pull signals, that will ensure a smooth flow of products across the organisation. The
signal for a resource to produce could be generated simply when the downstream
resource draws a unit of WIP from the buffer placed between the two resources.
y
The Japanese recognised that for the entire process to flow smoothly, it was
necessary for suppliers to participate in this flow process, responding to the pull
nl
signals. The Japanese management system was thus built on partnering arrangements
O
that had mutual benefits for both the automakers and their suppliers. For instance,
suppliers were provided ample visibility on the automakers’ assembly schedules, and
these schedules were generally adhered to quite stringently. That made it easier for the
n
suppliers to plan their own production schedules accordingly and deliver supplies in
io
a timely manner without having to hold a lot of finished goods inventory.
at
5.2 CREATING FLOW: THE TOOLS AND TECHNIQUES OF LEAN
THINKING ul
The preceding discussion clearly demonstrates the value of creating flow within and
irc
across organisations. Within the organisation, the important tools and techniques
(‘lean tools’) used to promote flow are:
C
➣ 5S
➣ Flow Charts
d
➣ Takt Time
ite
➣ Average Labour Content and Minimum Manning

➣ Mixed Model Scheduling and Small Batch Production
im
➣ One-Piece Flow
➣ Cellular Layout
rL
➣ Standard Work
➣ Pull Replenishment
Fo
➣ Point of Use Material Storage

➣ Mistake Proofing and Method Sheets
➣ Total Productive Maintenance
➣ Continuous Improvement and the Pursuit of Perfection
Although these lean tools were originally developed for manufacturing
organisations, many service organisations are finding them very useful as well. As we
discuss these tools, think about how they apply in a service setting. For instance, the
final product might be a piece of paper that reaches its destination in a more timely
fashion. How often has one of your operators sat idle waiting for the next order to
present itself? How often have you delayed a service delivery waiting for the
paperwork to be completed?
y
Before we apply these tools, we must address the scope of the lean implementation.
nl
Organisations often create lean cells to put out products at a rapid rate, only to find
O
these products queueing up at a downstream work centre. The questions that must
first be addressed are as follows: should the organisation implement lean on all its
processes and activities or should it focus on a subset of its processes? Similarly, should
n
it implement lean on all its products or on a subset of them? It is recommended that
io
the organisation choose one product family at a time and implement lean on all the
processes and activities that apply to this product family, before moving on to the
at
next product family.
5.2.1 5-S
ul
irc
The term 5-S is used to denote a systematic process for organising the workplace
based on five simple, yet powerful, activities, each of them represented by a Japanese
C
word that begins with the letter ‘S.’ The five Japanese words, with their English
translations, are Seiri (tidiness), Seiton (organisation), Seiso (cleanliness), Seiketsu
(neatness), and Shitsuke (discipline). Figure 5.2 presents the 5 S’s and their definition.
d
Corresponding to these five Japanese words, five English words that begin with the
ite
letter S and convey nearly the same intended meaning are also presented in Figure
5.2. Some organisations add a sixth S for Safety. Some of the intended meanings of
im
the original Japanese words are, however, lost in the translation, especially when
trying to capture the meaning with English words that begin the letter S.
The first step, Seiri, refers to taking out unnecessary items and discarding them.
rL
Literally, it means organising something that is disorganised. Every item in the work
area is classified as either a necessary, unnecessary, or a ‘red tag’ item. Red tagged
Fo
items literally have a red tag tied to them, indicating that it is unclear whether or not
the item is needed. Items marked as unnecessary are disposed of immediately,
following which an ‘auction’ is held wherein red tag items are displayed for anyone to
claim them as necessary. Red tag items that no one claims are discarded.
The idea behind the second step, Seiton, is to arrange the necessary items in a
proper order so that they can easily be picked up for use. In Japan, the words Seiri and
Figure 5.2: The 5 S’s and Their Definition
y
nl
O
n
io
at
ul
irc
Seiton are often used in combination as Seiri-Seiton because there is not a big
difference in their meanings. In any case, the second step is intended to create storage
C
systems and provide visual information about what is stored and how much should
be stored in a given spot. Standard, convenient locations are created for tools and
d
devices used in the work area. For example, tools are typically hung on boards, with a
ite
‘shadow box’ (a silhouette) of the tool painted on the board where the tool is to be
hung. Drawers are sometimes filled with styrofoam, with cutouts of the items that are
stored in the drawers.
im
The third step, Seiso, prescribes keeping the area clean, and to take pride in a
workplace that is organised and kept in good condition. This step goes beyond
rL
simply making the area more pleasant to work in by sweeping the floor and cleaning
up leaks and spills. It includes checks for malfunctioning machinery, loose parts on
machines, etc. Aside from making the area more conducive to work in, the Seiso step
Fo
provides other practical benefits. For example, if machines are kept clean, oil leaks
will be discovered before a catastrophic equipment failure; if the aisles are kept clean
and free of any oil spills, the chances of accidents are minimised, and so on.
The fourth step, Seiketsu, provides the basis for standardisation. Literally, Seiketsu
refers to a condition where there is no smear, stain, etc. This step covers both personal
and environmental cleanliness, and defines the standards by which personnel must
measure and maintain ‘cleanliness.’ This step prescribes what the normal condition
should be, as well as how an abnormal condition should be corrected. Visual
management is an important ingredient of Seiketsu. Colour-coding and
standardisation of colours are used for easier visual identification of problems.
Personnel are trained to detect such problems using one or more of their five senses
y
and to correct them immediately.
nl
The final step, Shitsuke, relates to building discipline that will sustain the first four
O
steps. Literally, Shitsuke refers to making a person keep a rule or order through
training. The rationale is that it is often easier to clean up an area than it is to keep it
clean. Thus an integral part of a 5-S programme should be a system for maintaining
n
the first 4 S’s. The Shitsuke step commits to maintain orderliness and to practice the
io
first 4 S’s continually, and it is vital that this step has full support from top
management. Initially, it is very likely that top management has to provide the right
at
incentives for this step to take place on a regular basis. Once the inertia is overcome,
however, there is a lasting effect, and the process becomes self-sustaining.
ul
The 5-S programme improves safety, work efficiency, productivity and establishes
a sense of ownership. These activities ensure a clean and orderly working
irc
environment and help employees become aware of their working environment and
the condition of the tools and machinery they use. In many organisations, the
C
benefits of a 5S programme are often so dramatic that there is a real danger that the
organisation may step away from a full blown lean implementation at this stage,
d
thinking that its mission is complete. It is, therefore important to note that the 5-S
programme is just the starting point in the lean journey. Equally important to note is
ite
that the 5-S programme can actually hinder the lean journey if the vital fifth step
(Shitsuke) is not practiced. A good 5-S implementation builds the foundation for
im
continuous improvement.
rL
5.2.2 Flow Charts

Anyone can visualise a process step. However, in a work environment it is often easy
Fo
to miss the simple flow of a process from one step to the next. The flow chart is a
powerful visual tool that can describe practically any process, be it a manufacturing
process or a service process. Flow charts enable us to quickly identify the process steps
we must eliminate in our drive for simplicity and waste reduction. Identifying and
eliminating (or at least reducing) non-value-added activities is key to streamlining a
process.
There are several alternate methods for capturing value-added and non-value-
added activities using flow charts. One way is to use a process flow chart that captures
the logical sequence of activities involved in delivering the product, in conjunction
with a spaghetti diagram that depicts the physical movement of products through the
plant or office. A process flow chart typically uses standard symbols, as presented in
y
Figure 5.3 below, to identify different activities. The value-added activities are
nl
colour-coded green and the non-value-added activities are typically colour-coded
yellow.
O
n
Figure 5.3: Process Flow Charting Symbols
io
at
ul
irc
C
d
ite
im
Deciding what is a value-added activity and what is non-value-added can

rL
sometimes be a contentious issue. Does the marketing function add value? And does
the logistics function add value? If so, which logistics activities are value-added? One
Fo
way to resolve this issue is to define a value-adding activity as an activity that either
actually transforms the product or one that the customer would be willing to pay for
it to happen.
Figure 5.4 presents a process flow chart for a mortgage loan application. In this
figure, the value-added ratio is obtained as the ratio of the actual value-added work
expressed in time units divided by the elapsed time for the process to complete. The
summary data in the figure presents the value-added ratio as 1.8 per cent; a
remarkably low number, that is, unfortunately, all too typical of the value-added ratio
for a majority of processes in the real world. It is also indicative of the amount of
waste that is prevalent in the system, and the huge opportunities there are for lean
efforts to remove waste. While it is difficult to specify what the value-added ratio
y
should be because it depends on the industry, a ratio of 10 per cent has sometimes
nl
been specified as a goal—see, for instance [G. Conner, 2001].
O
Figure 5.4: A Process Flow Chart for a Mortgage Loan Application
n
io
at
ul
irc
C
d
ite
im
rL
Fo
The formula described above, however, does not consider the actual effort
expended in delivering the product. For instance, it is quite feasible that some of the
steps, value-added or otherwise, may involve multiple employees working on the
activities. In such situations, a more meaningful value-added ratio would have the
total man-hours of value-added effort required to produce one unit in the numerator
and the sum total of all the man-hours employed by the organisation to produce one
unit in the denominator.
A diagram that is sometimes used to represent the distance travelled by the product
y
is the spaghetti diagram. Figure 5.5 presents a spaghetti diagram for the same
mortgage loan application process. The spaghetti diagram, used in conjunction with
nl
the process flow chart, can provide the analyst with valuable information in the effort
O
to streamline the process. While the process flow chart will reveal the non-value-
added activities, the spaghetti diagram will highlight the extent of travel, including
any back-tracking that the product may undergo. It may suggest how the process
n
could be streamlined through a better process layout.
io
at
Figure 5.5: The Spaghetti Diagram for a Mortgage Loan Application
ul
irc
C
d
ite
im
rL
Fo
The flow chart should be preferably drawn after the 5S programme is in place. If
the flow chart is drawn before the 5S programme is initiated, the lack of attention to
simple housekeeping may obfuscate some of the non-value-added activities that the
organisation really needs to focus on. For instance, the operators could be taking
more time to change tooling simply because they are difficult to locate and that may
be reflected in the time standards developed for the tool changeover activity.
y
nl
5.2.3 Takt Time
If someone compiled a dictionary of terms used in lean, they would notice a slew of
O
Japanese words in the compilation. For instance, we have already discussed the 5 S’s,
kanbans, and kaizen. Takt, however, is a German word for a musical meter, that is
n
used in lean, even in Japan. When German aerospace engineers helped Japan build
io
aircrafts during the 1930s, they used the word takt to present an analogy of a
conductor waving his baton to set the rhythm for the entire orchestra. After World
at
War II, Toyota adopted this word and the accompanying concept as the basis for
linking its production capacity to customer demand in the Toyota Production
ul
System. The customer demand became the cadence that dictated the pace of
operations on the shop floor.
irc
How do organisations apply takt time? First, it should be recognised that takt time
represents the customer demand. Takt time is calculated as follows:
C
Takt time = Available time per period/Demand per period.

The duration of the ‘period’ could be minutes, hours, days, or weeks (so long as the
d
same unit of time measurement is used in the numerator and the denominator). To
ite
illustrate the concept, we start with a simple example of an organisation working on

a single product.
im
We consider an insurance agency, Hungama Inc., that processes applications for

property insurance. Processing an application involves four steps: (a) Data gathering/
data entry (‘Distribution’), (b) Risk analysis (‘Underwriting’), (c) Computing the
rL
premium (‘Rating’), and (d) Policy Writing. Suppose that the customer demand for
this product is 20 policies per day, and suppose that Hungama operates a single shift
of eight hours duration.
Fo
The first step in calculating takt time is to determine the available time per shift.
Typically, the essential breaks such as lunch and bio breaks are deducted from the
duration of the shift to arrive at an available time of, say, six hours and forty minutes
(400 minutes) each day. Then the takt time is 400/20 = 20 minutes per unit. In other
words, the organisation has to process one policy every 20 minutes. (If the
organisation operated two shifts each day, then the available time per day is 800
minutes, and the takt time will be 800/20 = 40 minutes per unit.)
Continuing with this example, since the (external) customer demand is for one
unit every 20 minutes, the organisation should match its (internal) resources to meet
y
this demand. Suppose the average time it takes to process the four steps (the ‘cycle
times’) are 14 minutes, 22 minutes, 6 minutes and 12 minutes, respectively. Suppose,
nl
too, that there is an individual dedicated to each of these four steps. Figure 5.6 shows
O
a ‘Load’ chart depicting how these individuals are loaded. The figure has two parts,
the first part corresponding to the current loading, labelled ‘Before Load Levelling.’
n
io
Figure 5.6: Load Chart for Hungama Inc.
at
ul
irc
C
d
ite
im
rL
One of the main uses of the takt time calculation is to match internal resources to
external demand, as alluded to earlier. When we examine how the operators are
Fo
loaded in Hungama, it is apparent that the operator in charge of underwriting is

overloaded because he is loaded beyond the takt time. In order to meet the takt time,
this operator is presumably either working overtime, or working at a faster rate than
called for, in which case there are opportunities for the operator to make errors. The
load chart thus helps determine whether the work is assigned equitably and whether
anyone is overloaded. At the same time, it also helps determine whether there is slack
in the system, which is apparent in Figure 5.6 (a), since operators 1, 3 and 4 are
underutilised.
What are the options available for Hungama at this stage? If some of the
Underwriting work is reallocated to another operator, operator 2 could now work
under takt time. Also, the workload for operators 3 and 4 could be combined and
y
assigned to one operator. The second part of Figure 5.6 shows the ‘After Load
nl
Leveling’ scenario, where a task that accounts for five minutes of work is moved from
operator 2 to operator 1. The loads on the three operators are now 19 minutes, 17
O
minutes, and 18 minutes, respectively.
The preceding discussion assumes that the time to perform each activity is
n
accurate. That is an issue that must be resolved quite early in this decision process,
io
and may require some time studies to be conducted. Once this issue is resolved, the
next issue is to decide how the individual operators are loaded. If the operators are
at
underutilised, there is a likelihood that the work expands to fill the time available,
resulting in inefficiencies. On the other hand, loading each operator close to 100 per
ul
cent may result in lots of opportunities for error, particularly if the workload is highly
variable with prolonged periods where the operator is loaded beyond capacity.
irc
Ideally, we want to buffer variation with capacity, not inventory. Loading operators
close to 100 per cent leaves no room for accommodating increased customer demand
C
without reallocating resources. Note that the takt time goes down when the demand
increases. If the demand at Hungama increases by 10 per cent to 22 policies per day,
then the takt time goes down to 18.18 minutes per job and operator 1 becomes
d
overloaded. That said, in some situations it is advisable to keep operators loaded close
ite
to 100 per cent because that may motivate them to find creative ways of managing
the workload so that the resulting cycle time is well within the allotted takt time.
im
The above example used task times that were discrete. If it was possible to sub-
divide tasks even more finely, it would be possible to load each operator as close to
takt time as desired. If such a fine division of tasks were possible, what would be the
rL
best way to load each operator? In the above example, the total workload is
(19+17+18) = 54 minutes, and so the tasks could be allocated so that each operator
had a cycle time of 18 minutes. However, as discussed above, that would uniformly
Fo
under-utilise each operator. A better alternative may be to have the first two operators
loaded with tasks that add up to the takt time of 20 minutes, leaving the third
operator loaded only up to 14 minutes. Since the third operator has a loading that is
significantly less than takt time he could help the other operators if they fall behind
schedule. This operator could also be the one located at the end of the cell, where he
can perform the material handling activity, moving parts into and out of the cell.
Working with a takt time has a number of advantages. If each operator is paced to
takt time, that automatically limits one of the primary causes of inventory,
overproduction. Limiting overproduction also stabilises the system, preventing the
frequent stops and starts that inhibit a smooth flow. When a team of operators are
asked to pace their work according to takt time, there is a heightened awareness of the
y
output rates and potential problems that detract from achieving the desired output
nl
rate. The operators obtain immediate feedback if they miss takt time on a given cycle
and make corrections accordingly. If feedback is only provided after many cycles,
O
then the window of opportunity to correct errors may have passed. At the same time,
takt time should be applied judiciously. This concept is more applicable in a flow
n
shop that processes a set of products with relatively predictable demand that does not
io
fluctuate a lot. Takt time may not be very relevant in a job shop environment, but
calculating takt time is still useful because it helps determine the number of operators
at
that must be assigned to a process.
How often should the takt time change? This question depends on the industry.
ul
Consider a manufacturing cell. If the takt time decreases, then it is not simply a
matter of reallocating work to different operators. There might be equipment related
irc
issues as well. So, if takt time is adjusted daily, that could well result in chaos. At the
same time, if the organisation is not flexible enough to react quickly, that may result
C
in missed opportunities to fill demand or, conversely, lead to inventory build-ups.

The key is to distinguish between ‘noisy’ data and trends. For example, a ‘run’ test
could be applied to determine when the takt time should be changed. Consider the
d
sequence of demands presented in Figure 5.7. The takt time (equivalently, the desired
ite
production rate) could be re-evaluated if the demand exceeds the set production rate
for, say, five consecutive days as shown in the figure.
im
Consider how takt time works for the assembly operations at Dell. Dell assembles
its desktop computers, in response to customer demand, using a number of assembly
lines that operate in parallel. Suppose that, based on current customer demand, it has
rL
set its assembly takt time at 15 seconds per computer. Dell typically allocates one or
at most two operators to assemble a computer on each assembly line. Suppose each
Fo
assembly line has two operators and that each takes two minutes (120 seconds) to do
their respective tasks. Since a takt time of 15 seconds does not provide enough time
for the assembly operators on a given line to complete all assembly operations, Dell
runs a number of assembly lines in parallel. If each assembly line puts out one
computer every 120 seconds, then Dell needs eight lines to run in parallel, in order to
meet a takt time of 15 seconds (120/8 = 15 seconds).
Figure 5.7: Re-Calibrating Takt Time
y
nl
O
n
io
at
ul
irc
Now suppose that on a given day Dell finds that it does not have enough orders on
hand to warrant a takt time of 15 seconds per computer, but instead the takt time
C
works out to 16 seconds per computer. (Note that the takt time goes up when the
demand goes down.) Dell has several choices at this stage. One alternative is to shut
down one assembly line. But with 7 lines, each line assembling one computer every
d
120 seconds, it will take 17.14 seconds to assemble a computer (120/7 = 17.14
ite
seconds), and that would not meet customer demand. Another alternative would be
to ask each assembly line to work slower. A third alternative would be to run all eight
im
lines at full speed, continuing to produce one computer every 15 seconds. This
alternative causes minimal disruptions in the assembly process but the
overproduction will result in some computers that are built to stock. These
rL
computers can either be sold at a discounted price, as Dell usually does, or it could
serve as a buffer for a possible demand increase the following day.
Fo
To give another example, Toyota examines its production rate at 10 day intervals
and adjusts the takt time if need be. Most of the time, Toyota may not need to change
the production rate or if it does may decide to do so by changing the length of the
workday (by working overtime), because changing the line rate on an automobile
assembly line is a major undertaking. On the other hand a small manufacturing cell
may change its takt time weekly or even daily, and adjust the number of operators
accordingly.
Takt time is one of the most misunderstood concepts in lean manufacturing. I have
been to several organisations where they prominently announce their takt time to be,
say, ‘100 pieces per hour.’ (Recall that takt time is time per piece, not pieces per time.)
It is also not uncommon for managers to state that their machines have a takt time of
five minutes. (Takt time is a measure of external demand, and has nothing to do with
y
machine capacity.) In some instances, takt time has also been confused with lead time
nl
(or flow time). Quite often this is because organisations simply go through the
motions and effect lean in a piecemeal manner without fully understanding the
O
implications. Remember that the primary purpose of takt time is to match external
demand with internal capacity. To sum up our discussion on takt time so far:
n
➣ Takt time is used to represent the customer demand. It is expressed in terms of
io
the time available to make one unit that will keep pace with customer demand
➣ The primary purpose of calculating takt time is to match external demand
at
with internal capacity
ul
➣ Takt time should be adjusted when it is clearly evident that the customer
demand has changed. Adjusting takt time too frequently to respond to small
irc
fluctuations could result in chaos. The key is to distinguish between ‘noise’
and a trend.
C
So far, the discussion on takt time has been based on a single product. The takt
time calculation does not change when there are multiple products involved, but the
calculations involving operator loading become just a little more involved. We
d
demonstrate how the calculations proceed by expanding on the Hungama, Inc.

ite
example.
Hungama Inc. Revisited

im
Hungama Inc. processes four different kinds of property insurance applications:

RUNS (requests for underwriting), RAPS (requests for price quotes), RAINS
rL
(requests for additional insurance on an existing policy) and RERUNS (annual

renewal of existing policies). The daily demand for each type of application is
provided in Table 5.1 below.
Fo
Table 5.1: Demand Data for Hungama Inc.
RUNS RAPS RAINS RERUNS
Jobs/Day 4 12 6 18
per cent of Total 10 per cent 30 per cent 15 per cent 45 per cent
Each product type requires some or all of the four steps indicated earlier, namely,
distribution, underwriting, rating, and policy writing. The workload varies by
product type. Table 5.2 presents the average labour content for each product type.
y
Table 5.2: Operation Times for Hungama Inc.
nl
RUNS RAPS RAINS RERUNS
O
Distribution 58 min. 50 min. 44 min. 28 min.
Underwriting 43 min. 40 min. 23 min. 19 min.
n
Rating 72 min. 65 min. 68 min. 75 min.
Policy Writing 67 min. 0 min. 55 min. 50 min.
io
Total Labour 240 min. 155 min. 190 min. 172 min.
at
The takt time for this example is simply determined by adding all the demands to
obtain a total daily demand of 40 units. Note that even though the time requirements
ul
for each product is different, this is not a matter of concern at this stage since we are
only looking at external customer demand, not internal resource requirements.
irc
Assuming that the organisation works 400 minutes per day as before, the takt time is
obtained as 400/40 = 10 minutes per unit.
C
5.2.4 Average Labour Content and Minimum Manning

d
The next step is to compute the minimum number of operators required to sustain
ite
operations. From the data given in Tables 5.1 and 5.2, the average labour content for
a job is obtained as the weighted average of the labour content for each product type.
im
This gives
Ave. Labour Content = 0.10 ´ 240 + 0.30 ´ 155 + 0.15 ´ 190 + 0.45 ´ 172
rL
= 176.4 minutes
Since the takt time is 10 minutes per unit, we need a minimum of 176.4/10 =
Fo
17.64 operators, or 18 operators to manage customer demand. Clearly, there will be

some sharing of tasks since there will be at least 18 operators to do the four different
activities. However, there are workload balancing issues that need to be resolved. As a
starting point, we could determine how many operators need to be allocated to the
distribution activity. The average labour content for this activity is 0.10 ´ 58 + 0.30 ´
50 + 0.15 ´ 44 + 0.45 ´ 28 = 40.0 minutes. Since the takt time is 10 minutes, the
minimum number of operators allocated to this task will be 40.0/10 = 4 operators.
The average labour content for the underwriting, rating, and policy writing
activities are computed similarly, and work out to be 28.3 minutes, 70.65 minutes,
and 37.45 minutes, respectively. Therefore, these three activities require a minimum
of 28.3/10 = 2.83, 70.65/10 = 7.065, and 37.45/10 = 3.745 operators, respectively.
Workload balancing and related issues may thus result in more than 18 operators
y
being needed. A load chart can now be constructed to determine how much each
nl
operator is loaded.
O
This exercise demonstrates the importance of scheduling the jobs in a judicious
manner and avoiding large batches in order to achieve scale economies. Consider, for
instance, the operators involved in policy writing. The RUNS, RAINS and RERUNS
n
demand at least 50 minutes of effort per job. However, the RAPS do not require any
io
time from the operators. If the jobs are scheduled using economies of scale, there will
be long periods during which the operators will be overworked because they are
at
working on a RUN, RAIN or a RERUN. This will be followed by a period during
which they will be idle because RAPS are being processed by the other operators in
ul
Hungama. On the other hand, if a mixed-model scheduling approach is used, it is
possible to level-load the operators and manage them much more effectively.
irc
5.2.5 Mixed Model Scheduling and Small Batch Production

C
A characteristic of an efficient production system is that products flow smoothly

through the organisation with no delays. The production schedule plays a significant
d
role in effecting this smooth flow because it dictates the frequency with which
ite
products are scheduled at the various resources. In a perfect world, the organisation’s
customer would pull a product from the final assembly station and that would
im
generate a signal on each upstream resource to produce exactly what was pulled by
the customer. However, the reality is that quite often production constraints such as
changeovers, material availability, operator availability, and so on would restrict how
rL
products flow through the organisation. In particular, the changeover time

significantly influences the extent of batching. Furthermore, plants that make a
variety of different products usually tend to produce the different products in large
Fo
batches to exploit economies of scale.

Large-lot production, of course, sends a ripple effect throughout the organisation.
In the beer game that we discussed in Chapter 2, a single change in demand at the
retailer was enough to cause huge variations in demand four stages upstream, at the
factory. Likewise, within the organisation, a small change in demand at the most
downstream operation can cause significant demand variation at upstream resources,

and batching only exacerbates this problem. Typically, these demand variations are
absorbed by carrying large amounts of WIP between each stage. Thus, in-process
buffers are one way to ‘level’ the production, and work with large batches. Needless to
say, larger WIP inventories increase the lead time. There is a better way to level the
y
production schedule. It is simply to produce every product as quickly as possible, at
nl
the same rate at which customer demands are made. This, in essence, is mixed-model
scheduling. The Japanese term for mixed-model scheduling is heijunka, which refers
O
to distributing the production of different product types evenly over the course of an
hour, day, week, or month.
n
Suppose, for instance, that an organisation makes three products, A, B, and C each
io
of which requires 10 minutes of assembly time. Assume that the final assembly shop
works 10 hours each day, five days a week. Suppose the customer is demanding these
at
products at a rate of three, two, and one, per hour respectively. The customer for
these products could be demanding them every hour, but the organisation may
ul
choose to produce these products in large batches for reasons as indicated above.
Suppose the final assembly shop decides to produce one week’s demand of each
irc
product each time. In other words, the weekly schedule would be as follows: 150 A,
100 B and 50 C. Instead of receiving products every hour, the customer will receive
C
the products once a week. In other words, either the customer or the final assembly
shop will carry an average finished goods inventory of 75 A’s (half the batch size, on
average), 50 B’s and 25 C’s. If the final assembly shop had, instead, produced
d
according to a mixed-model assembly schedule of 3 A, 2 B, 1 C, the finished goods

ite
inventory would be negligible since that exactly matches the customer’s hourly
demand rate. The six units produced in an hour could be even more finely sequenced
im
as follows: A B A C A B.
The above exercise assumed that the assembly time is the same for each product.
Mixed-model scheduling has an even greater impact on smoothing flow when
rL
different products have different processing time requirements as discussed earlier in

the context of the Hungama Inc. case. No doubt true mixed-model scheduling
Fo
requires set-up/changeover times to be minimal. As changeover times become more

significant, it can be argued that the batch sizes should correspondingly increase.
While there is some truth to this statement, we can still achieve short-cycle
scheduling by producing in small batches in the presence of set-up times.
To sum up the discussion in this section, producing products in large batches:

➣ Creates an uneven workload
➣ Creates uneven demand for upstream processes, making pull impossible
➣ Causes production to be out of sync with customer demand.
y
On the other hand, mixed model production
nl
➣ Creates a smooth work-load:
O
➣ Creates a smooth demand for upstream processes
➣ Allows production to match customer demand.
n
While mixed-model scheduling helps level the workload at each work centre, the
io
true benefit of such schedules is realised when they are used in conjunction with
another lean concept: one-piece flow.
at
5.2.6 One-Piece Flow ul
In the extreme case, mixed-model scheduling refers to the case where each successive
irc
item processed at a resource could be a different product type. In other words, the
products are processed in lots of sise one. One-piece flow refers to the concept of
C
moving products one unit at a time between workstations. This is in contrast to the
other extreme where we might process an entire batch of parts at a workstation before
d
moving the batch to the next downstream workstation. Mixed-model scheduling,

when used in combination with one-piece flow, thus results in keeping WIP
ite
inventories at the lowest possible levels.

The goal of one-piece flow is to reduce the lead time or equivalently reduce WIP
im
inventory. Little’s Law states that the average lead time is equal to WIP/throughput.
Thus, if you have a week’s worth of WIP inventory stored at an output queue, waiting
rL
to be transferred to the next station, it means you have added one week to the average
lead time at that stage.
To elaborate on how one-piece flow affects the lead time, consider the example
Fo
shown in Figure 5.8, consisting of a process with four stages of operation. Suppose
the production rate at each stage is 100 per week, and that parts are transferred 60
units at a time, as in part a) of Figure 5.8. Since the total WIP inventory in the system
is 240 units, Little’s Law tell us that the average lead time = WIP/throughput = 240/
100 = 2.4 weeks.
Figure 5.8: One-Piece Flow and its Impact on Lead Times
y
nl
O
n
io
at
ul
irc
C
On the other hand if the items are transferred one piece at a time, as shown in part
d
b) of Figure 5.8, then the total WIP inventory in the system is four units, and so the
average lead time = 4/100 = .04 weeks.
ite
One-piece flow also helps improve product quality, because it shortens the
duration of the feedback loop. When parts are transferred one at a time between
im
workstations, the downstream workstation is able to determine if the upstream

workstation is producing defective items, almost immediately. As opposed to that, if
rL
one week’s worth of WIP is transferred at a time, the entire week’s output could be
defective without anyone noticing this problem. The feedback delay is thus at least
one week.
Fo
It must be clarified that one-piece flow does not necessarily mean that just one
piece or one part is transferred between two processes each time. The unit of transfer
could well be a pallet of parts, although clearly the smaller the number of units
transferred at a time, the lower the resulting WIP will be. One-piece flow is an ideal
that the manager should aim for because it minimises the hand-off time. Products do
not have to be placed in the output queue of the upstream workstation, waiting for a
batch of products to accumulate before the transfer to a downstream workstation is

made.
While one-piece flow is an ideal that the manager should aim for, it might become
counterproductive when the cell processes a large variety of products with different
y
processing times and routings. There are a number of other situations where one-
piece flow is simply not practical. For instance, if the upstream process is a heat-
nl
treatment operation that necessarily has to produce a batch at a time, it may not be
O
very meaningful to move these parts one at a time since they would anyway come out
of the furnace all at once. Similarly, if the downstream operation requires a set up
each time it begins work on a new product, then one-piece flow may not be the right
n
approach to use, to move parts to this operation. Finally, if material transfers are done
io
by the operators themselves rather than through an automated conveyance system,
then it does not make sense for the operator to be moving the parts one piece at a time
at
if the transfer time is close to the processing time per part. In other words, if material
handling costs are high, then it may be more economical to transfer the parts in small
ul
batches rather than one piece at a time.
To implement one-piece flow, it is desirable to have little variability in the process
irc
times and the process quality at each of the steps linked in the one-piece flow. If one
step is delayed it stops the flow of the entire process. The processing steps must also
C
be located adjacent to each other, in order to facilitate moving only one piece at a
time. One-piece flow is thus facilitated by a cellular layout.
d
ite
5.2.7 Cellular Layout

One piece flow is significantly enhanced when the various processes are organised in
im
a cellular layout (alternately, a product layout). A cell consists of the operators and the
workstations required for performing the steps in a process segment, with the
workstations arranged in the processing sequence. The cellular layout is to be
rL
contrasted with a process layout where the workstations are grouped by departments
or functions. When these workstations are placed close together, then the products
Fo
moving from one workstation to the next do not have to traverse long distances, and
that facilitates one-piece flow. In addition to assisting one-piece flow, keeping
workstations close to one another allows the operator at the downstream workstation
to see what is being produced by the upstream workstation. That helps eliminate
paperwork, which may otherwise be necessary to coordinate the different
workstations if they were separated from one another by a significant distance.
A common layout used for manufacturing cells is a U-shaped configuration. This

layout has a number of advantages. First, it provides more flexibility in allocating
tasks among the operators. Consider, for example, the assembly cell shown in Figure
5.9, which has eight assembly tasks, each of duration 15 seconds. The takt time is
currently 15 seconds, and hence each assembly task requires one operator.
y
nl
Figure 5.9: A U-Shaped Cell
O
n
io
at
ul
irc
C
d
ite
im
Now, suppose the takt time increases to 30 seconds. Since each assembly task still
requires only 15 seconds, it is possible to operate the cell with four operators,
allocating two assembly tasks to each operator. A U-shaped cell allows more
rL
flexibility in the allocation of tasks. For instance, it is now possible to allocate tasks 1
and 8 to one operator who is now responsible for monitoring the parts coming in and
going out of the cell. Such an allocation of tasks would not have been possible with a
Fo
straight-line layout. This layout also allows easy replenishment of materials from
outside the cell. The U-shaped cell also promotes teamwork because the operators are
located closer to one another.
Since the U-shaped cell gives a lot more flexibility in allocating multiple tasks to
operators, there is a better utilisation of manpower. The U-shaped cell facilitates the
re-allocation of tasks among operators when an operator is added or removed in

response to a change in the takt time. Other common cell configurations include the
T-shaped cell, the L-shaped cell, or a serpentine arrangement (a series of adjoining U-
shaped cells).
y
In summary, some of the benefits associated with one-piece flow and cellular
manufacturing include:
nl
➣ WIP reduction
O
➣ Better space utilisation
➣ Lead time reduction
n
➣ Productivity improvement—more flexibility in allocating tasks to operators.
io
Production lines can be re-balanced more easily to accommodate any
absenteeism
at
➣ Quality improvement—provides immediate feedback on defects
ul
➣ Enhanced teamwork and communication—Employees can better support
co-workers who have fallen behind in their production schedule
irc
➣ Better visibility of all tasks and operations
The next step is to identify the best method for performing a particular task and
C
then develop a standard work procedure that anyone working on this task should
follow. This is the concept behind Standard Work.
d
ite
5.2.8 Standard Work

Standard work relates to a clear specification as to how tasks should be performed.
im
The tasks are first organised in the best known sequence, and by using the most
effective combination of resources. This task sequence and combination of resources
rL
is well documented and every operator is required to adhere to this task sequence and
use the same resources. The intent is not to take away the creativity of performing
tasks from the operator. Rather, the intent is to make every operator follow
Fo
recognised best practices.

There are a number of reasons why standard work is important. First and
foremost, standard work promotes consistency and promotes continuous
improvement. If each operator performs tasks his or her own way, it would be very
difficult to improve a process. A documented standard makes it easier to effect
continuous improvement and increases the likelihood that the results will be
consistent. Standardised work improves safety. Unsafe practices are mitigated when
all operators are asked to follow the same routine when performing assigned tasks.
Moreover, with standards established, it now becomes easier to measure performance
fairly. It is possible to establish a ‘fair’ output rate and judge everyone by the same
standards. Standard work is particularly useful while training new employees.
y
Standard work recognises that the best practice may be a moving target.
nl
Sometimes the operators performing the tasks may come up with a better method for
O
accomplishing the tasks and therefore standard work practices could change over
time. It is important to note that standard work represents a set of tasks allocated to
an operator. Thus, standard work will change when the takt time changes or the
n
model mix changes. The goals in setting standard work are to give each operator an
io
amount of work less than or equal to the takt time while creating a compact foot print
for each operator. The operators should be involved in developing standard work.
at
That way, they will be more likely to do their job correctly and will help with
continuous improvement efforts in the future. Many organisations have created
ul
standard work charts for different takt times, requiring various number of operators.
These organisations are able to respond very quickly to takt time changes with
irc
minimal disruption since the standard work plan is already prepared.
C
5.2.9 Pull Replenishment and Kanbans

Pull replenishment provides a very convenient means of controlling the flow of
d
products through the various production processes. The goal of pull replenishment is
ite
to contain inventory. Pull replenishment is achieved in most implementations using

kanbans (the Japanese word for ‘sign’ or ‘signal’) although there are other methods for
im
effecting pull. Pull replenishment uses kanbans and is quite simple in concept. The
basic idea is to transfer production responsibility to the operators themselves, rather
than have a production controller decide in advance what each operator should be
rL
producing during a given time period. The decentralised control of operations is

accomplished by having each downstream operator signal their upstream operator(s)
when parts are needed; kanbans provide this signalling mechanism. To use kanbans,
Fo
the following conditions must typically be met:

➣ Demand for the item must be relatively repetitive.
➣ Lead times must be relatively short.
➣ Components must be available so an item can be produced on demand when
the visual signal is generated.
The first step in designing a kanban system is to determine the amount of

inventory that will be used to buffer the downstream operator from the upstream
operator. Two types of kanbans that are typically used are ‘in-process’ kanbans and
‘Material’ kanbans. In-process kanbans are typically used when the upstream and
downstream operations are fairly close to each other so that the upstream operator
y
can be triggered simply by a visual signal from the downstream operation. Typically,
nl
in-process kanbans are implemented simply by allocating a physical location between
the two operations and specifying the maximum amount of WIP inventory that can
O
be placed in this location by the upstream operator. If the in-process kanban is full,
the upstream operator must stop production until inventory is drawn from that
n
location. It is common practice to paint a square or a rectangle on the shop floor
io
between the two operations, indicating where the part (or a container of parts) should
be placed. If there are multiple part types, each might have a different colour code. Each
at
location or container has a number allocated to it, indicating the WIP inventory limit.
Material kanbans are used in a variety of ways. They are used by a production
ul
facility to signal replenishment of material from a supermarket/warehouse (a
‘withdrawal’ kanban), from another production facility (a ‘production’ kanban), or
irc
from a supplier (a ‘supplier’ kanban). A production kanban is used in place of an in-
process kanban if the upstream and downstream processes are separated by a
C
significant geographic distance, since the upstream operator has no visibility on the
needs of the downstream operator. The replenishment signals are usually transmitted
either through a card or an electronic signal. Each card requests a specific number of
d
items to be replenished. At any point in time, the number of kanban cards (or,
ite
equivalently containers of parts) that are in circulation is determined using the

following formula:
im
D × KCT × (1 + SF )
Number of kanban cards = +1 ,
Kanban size
rL
—
where D is the average demand per time period, KCT denotes the replenishment lead
time (kanban cycle time), and SF is a safety factor to buffer the combined variation in
Fo
demand, the variation in replenishment lead time and variation in quality of the
upstream supplier. The above formula assumes that the signal to replenish a container
of parts is triggered when the container becomes empty. This formula has two
unknown values—the number of kanban cards in circulation and the kanban sise (the
term commonly used to denote the number of items in a container). Typically, the
kanban sise is determined based on consultation with the supplier (or the upstream
facility or the supermarket as the case may be), based on a variety of factors including
standard container sizes, ergonomic issues, and so on. The formula is used to
determine the number of cards in circulation. Alternately, some organisations decide
to use a ‘2-bin’ system of replenishment, that is, they start with two kanban cards and
use the formula to determine the kanban sise. If that number is unsatisfactory, then
the number of cards is increased by one and the computation is repeated until a
y
satisfactory kanban sise is determined. Note that there should be a minimum of two
nl
cards, since there should always be one container at the workstation for the operator
to pull parts from. The withdrawal kanban signals are typically generated from a
O
production facility that keeps materials at point of use.
n
5.2.10 Point of Use Materials Storage
io
Point of use material storage places the materials at the point of use rather than in a
at
central warehouse. The upstream process or external supplier delivers the material
direct to the point of use, typically based on kanban pull signals. Typically materials
ul
are stored on flow-through racks located outside the cell. Flow-through racks are
storage locations with shelves that can be replenished from the back and consumed
irc
from the front, like the beverage coolers in convenience stores. Flow-through racks
preserve first-in first out material usage and allow material to be replenished from
outside the cell. For material that cannot be stored at point of use, supermarkets
C
(nearby storage locations) are often used in place of centralised storage. For example,
a supplier may deliver a large batch of parts used by several cells, once per day. Those
d
parts may be stored in the supermarket located near the cells and be pulled out in
ite
hourly quantities. A cell that produces in small batches may have only the materials
needed to make one product at the point of use at any given time. The materials for
other products will be stored at the supermarket.
im
5.2.11 Mistake Proofing and Method Sheets

rL
Mistake proofing and method sheets are tools used to prevent quality problems.
Mistake proofing or poka yoke is aimed at developing techniques to prevent defects
Fo
from being passed to the next process. Mistake proofing requires quality checks to be
built into the operations and equipment, using appropriate sensors to detect errors
and stop the process when necessary. Combined with other lean tools, mistake
proofing works to ensure that 100 per cent quality is built into the process and
product. An example of mistake proofing is the three-prong electrical plug. There is
only one way you can plug it into a wall socket.
Method sheets are visual instructions located at a workstation, showing how a job
must be performed, the quality checks necessary, and the tools to be used. The
instructions show pictures of each step to be performed. The goal is to make
instructions so clear and unambiguous that a new operator can immediately
understand them.
y
nl
5.2.12 Total Productive Maintenance
O
Total Productive Maintenance (TPM) is a term used to denote the systematic
execution of maintenance by all employees. The goal of a TPM programme is to
n
significantly increase productive capacity and decrease process variation while, at the
io
same time, increasing employee morale and job satisfaction. TPM is primarily an
equipment management strategy, originally developed by Toyota in 1970 to support
at
their Toyota Production System and it evolved from the Total Quality Management
(TQM) movement pioneered by Deming.
ul
TPM resembles TQM in a number of ways and the similarities are not
coincidental. Traditionally, maintenance can be classified as Corrective, Preventive,
irc
Predictive or Proactive. Corrective maintenance waits until a failure occurs and then
remedies the situation as quickly as possible. An example of corrective maintenance is
C
a sensor failing in an automobile. Preventive maintenance aims at maintaining

equipment at regularly spaced intervals to keep an otherwise troublesome failure
mode at bay. An example of preventive maintenance is changing the engine oil for the
d
automobile. Predictive maintenance examines vital signs displayed by the equipment,

ite
using instruments such as vibration analysers, to determine the health of the

equipment. For instance, predicitive maintenance would analyse engine wear. Finally,
im
proactive maintenance analyses why defects occur in, say, a machining operation, and
then designs the problem out of the machine. When the problems of maintenance
were examined as a part of the TQM programme, the general concepts did not seem
rL
to fit or work well in the maintenance environment. To solve this problem the
original TQM concepts were modified to suit this environment. These modifications
elevated maintenance to the status of a separate, yet integral, part of the overall
Fo
quality programme and gave it the name TPM.

TPM is a manufacturing led initiative that emphasizes the importance of people, a
continuous improvement philosophy, and the importance of production and
maintenance staff working together. The focus is on maintaining the equipment and
the processes that support manufacturing. It is a vital part of any lean implementation.
Because there is a natural tendency for people to follow the ‘if it ain’t broke don’t fix
it’ philosophy, TPM requires top management commitment to sustain the initiative.
5.3 CONTINUOUS IMPROVEMENT AND THE PURSUIT OF
y
PERFECTION
nl
Complacency can become the toughest challenge in any lean transformation process.
Lean is not a one-time implementation effort, nor is it a quality programme of the
O
month. It is an ongoing journey that requires a sustained effort at continuous
improvement. The transformation to lean is not just an application of a few
n
techniques; rather, it is a whole new way of looking at the operations of the
io
organisation. Because lean tools and techniques are often quite different from
traditional tools, there must be a sustained effort to operate with these tools. As we all
at
know, organisations that do not aim to continue an upward momentum will falter
and fall behind their competitors. It is essential to continuously re-examine processes
ul
and look for ways to take out waste and non value-added activities if organisations are
to gain significant financial performance improvements.
irc
To sustain lean implementations, it is therefore necessary for organisations to
constantly initiate kaizen events that promote continuous improvements. At the same
C
time, there is a need to promote kaikaku, the radical re-design of processes and
methods geared for achieving breakthroughs in performance and growth. (This step
is often referred to as a kaizen blitz.) Once a kaikaku step is applied, kaizen becomes
d
a powerful follow-up drive to perfect the processes and methods and to continue to
ite
adapt and be relevant.

As Womack and Jones [1996] suggest, ‘dreaming about perfection is fun.’
im
However, lean thinking provides ways to make these dreams a reality. Applying lean
thinking and implementing the tools and techniques to create flow result in a
fundamental change in the way the organisation thinks about its operations. The
rL
employees soon realise there is no end to the pursuit of perfection—efforts aimed at

reducing effort, time, space, cost, and mistakes in the process of producing and
Fo
delivering a product. When products flow faster through the organisation, they
expose hidden waste in the value stream. The harder you pull, the more the obstacles
are revealed so they can be removed.
While lean implementations must have commitment and support from top
management, the shop floor personnel are also critical to the success of lean. Many
organisations have initiated their lean efforts from the bottom up. Continuous
improvement (kaizen) events play a vital role in getting them engaged in the lean
journey, and that pays dividends. Managers at all levels become lean thinkers and
change agents. A truly lean organisation makes it much easier for everyone, including
shop floor employees, supervisors, lean champions, subcontractors, or first-tier
suppliers, to discover better ways to create value. Because the feedback loops are
y
significantly shortened, there is a faster feedback to employees, providing a more
nl
conducive environment for employees to pursue perfection.
O
CONCLUSIONS
n
io
Thomas J. Watson, Jr. once said ‘Whenever an individual or a business decides that
success has been attained, progress stops.’ Lean is a journey—an ongoing journey that
at
requires a sustained effort to maintain the momentum. At the same time, once the
initial resistance is overcome, it becomes much easier to maintain the tools and
ul
techniques of lean. Once there is a sharpened awareness, among the employees, of the
waste present in the system, there will be a concerted effort to maintain the
irc
momentum if the right incentives are provided. All types of organisations can benefit
from lean thinking, regardless of whether they are involved in manufacturing,
C
process, distribution, software development or financial services. Some of the

important points that organisations should keep in mind when embarking on a lean
journey are as follows:
d
1. While the goal of lean has often been identified as the removal of muda
ite
(waste), it is important to note that removing muda is just a means to an end.

2. The real goal of lean is to reduce lead time. Lean is all about lead-time
im
reduction and creating flow.

3. Some of the more important steps that organisations can take to create flow
rL
are takt time, standard work, pull replenishment and 5S. Incidentally, all of
these are steps that readily apply in a service setting as well.
4. Steps such as 5S and takt time are often misunderstood or misused. For
Fo
instance, the fifth step in the 5S programme, Sustain, is vital for sustaining
the momentum. Once a 5S programme is put in place, some organisations
neglect to practice this step. Lean is a journey one that really never ends.
5. Organisations that embark on the lean journey typically should therefore
start by identifying a product family they can apply lean principles to.
6. Lean thinking must be applied to all the processes in the organisation that
work on the selected product family(ies). The idea is not to simply lean out
some of the process steps and create a few islands of excellence. While some
waste may be removed in the process of creating such islands, the products
flowing out of these islands will end up stored elsewhere in the organisation.
y
In particular, they will queue up in front of the constraint resources. (This
nl
points out the need to apply lean thinking in conjunction with the Theory of
Constraints.)
O
➣ Finally, the lean supply chain is the ultimate goal. It is the responsibility of
the lean organisation to collaborate with upstream and downstream supply
n
chain members to successfully develop the lean supply chain.
io
at
ul
Conner, G., (2001), Jack Sprat Speaks: Tips on Lean Manufacturing, http://
www.thefabricator.com, August.
irc
Ohno, T., (1988), Toyota Production System: Beyond Large-Scale Production, Cambridge,
MA: Productivity Press.
C
Womack, J. P. and D. T. Jones, (1996), Lean Thinking, New York: Simon & Schuster.
Case references : The impact of lean thinking principles can be applied in the cases Titan
d
Industries Limited and Western Oil Limited (A), with special reference to product
ite
variety.
im
rL
Volpens Inc., is a company that assembles and markets ballpoint pens in three colours. It is
in the process of implementing a pull system for its operations, and has identified the
following monthly demand for its pens and shells.
Fo
Blue Pens: 3200 Red Pens: 2000 Yellow Pens: 1600

Blue Shells:1200 Red Shells: 1200 Yellow Shells: 800
The plant works five days a week, and runs eight hours each day. Coffee and lunch
breaks account for 80 minutes of a workday. (Time available per day is 400 minutes.)
Assume that there are exactly four weeks in a month.
Use the above data to answer questions 1 through 4.
1. What is the takt time?

2. At the final assembly shop, the total labour content for assembling a pen is 100
seconds, while the corresponding number for a shell is 30 seconds. What is the
y
minimum number of operators that Volpens must have in its final assembly shop in
nl
order to meet customer demand?
3. Volpens, Inc., wants to develop a kanban plan for yellow barrels. The lead time for
O
yellow barrels is two weeks. The supplier wants to ship the barrels in pallets that can
hold five dozen barrels. How many kanbans are needed for the yellow barrels? Assume
that a replenishment signal is faxed to the supplier as soon as a pallet becomes empty.
n
Use a 20 per cent safety factor. (Note: A yellow barrel is needed for each yellow pen/shell.)
io
4. Volpens wants to develop the mixed model sequence. Help Volpens get started on this
process by identifying the smallest possible repeatable sequence, that will include all
at
six models assembled by Volpens, and determine the number of units of each model
that should be present in this sequence. (Note: I am not asking you to develop the
ul
complete mixed-model sequence. That is, I do not need the exact order in which the
pens/shells will be assembled.)
irc
For the next four questions (questions 5 through 8), you will use the data provided
below.
C
The Pleasant Valley Health Clinic treats patients with respiratory illnesses. It classifies
patients according to four different types of respiratory problems (commonly referred to as
Diagnostic Related Groups, or DRGs): Bronchiolitis (BRO), Pneumonia (PNE),
d
Pharyngitis (PHA) and Sinusitis (SIN). Based on data averaged over the past six months,
ite
the number of patients it treats under each category is as follows: BRO: 15, PNE: 24, PHA: 25,
SIN: 36. The clinic works two shifts, from 7:00 am to 3 pm and from 3 pm to 11 pm. During
a shift, the staff is given a half hour lunch break and two rest breaks of 15 minutes each.
im
Each patient requires four process steps: Check-In (weigh-in, blood pressure check, etc.),
Evaluation (by a physician), Testing (X-rays, administering respiratory instruments like
rL
Pulse Oximeters, etc.), and Assessment (diagnosis and future scheduling). Average task
times for the processes (in minutes) are given in the table below:
Fo
BRO PNE PHA SIN
Check-In 30 min. 30 min. 30 min. 30 min.

Evaluation 20 min. 15 min. 10 min. 12 min.
Testing 45 min. 40 min. 15 min. 15 min.
Assessment 15 min. 15 min. 10 min. 10 min.
Work Content 110 min. 100 min. 65 min. 67 min.
In this clinic, tasks 1, 3, and 4 are typically handled by Licensed Practitioners (LPNs)
and Registered Nurses (RNs). Task 2 is handled by a physician. For simplicity, in these
exercises we will assume that all tasks are administered by RNs. For simplicity, again, we
will assume that the arrival rates of patients are constant through both shifts.
5. What is the takt time for Pleasant Valley Health Clinic?
y
6. Determine the average work content (‘labour content’).
nl
7. Determine the minimum number of RNs required using the average total work
O
content and the takt time for given demand (‘minimum manning’).
8. If the per cent demand for each request type remained unchanged, but the total daily
demand increased to 120 patients, how would the answers to 1 and 2 change?
n
io
at
ul
irc
C
d
ite
im
rL
Fo
y
CHAPTER
nl
6
SCM Across Organisations:
Upstream Interface
O
n
CHAPTER OUTLINE
io
Introduction 6.5 Contracts
at
6.1 Procurement 6.5.1 Illustrations
6.2 Strategic Issues in Procurement :
ul 6.5.2 Service Measures
Sourcing 6.5.3 Penalty Costs
6.2.1 Single Source versus Multiple 6.5.4 Contracts and Supply Chain
irc
Sources Goals
6.2.2 Long-term Contracts versus Spot 6.6 Markets and Auctions
Purchases (Markets)
6.6.1 Illustrations
C
6.2.3 Criteria for Supplier Selection and

6.6.2 Price Discovery
Assessment
6.6.3 Physical versus Electronic
6.2.4 Global Sourcing
d
Auctions
6.2.5 The Logistics of Procurement
References
ite
6.3 Operational Issues in Procurement :

Exercises
Purchasing
6.3.1 Quality Regimes (Change from
im
Inspection to Certified Quality)

6.3.2 Dynamic Prices
6.3.3 Electronic Exchanges and Auction
rL
Based Purchasing
6.3.4 The Role of Radio Frequency
Based Identification and
Information Technology
Fo
6.4 Items of Procurement

6.4.1 Raw Materials
6.4.2. Components
6.4.3 Conversion Capacity
6.4.4 Services
SCM Across Organisations: Upstream Interface 155
INTRODUCTION
T
he previous chapter pre sumes that the remaining part will be
sented some principles for bought or sourced. This goes back to
y
managing a supply chain the very core of specialisation, compe-
that would apply to an organisation. tence and factor endowments (mainly
nl
These principles are largely influenced physical) that determine what eco-
by the lean manufacturing paradigm. nomic activities take place in a given
O
We now address supply chain man- location. So Japan has steel plants
agement across organisations. Man- with iron ore sourced from India and
agers and analysts realise that it is the South America (and Japan often sells
combined activities of a number of steel to other parts of the world). The
n
supply chain partners that results in implication is that Japan is, in this
io
value being generated for end custom- case, forced to procure this raw mate-
ers. Management and IT tools are now rial. The U.S. has refineries which
available that allow, at least in prin- source oil from various parts of the
at
ciple, for better end-to-end manage- world including its own oilfields, as a
ment of resources, such as for tracking strategic measure. India sources engi-
the flow of commodities and sharing
information among supply chain part-
ul
neering goods for construction and
projects from some parts of the world
ners. because of the technology and cost-
irc
effectiveness involved in those prod-
Even though a number of these use-
ucts. The financial sector outsources
ful supply chain management tools are
technical services from India and East-
still being developed and refined, when
C
ern Europe for cost considerations.

we go beyond data visibility and try to
More recently there are a few instances
pin down decisions, the following di-
involving outsourcing of engineering
lemma emerges. When there is more
d
goods such as automobile compo-

than one organisation, it is not clear
nents from India.
ite
who is doing the managing and more

fundamentally, whether there is any- The ‘make versus buy’ decision was
thing to manage. The market as an in- initially discussed in an era where
terface between two firms may be too technology was scarce and, provided a
im
restrictive if it is the only mechanism reasonable product could be made,

available for managers. As a first step, goods could essentially be pushed into
we therefore look at some issues to do the market. Over time, similar deci-
rL
with actively managing activities sions needed to be made on the mar-

across a firm’s boundaries. ket side, as to whom to sell to and
The classical managerial decision at what to sell. In some cases, additional
Fo
the top of the hierarchy is the decision activities were taken on board to ad-
of ‘make versus buy’, including in a dress needs of consumers (such as
very general sense, the decision as to packaging). For some other cases,
what business to be in. A decision by a such as advertising and distribution,
firm to attempt to add value in some other partners were needed. In yet other
aspect of a product or service pre- cases involving modern international
supply chains, organisations are find- side, whom to collaborate with for dis-
ing the need for a whole host of trade- tribution. We first discuss sourcing and
y
related partners such as customs procurement. We follow that up with a
agents, export-import houses, trans- discussion on distribution, and include
nl
porters, warehousing agents, etc. a discussion on contracts, which pre-
sents an important way to deal with
O
The net impact of this is that a host supply chain partnerships in today’s
of decisions have to be made, both world. We take a broader view of pro-
upstream and downstream by any curement, to include capacity procure-
given player in a supply chain, as to
n
ment and services procurement, in ad-
what to procure and whom to procure dition to the procurement of material
io
from on one side and, on the other and components.
at
ul
irc
C
d
ite
im
rL
Fo
6.1 PROCUREMENT
The activity of procurement is economically justified by a number of reasons,
including availability of material, capability of conversion, and cost effectiveness of
conversion—due to superior technology or lower factor costs such as labour. In some
y
instances, the procurement activity is justified by the presence of regulatory regimes.
nl
The last one is a contentious activity worldwide, where ecologically harmful activities
are farmed out to regions that have less legislation and controls in place.
O
The materials procured can be raw materials, components or semi-finished
assemblies or complete assemblies, complete kits or of course, the entire product,
n
which is then traded. Procurement generally would have two aspects, a strategic
io
element in the form of sourcing and an operational element in the form of the
purchasing activity.
at
6.2 STRATEGIC ISSUES IN PROCUREMENT: SOURCING
ul
Sourcing refers to the strategic aspect of procurement, including the identification
irc
and selection of sources, the structure of the sourcing arrangement and the
geographical spread of the activity. A simple view of the sourcing activity could be the
C
matter of selecting, from among numerous options, where and how materials and
services are procured. The selection is typically based on some economic criteria.
However, even for this relatively straightforward decision there are a number of issues
d
that should be considered. Some of these are discussed below in more detail.
ite
6.2.1 Single Source versus Multiple Sources

im
As the term implies, single sourcing refers to the procurement of a given material or
service from a single source. The known advantages of single sourcing are the
rL
possibilities of large volumes of business and therefore lower prices, ease of long-term
relationships and collaboration in design and other initiatives and ease of traceability
(for quality related diagnosis and commercial matters like warranties). A single
Fo
source is more likely to be persuaded to align strategically with the buyer, make
investments in resources and technology on the buyer’s behalf, and form closer
relationships with the buyer, all of which have some clear advantages.
The flip side of the coin is that a single source makes the buyer vulnerable to the
performance of the single supplier. There is not much insurance against geographical
or other physical disturbances. More importantly, the sources for technological and
managerial innovations are now confined to a single source. These considerations and
the persistent search for cheaper sources lead organisations to look for, and contract
with, multiple sources. In recent times global sourcing possibilities because of
reduced trade barriers and also because of technologies such as the internet, may have
led to an opening up of the supplier base, for many multinational operations. This
y
may settle down to a new smaller set of suppliers (if not a single source), in many
nl
cases.
O
6.2.2 Long-term Contracts versus Spot Purchases (Markets)
n
Broadly speaking, we can identify two extreme modes of sourcing. One is through a
io
contract, which would typically be used when supply sources are uncertain—say,
because markets are not well developed, quality assessment of the supplied products
at
is difficult, and when there is a longer time frame intended for the transaction. The
other extreme is through spot purchases that take place in a market with many
ul
suppliers present, where each transaction involves a short time duration and where
the quality of the product is standardised and easy to assess.
irc
Sourcing arrangements with suppliers can be over different time frames. The issues
are somewhat similar to the single source versus multiple sources discussed above. A
C
long-term contract allows a supplier to offer discounts, as his capacity utilisation is

assured and he is willing to pay something for the risk free operation. On the other
hand, spot purchases may permit the exploitation of locally good prices, with the
d
downside that prices could go up as well. The other advantage of longer-term

ite
contracts is the ease of making logistical arrangements over a period of time.

A brief discussion of these issues in the context of global sourcing is made in
im
Section 6.2.4 below.

rL
6.2.3 Criteria for Supplier Selection and Assessment

Let us consider a number of suppliers, each of whom is technically qualified to supply
Fo
an item or a service. Apart from the unit price (cost to the buyer), which will always
remain the most important criterion, buyers are realising the importance of many
other factors. It can be argued that all these other factors (like quality, reliability of
delivery, matching with schedules of buyer, handling efficiency, etc.) add to cost
when suitably defined. While this is true, it does make the various criteria more
transparent and may be more convenient to monitor and assess. A criterion that is not
easy to put in cost terms is the design and innovation capability and flexibility
provided by the supplier. This has to be assessed in a strategic manner.
Apart from this, there is the obvious difficulty of somehow converting all these
factors to a common cost denominator. Multi-criterion methods and techniques to
y
discover weights and overall performance can be suggested here, such as the Analytic
Hierarchy Process (AHP) and Data Envelopment Analysis (DEA).
nl
A similar set of criteria is actually relevant for continuing supplier assessment over
O
a period of time. Monitoring of supplier performance, with benchmarks for key
indicators is a part of the sourcing strategy of any organised supply chain player,
today.
n
io
6.2.4 Global Sourcing
at
Global sourcing was always there for commodities, since producing regions are
strongly influenced by availability and extraction capability. At the other end,
ul
retailers have been importing finished goods for local consumption based on price
irc
advantages of the supplier (perhaps due to specific technologies and scale benefits).
What is relatively new is for manufacturing firms to source intermediate
manufactured goods—not raw materials or finished goods, but products that go into
C
other products and are either further manufactured or assembled as part of a larger
product, equipment or plant. These sourcing decisions are largely driven by a desire
d
to reduce costs. Indeed, the procurement function in many organisations is measured

ite
by a metric called purchase price variance (PPV). Essentially, the PPV approach
compares the current cost of procuring the material with the landed cost of
procurement from an overseas supplier. However, unless these decisions are very
im
carefully considered from a systems perspective, it could result in some serious

problems.
rL
Consider, for instance, the ramifications of sourcing a product from an overseas

supplier, as opposed to sourcing the product locally. There are a number of hidden
costs that need to be carefully considered. One of these is the additional pipeline
Fo
inventory that will inevitably accompany such a decision. Thus, if the lead time for
the local supplier was one week, then the pipeline inventory will be one week’s supply
of material. If the lead time for the offshore supplier (due to increased distance) is
four weeks, then the pipeline inventory will be four weeks of material. If the product
is a commodity product, then the effect of lead time may, arguably, be small.
However, as the beer game clearly demonstrates, the bullwhip effect will be much
more pronounced. Furthermore, there is a cost due to a loss of flexibility which is

extremely difficult to measure.
Many organisations are now realising that they need to take a more systemic view
of their offshoring decisions. They are trying to come up with a “total cost of
y
ownership” (TCO) instead of simply using the PPV metric.
nl
If the outcome of a careful TCO analysis is to still source the material from an
overseas country, then it requires follow up action. Large firms would now need to
O
segment their procurement activities into two or more categories, one through
dedicated suppliers, located anywhere in the world and second, through exchanges,
from a pre-approved set of global suppliers who bid on supplies for specific
n
requirements. A third category of occasional procurement can be through spot
io
purchases. The criteria for deciding which items are procured through which mode
would depend on factors, such as the need for strategic investments by dedicated
at
suppliers (which would then limit purchasing options by the buyer as part of the
incentive for the investment decision by the supplier). Another factor would be the
ul
standard design and specification of the items or materials to be procured, which
would be amenable to purchase through an exchange.
irc
6.2.5 The Logistics of Procurement

C
Materials that are procured have to be moved physically to their point of use. This has
d
to happen in a timely and effective manner to keep costs in control. The gamut of
issues to do with their logistics could well be the concern of the procurement
ite
function, especially when the buying firm is a larger, more organised one. Even
otherwise, firms may find it advantageous to influence the logistics of the supply, for
im
example, the warehousing locations that are used, the mode of transport or the
timing of shipments.
rL
These issues per se may equally well be the concern of the supplier firm (as part of
its distribution function), and with this in mind, we take them up in the next chapter.
If they are managed by the buyer, the span of decision making would be different, but
Fo
the substantive issues (costs, criteria for decisions in this area, etc.) would be the
same. For example, Tata Motors may like to influence the location of a warehouse
near JIT supply of components from key suppliers. An organised sector buyer for
food processing may take the initiative in mode of selection for transport and help in
getting good rates.
6.3 OPERATIONAL ISSUES IN PROCUREMENT: PURCHASING

Once the strategic (and tactical) part of procurement, which we have called sourcing,
is done, the remaining operational part of the activity is referred to here as
purchasing. As a commercial activity, purchasing is well understood in the business
y
world. Some relatively new elements to the purchasing activity are now discussed.
nl
O
6.3.1 Quality Regimes (Change from Inspection to Certified
Quality)
n
The Total Quality Management as well as the JIT revolution in manufacturing has
led to changes in the purchasing function in the organised sector. The notion of
io
purchasing being tied up to inspection and the provision of a store for incoming
at
materials (with areas for inspection, return, rework) is giving way to quality as source,
self-certified supply (perhaps with third party audits occasionally) and direct supply
ul
to points of use (lines or shelves as the case may be).
irc
6.3.2 Dynamic Prices
Purchasing in the electronic age means that prices of items may fluctuate from time
C
to time, sometimes day to day. Systems of purchasing have to be oriented to this, if

beneficial to the firm. The possibility of hedging against this uncertainty through
d
contracts or through futures is a realistic option, with the caveat that such strategies
ite
also lead to fluctuations in inventories and quantities in the pipeline, adding to the
causes of the bullwhip effect.
im
6.3.3 Electronic Exchanges and Auction Based Purchasing

rL
For commodity purchases and in selected industry niches, procurement through

electronic exchanges and partially automated auctioning mechanisms is now a reality.
This calls for strategies on the part of the buyer (and of course, the seller) to optimise
Fo
performance in this setting. A challenging issue in this context, which is very relevant,
is the combinatorial auction, where there are significant benefits in joint supply (and
therefore joint purchases) of items. This now has a large literature and some
applications that have succeeded in bringing about efficiencies both for suppliers and
buyers.
Examples of electronic exchange based purchasing can be found in the

pharmaceutical industry, construction industry and chemicals industry. The
automobile industry has made attempts to set up standards to facilitate such
procurement, but the attempt is yet to stabilise into a system where a large number of
different buyers can participate.
y
nl
6.3.4 The Role of Radio Frequency Based Identification and
O
Information Technology
An enabling technology that permits tracking of items across the supply chain and
n
potentially well into the usage cycle of the customer is RFID (Radio Frequency based
io
Identification). Through a combination of tags and readers at various locations in
supply chains, this technology promises a previously unattainable level of supply
at
chain visibility. Because of high profile initiatives of retail giants such as Walmart and
Target, suppliers worldwide are being forced to move to pallet or carton level
adoption of RFID.
ul
The general perception is that large retailers have more to gain from this as their
irc
availability and supply efficiency is improved. Questions that are being debated are
whether suppliers have any direct benefits in this technology, and what is the
C
optimum time to adopt the technology.

Associated with the RFID question is the issue of IT system alignment of the
d
supplier with the planning systems of the buyer. ERP systems and SCM software
platforms promise planning options across enterprises, but there are many decisions
ite
that supply chain partners have to take for this to happen.

im
6.4 ITEMS OF PROCUREMENT

We now look at another dimension of procurement, which is to do with what is being
rL
procured. Broadly, we can think of procurement of raw materials or components,

conversion capacity and services.
Fo
6.4.1 Raw Materials

Raw material procurement is now global as trading barriers have reduced, currency
exchange and a global financial system in place (barring a few pockets), and where
there is commodity standardisation so that customers in one part of the world can
procure and use raw materials from another part. The sources of major raw materials
are obviously limited to those areas where geographical conditions permit extraction
and transport. Logistical costs of handling and movement dominate the
considerations at this stage in the supply chain, although some conversion (e.g.
washing of coal, or sorting and grading of agri-commodities or pre-processing of iron
y
ore) could be there in some cases.
nl
One reason that direct logistics costs form a large part of the costs at this stage, is
O
because the material is in pre-processed form and is bulkier. Also, since the value
addition is not so high at this stage, service related costs are comparatively lower.
Transportation costs are an important consideration. An important point is that
n
different transport modes may provide the opportunity of different unit freight rates,
io
but what matters is the end-to-end movement cost of the commodity. This results in
low unit freight modes (such as water movement and rail movement) having to
at
provide interfacing facilities to be able to compete with end-to-end modes (such as
road movement). In trans-continental movement, the economies of scale in cost are
ul
so significant that large ocean going carriers are the dominant mode, even though
there are significant batching costs.
irc
6.4.2 Components
C
In the context of discrete manufacture, component procurement is a key strategy to

reduce complexity, to cut costs (by looking for low cost sources) and sometimes to
d
take advantage of high technology suppliers worldwide. The standard example of the
ite
automobile is very relevant here because a typical automobile has thousands of

components, and these (often in sub-assembly form or in kitted form) are procured
im
from a variety of sources by manufacturers. For components, the design and fit into
the final product assembly and the resultant performance is quite crucial (as opposed
to logistical costs). A general principle is that at stages where there is active value
rL
addition through technology, closer communication is required, between buyer and

seller. The information component in the procurement activity is much more because
apart from designs, specifications and quality records, there is the issue of
Fo
synchronisation with assembly schedules.

This concept is extended from components to sub assemblies to assemblies (e.g. a
pump manufacturer procuring assembled motors from a supplier), to of course the
entire product (which then makes the procurer a trading and marketing partner in
the supply chain).
6.4.3 Conversion Capacity

Another resource that can be procured is conversion capacity. Here, the buyer retains
the engineering and technology of the process (designs and specifications) and only
procures standard capacity (like machine hours on standard equipment). The cost
y
advantages are significant if the capacity is used by multiple users, in which case the
nl
fixed costs are spread across different users. The main issue for procurement is the
reliable availability of this capacity when required. If this is to be formalised, it leads
O
to different types of contracts (discussed later in the chapter).
n
6.4.4 Services
io
An increasingly important element of procurement is that of services. This can range
from logistical services (the emergence of third party logistics service providers and
at
developments thereof ), to certification and audit services, to specialised services in
the areas of maintenance, after sales support etc.
ul
irc
6.5 CONTRACTS
We look at contracts as a means of achieving supply chain related goals through
C
working with multiple organisations. Contracts provide the language and the
mechanism to specify and enforce supply chain performance measures across the
d
boundaries of firms. Contracts are of many types, to do with quantity commitments

(so as to plan capacity and utilisation), quality, delivery (so as to cut logistical costs
ite
due to uncertainties) and so on. Supply chain contracts are extensions of legal and
trade contracts which are more to do with specification of the transaction, ownership
im
issues, regulation and safety of consignments, supply etc. The enhancements come
about by analysis of the relevant supply chain and understanding its performance
measures and their link with contract parameters.
rL
Shippers who need logistics service providers and firms engaged in logistics
services today are grappling with a fairly fundamental issue of business significance. It
Fo
is the issue of defining service measures and contract parameters that would enable
them to operate fairly (vis-à-vis supply chain partners), while being competitive (vis-
à-vis other supply chains). This is true the world-over, but is of increased significance
in the emerging markets of Asia. One reason is that the large, and diversified markets
that are emerging in this part of the world are not necessarily very high value-seeking
ones and manufacturers and service providers have to provide customised products
and services but with lower margins. Added to this is the fact that assets and capital
are relatively scarce in this region and infrastructure is comparatively strained and
unreliable.
In concrete terms, this means that logistics service providers have to plan for cost
optimal asset utilisation plans, while at the same time account for the supply chain
y
consequences of poor service parameters. Eventually, this would have to be quantifiably
nl
captured and be part of contractual agreements between providers and users of logistical
services.
O
6.5.1 Illustrations
n
Let us begin by examining a few examples that illustrate the importance of this issue.
io
at
6.1
ul
An export oriented iron ore trader negotiating with the railways in India for supply of
ore from mines to port.
irc
In earlier days, if the Indian Railways (IR) were a feasible mode of movement of
iron ore, costing was done as per railway classification of commodities. This costing is
C
based on a complex mix of haulage cost for railways, ability of customers to pay,
strategic significance of the commodity and finally the lobbying power of the
d
industry in question. In a few cases, there would be a favoured customer status based
on volumes provided and some benefits in case the shipper participated in asset
ite
creation such as the Own-Your-Wagon scheme. But contracts were on quantities

only, and individual shipments would be carried as per traditional order placement,
im
indenting, rake supply as per availability, loading as per siding conditions (partly
controlled by IR and partly by the shipper), transit as per operating conditions on
rL
that part of the railway network, and finally unloading governed by conditions
similar to what is faced in loading (and additionally governed by port conditions in
this example). IR would attempt to protect itself against the risk of inefficient asset
Fo
utilisation (of wagons or rolling stock), by charging demurrage to customers who

detained railway rakes beyond a permissible limit, and by charging wharfage to
customers who utilised railway storage space.
At one level, these charges (demurrage and wharfage) can be considered simply as
asset utilisation charges that any service provider would use. But if these assets are
used dynamically for many different customers and usage by one customer impacts
reliable service offering to others in the system, the equation changes dramatically.
Demurrage is then a dynamic quantity reflecting opportunity costs or shadow prices
for different customer supply chains. Reflecting this accurately is one challenge facing
IR.
y
A more serious issue today, is that a project investor would ideally like a two-sided
contract, which would protect it against risks in a quantifiable manner. This would
nl
imply that the shipper would be willing to pay a premium in freight, but only for
O
reliable service. The flip side of a premium for reliable service is a penalty for
unreliable service. All service providers would have to gear up to be able to compete
in such situations.
n
io
6.2
at
Consider an express cargo mover specialising in re-location of household effects.
ul
Timing of its movements to enable customers to plan their moves is quite important.
Such a ‘move management’ company may not own all the resources required for the
irc
activity and may in turn hire assets from the market. While this keeps costs under
control and provides flexible capacity, it does provide for occasional unreliability of
C
operations (flexible capacity is inherently unreliable as investments are made with

multiple customers in mind and there is occasional pressure on resources). In such a
situation, a response of ‘This very rarely happens!’ or even ‘This has never happened
d
before!’ is not reassuring to customers, even if the statements are true. The immediate
ite
customer would like evidence of a contingency plan, which could include different
types of service.
im
With modern day customer referral systems and reviews on the internet, it is not
convincing to proclaim 100 per cent service, which customers would realise is a pipe
dream. More reliable reporting of service levels (hopefully good enough to inspire
rL
confidence, and increasing with time), with back up plans in case of failure, can be
more effective.
Fo
6.3
An owner of a small fleet of trucks receives a request from a larger transportation

company to provide resources for an ‘urgent’ shipment, which has got a deadline to
meet at a port. The transportation company is liable for costs of movement by a more
expensive, express mode, if the shipment is not made on time. It wishes to pass on
some of this risk to the trucker. Should the trucker agree or not, and if so, what would
be the arrangement?
Note that we are not highlighting the long-term effects of poor service (in the first
y
case, the iron ore company may develop an alternate mode over time, in the second
case, the logistics service provider may lose clients and in the third case, the small
nl
trucker may lose business with the transportation company). We are trying to see if
O
the short term, immediate consequences of service can be accounted for in a supply
chain transaction and set of activities. To an extent, this is healthy as it aligns short-
term incentives for action, with long-term goals.
n
io
We now explore some conceptual and managerial issues in this context.
at
6.5.2 Service Measures and Performance Measures in
Procurement ul
There are a number of service measures that can be used to monitor the effectiveness
irc
of logistics operations. Type-one and type-two service measures look at the number of
shipments or transactions that are on time, and the total quantity of supply that is on
time, respectively. Such measures are also used in measurement of distribution
C
efficiency in supply chains moving, for example, consumer goods. The type-one
measure respects each transaction and in a sense protects smaller customers, and the
d
type-two measure takes a more aggregate view of service.

ite
A minor modification of service measures of this type is to include an allowance of

some kind (such as the punctuality statistics published by Indian Railways(IR)). This
im
is actually unnecessary in real terms, as it depends on a self-defined allowance or

window of tolerance.
An internal consequence of using such measures in planning is that managers
rL
would factor in a safety time in operations, so that time reliability can be attained. An
example of this would be the cut-off time announced by courier collection centres for
Fo
making connections at hubs. When transport and other operations are moving
smoothly, customers who miss the cut-off times often find themselves
accommodated. Passenger train timetables on IR have considerable slack times before
arrival at the destination (which ensures good customer oriented on-time
performance) and at key intermediate points (which ensures good utilisation of
synchronised assets like crew and locomotives, sometimes at the expense of
unnecessary waiting on the part of customers). Such safety times and slacks are a part
of the premium many people are willing to pay for achieving reliability. One reason is
that synchronising actions across supply chain partners is more expensive than either
synchronising internally or incurring additional asset utilisation costs internally.
y
nl
6.5.3 Penalty Costs
A long-standing difficulty in the analysis and management of supply chains is to
O
quantify the costs of not achieving service targets. Penalty costs, shortage costs and
other similar costs are widely used in the analysis of inventory and other strategies,
n
but they are usually too notional and imprecise to form parts of practical planning
io
systems. Conceptually, such costs are clearly recognised at the strategic level of any
management. They include cost of express shipments or contingencies, costs of
at
storage of undelivered items, cost of returns, opportunity costs of lost sales and more
and more, the loss of customer good-will, reputation and business because of poor
ul
service. Managers therefore have increased the notional value of shortages and self-
imposed penalties for targets, compared to what they used some years ago.
irc
In retail environments, stock-outs are increasingly visible and given the average
rates of sales, one can reasonably estimate direct loss of revenue due to stock-outs.
C
One challenge of supply chain managers would be to translate perceived penalty

costs and shortage costs to meaningful service measures which internal operations
d
managers can factor into their planning. Logistics operations that are time-based are
ite
especially suited to measurement through service levels, as they are quick to measure
and display. The impact of investments on such a measure is however more difficult
to estimate and if required to be done a priori, would require tools such as simulation
im
or somewhat sophisticated stochastic models.

rL
6.5.4 Contracts and Supply Chain Goals

Finally, we highlight some of the ways the above concerns could be addressed in a
Fo
business setting. Contracts are a means of supply chain co-ordination and risk
sharing. Over a period of time, contracts have become increasingly meaningful and
significant. They first arose in supply chains to facilitate transfer of ownership legally,
then acquired commercial significance in booking capacity and assuring business
volumes, and subsequently have been used to specify quality norms in service. Today,
they play an important role in the significant areas of value addition and
competition, namely logistics services. A complete discussion of contracts of various

kinds is beyond the scope of this chapter, however.
Logistics contracts relevant to the discussion in this article would specify delivery
terms of handling and storage, quantities and dates of supply at origin, and desired
y
times of delivery. Penalties, one or two sided, could be part of the contract. Finally,
cost per shipment would always remain the most important part of a contract. Risks
nl
of many kinds could be covered in contracts, such as price protection against freight
O
rates. Considerable background analysis of operations (in addition to legal
implications, which dominated contract preparation in the past) is now required to
draw up a meaningful contract between supply chain partners.
n
Analysis shows that there is considerable scope for a supply chain driver to set
io
reasonable contract parameters that achieve overall supply chain goals as well as
incentivise local decision makers to choose ‘globally optimal’ operating parameters.
at
Contracts are finally agreed upon through a process of bargaining, negotiation and
co-operation between supply chain players. In some cases, inherently, supply chain
ul
goals align with optimal decisions through supply chain contracts that arise naturally
irc
through negotiation between supply chain partners. In other cases, incentives outside
a single contract may be required to achieve such goals, and these require innovative
measures. Contracts are again discussed in the context of distribution systems, at the
C
end of the next chapter.

d
6.6 MARKETS AND AUCTIONS

ite
In certain sectors such as the procurement of agricultural products, markets and

auctions are well-developed mechanisms of business. There would be a large number
im
of suppliers (e.g. farmers) and perhaps some number of buyers who would need to
interact in a systematic way to achieve supply chain effectiveness. Some dimensions
rL
of these mechanisms are now addressed.
6.6.1 Illustrations
Fo
Let us begin by looking at some illustrations of markets and auctions. Procurement of

copra by Marico all over the country is an example of an electronic auction. This
allows several small producers to interact with a key buyer (who accounts for some 25
per cent of the total purchases in this market). With self certification of quality based
on a small set of critical product quality parameters—such as moisture content—it
becomes an efficient means of procurement without physically moving material to a
site simply for effecting the commercial transaction. Now material can go straight to
the processing facility.
Procurement of tea in auctions in Kolkata and Guwahati are examples of physical
auctions. Here, the variety of products in the market is an important factor for buyers
y
who seek particular blends. So in addition to contractual buying from established
gardens (which is again picking up as a mode of procurement), auctions provide a
nl
convenient single point meeting of multiple buyers and sellers and outweigh some of
O
the logistical costs that are involved in this. The trade off between physical and
electronic auctions is discussed in Section 6.6.3.
Other examples of markets in the sourcing activity are tobacco, among
n
commodities, electrical and mechanical component markets (for standard
io
components for products such as motors, fans and the like) and the transport market
for trucks.
at
6.6.2 Price Discovery
ul
irc
Economic theory tells us that auctions are a good means to ‘discover’ the true price of a
commodity, as it allows for information exchange and revelation of the true costs and
values to sellers and buyers respectively. Although theoretically a single shot auction
C
would serve the purpose (with the modification that the winning person gets the second
highest bid—the so-called Vickery mechanism), in actual practice, a multi-call auction
d
may be more effective to facilitate information exchange among bidders.

ite
6.6.3 Physical versus Electronic Auctions

im
Physical auctions of commodities like tobacco, tea and other agricultural products
were a common phenomenon in past years and still have a value in terms of bringing
buyers and sellers to a common platform. It is being replaced in some instances by e-
rL
auctions where transactions are carried out on the internet, using a variety of
mechanisms to share information about bids, prices, quantities and other attributes
Fo
of a transaction between two parties. Electronic auctions, extending to commodity

trading, have some benefits such as simultaneity of transactions, information
visibility—provided for in different ways, decision making in one’s own managerial
context—which allows consultation and integration with other functions in a firm,
and finally, logistical cost effectiveness, as the commodity does not have to move
physically to a common location. This last point is true so long as the self certification
by sellers is reliable. This relates to the quality discussion in Section 6.3.1.
Arnold, T. J. R., S. N. Chapman, (1998), Introduction to Materials Management, Upper

Saddle River: Prentice-Hall.
y
Bapna, R. P. Goes and A. Gupta, (2003), Analysis and Design of Business-to-Consumer
nl
Online Auctions, Management Science, 49(1): 85–101.
Burt, D. N., (1984), Proactive Procurement: The Key to Increased Profits, Productivity, and
O
Quality, Englewood Cliffs: Prentice-Hall.
Corey R. E., (1978), Procurement Management: Strategy, Organisation, and Decision-
n
Making, Boston : CBI Publishers.
Grieger, M., (2003), Electronic Marketplaces: A Literature Review and A Call for Supply
io
Chain Management Research, European Journal of Operational Research, 144: 280–294.
at
ul
irc
1. For any typical assember of a complex product, e.g. an automobile assember, make a
list of components and sub-assemblies that you would propose for single sourcing
versus multiple source of supply. The criteria for doing this should be spelt out.
C
2. Continuing with the above situation, identify the items that can be purchased in spot
markets. What are the benefits of contracted procurement? For the automobile sector,
d
identify some mediums which facilitate spot purchases of standard items.

3. What is the implication of RFID technology for monitoring of inventory and the
ite
related inventory management system from the buyer’s point of view? How does it
affect supplier decisions? Is RFID technology driven by customer requirements or by
im
supplier capability? What are the benefits of an RFID system and investment in this
for a supplier?
4. For any auction mechanism that you are familiar with or can explore, try to see
rL
(a) What incentive suppliers have to reveal their true prices? (b) How information is
exchanged among the buyers and amongst the suppliers? (c) What are the
considerations for designing a good auction system (say for a standard component or
Fo
commodity)?
5. If a supplier enters a contact with a guaranteed service level, what are the implications
for its capacity planning? (You can assume a penalty of some amount if the availability
condition is not met.)
6. From an industry of your choice, identify whether incoming items are subject to
inspection and whether any scheme is followed for this (e.g. 100 per cent inspection,
sampling). Why are organisations moving towards quality at source and certified
quality instead?
7. As a supplier, would you like to offer stable prices or would you prefer varying prices
for your customers? What are the pros and cons of doing so? In, either case, think of a
mechanism to decide your price of supply.
y
8. Your purchase manager comes to you and says that she has negotiated a large discount
nl
in price from a supplier, provided the item is procured in bulk at a single time (for
example, the quantity discount is given because of a saving in transport cost by the
O
supplier). How would you react to this and in particular, how would you estimate the
overall savings as a results of this deal?
n
9. What are the benefits of outsourcing services, such as maintenance of equipment, in
your supply chain? Are there any drawbacks? What indicators would you use to judge
io
if you need to (a) enter into a structured contract, or (b) take over this task yourself in
future. For any service that you can define, try to design a structured contract with a
at
supplier (i.e. what conditions need to be specified, and how you would try to cost it?).
ul
irc
C
d
ite
im
rL
Fo
Fo
rL
im
ite
d
C
irc
ul
at
io
n
O
nl
y
y
CHAPTER
nl
7
SCM Across Organisations:
Downstream Interface
O
n
CHAPTER OUTLINE
io
Introduction
at
7.1 Distribution Management
7.2 Strategic Decisions in Distribution Management
ul
7.2.1 Channels of Distribution
7.2.2 Depth and Reach of Distribution
irc
7.2.3 Options in Logistics ManagementIn-House versus Third Party
7.3 Operational Management of Distribution
7.3.1 Material Accounting
C
7.3.2 Allocation
7.3.3 Inventory Planning and Control
d
7.3.4 Scheduling and Routing

7.3.5 Service Measures and Performance Measures in Distribution Systems
ite
7.4 Items of Distribution and Delivery

7.4.1 Semi-Finished Goods
im
7.4.2 Finished Goods

7.4.3 Services
7.5 Transportation, Storage and Warehousing
rL
7.5.1 Transportation Mode Choice and Multi-modal Transport

7.5.2 Developments in Storage Technologies
7.5.3 Warehouse Management
Fo
7.6 The Role of Contracts in Distribution and Marketing

7.6.1 Quantity Based Contracts
7.6.2 Time Based Contracts
Reference
Exercises
SCM Across Organisations: Downstream Interface 175
INTRODUCTION
I
t would be fair to say that the provisioning of requirements to the
spatial aspects of supply chain doorstep of the customer becomes
y
management have been more and more important. In many cat-
revolutionised by developments in dis- egories of goods today, the basic tech-
nl
tribution and logistics in the last two nology does change, but not as fast as
decades, with the pace of innovation in innovations in marketing, distribution,
O
these fields outstripping those in servicing, financing customer pur-
manufacturing and conversion tech- chase, and the accompanying areas of
nologies. An inevitable consequence packaging, storage, documentation,
of customer centric manufacturing and warranties and other value added ser-
n
supply of goods and services is that vices that meet needs of value seeking
io
distribution, including the last mile customers.
at
ul
irc
C
d
ite
im
rL
Fo
7.1 DISTRIBUTION MANAGEMENT

This chapter surveys some of the major developments in this area and techniques and
technologies for managing this increasingly important aspect of supply chains. The
chapter takes an operations oriented view of this activity rather than a marketing
y
oriented view and so we make only a passing mention of several important activities
nl
involved in this part of the supply chain, such as product positioning, display,
pricing, promotions, advertising, customer behaviour analysis, sales force
O
management and so on. We focus on material movement, storage, stocking and
related technological and methodological issues.
n
We note that distribution (from the view point of any one player in the supply
io
chain) has a big impact on the value of a product upstream in a supply chain, and
there, choice of mode and other strategic decisions of transport, handling and storage
at
need to be made. This is true for all bulk commodities (such as iron ore, steel, coal,
petroleum products). On the other hand, in downstream parts of a supply chain,
ul
where there are semi-finished or finished goods, logistics costs may not play a big
role, but they impact service measures and customer satisfaction in a big way.
irc
This chapter also sets the tone for the next one, which deals with the decisions and
management issues facing a logistics service provider. These players, who form a fast
C
growing business segment, especially in emerging markets, would plug into

distribution or supply needs of manufacturing or trading firms, so the issues
discussed in this chapter form the market context for those players.
d
ite
7.2 STRATEGIC DECISIONS IN DISTRIBUTION MANAGEMENT

im
The major decisions to do with distribution management at the strategic level are
structuring the channels of distribution and fixing the level of outsourcing for the
various activities in distribution. These are now taken up.
rL
7.2.1 Channels of Distribution

Fo
Even after the manufacturing part of the supply chain, there are often multiple stages
in a supply chain. For some products, these stages may involve repacking or break-
bulk operations, also called toll-packers in India, and other packaging related
activities. Although the internet and other means of direct marketing and supply are
now quite well organised with fewer intermediate stages of information and money
flow, there are often multiple stages of physical supply, for various reasons. After sales
servicing of products is one reason, and stocking a variety of products is often best
achieved by levels of the supply chain closer to the customer.
For commercial reasons, material is moved under own title close to various points
of final sale. For example, in India, central sales tax is applicable for sale by one entity
y
to another entity across state borders. Partly in response to this, many manufacturers
have their own branches, or outsourced agents (called Carrying and Forwarding
nl
Agents - CFAs) that operate in various locations. The material at these locations is
O
still owned by the manufacturing company. This has legal implications for liability, as
well as accounting implications in inventory accounting. From a supply chain point
of view, this introduces another stage in the physical supply chain, since material is
n
physically stocked at the CFA location.
io
Furthermore, it is not economical to supply directly in full truck loads or
equivalents, to final retail locations, in many cases. This has led to another level,
at
called the distributor, who maintains stock and does frequent supplies to the final
retail points, using milk runs (combinations of locations using a large vehicle) or
ul
other modes of supply. Traditionally in India, distributors are quite large and
commercially powerful and have long term equations with the manufacturing
irc
company. They play a key role in contact with the retailers, successful introduction of
new products and relationship with the final market as such. Again, from a supply
C
chain point of view, this adds another stage in the physical flow of goods.
d
7.2.2 Depth and Reach of Distribution

ite
An important decision to do with distribution management is the design of primary

and secondary stages of distribution in terms of depth and reach. The extreme
im
options are (a) a hub and spoke arrangement with one intermediate node and where
all movements are routed through this node, and (b) a direct shipment model.
Intermediate models with a small number of primary stocking points are also
rL
possible. Formal decisions of this type are based on mixed integer programming
models of network design.
Fo
Pure location models are important in this context while deciding on warehousing
options, similar to those while deciding on where to position manufacturing
facilities. With options of sub-contracting on the one hand, and leasing space or
capacity on the other, there are both long and medium term decisions to be made
regarding facility decisions in the distribution part of the supply chain. Apart from
factors that identify feasible options, the pure location costs (such as land cost or rent
or utility costs) need to be incorporated into multi-criteria decision analysis in these

location models. In case subsequent decisions are also planned, such as resulting
transportation and supply to multiple locations, the models become more complex
(involving both long run and medium run costs) and the class of location-allocation
models becomes useful. These have been successfully applied in various sectors such
y
as the food industry, electronic goods and these days, in retail supply chains.
nl
O
7.2.3 Options in Logistics Management—In-House versus
Third Party
n
A key decision facing firms is whether to take on the role of logistics (including
io
warehousing and storage) and transport themselves or leave it to their customers or
suppliers, or to enter into contracts with service providers and manage that interface.
at
Each option has its pros and cons and in Indian industry today, all of them have
successful (and unsuccessful) examples.
ul
The option of leaving transport and logistics to a customer or a buyer is adopted
either by smaller players, where a large player is controlling movement of goods, or
irc
some very large players whose core competence is the static value addition or trading
benefits (especially price) that they provide. It is more common for supplier to make
C
arrangements for transport than for buyers to make arrangements for transport,
partly because of commercial considerations and the point of change of title.
d
Managing In-House Logistics

ite
Managing own logistics has become a strong business strategy for major players in
many different segments. It allows them to control an end-to-end process, and for
im
large players, turns out to be cost effective since the internal business volumes are
sufficiently large to justify asset investments. Mahindra, Tata Steel and Godrej
Appliances are three players who have taken this route in India. In fact, what starts
rL
out as an internal service function soon turns to offering these services (logistics
management) to other group companies and even outside companies and in some
cases even competitor companies. Logistics organisations in the three examples
Fo
quoted above have evolved into strategic business units.
Procuring Third Party Logistics Services

The intermediate option of using a third party provider is an attractive option for
fast-growing firms who lack the managerial resources or the volumes of business to
own and operate logistics relevant to their needs, but who still need a professionally
managed operation to meet current standards of competition. The notion of service

measures discussed above, plus the cost economics of transport and storage are
relevant to be able to successfully manage this interface. Viewed as a procurement
activity (of logistics services), the discussion on contracts in the previous chapter is
relevant here.
y
Firms sometimes make a strategic decision not to enter into logistics related
nl
activities and prefer to procure these services even when volumes are potentially large.
O
Maruti and some of the big players in electronics, computers and white goods are
examples of such firms. Multinational firms in emerging markets often adopt this
strategy to ramp up business in a phased way without committing assets up-front.
n
With the emergence of high value outsourcing by such firms, there has been a growth
io
in the third-party logistics sector, which is the subject of the next chapter.
Broadly, decisions to do with procurement of third party logistics services is
at
intricately tied up with the design of the channels of distribution, discussed earlier in
this chapter. A specific factor that may be relevant is the mode of transport that is
ul
used. The onus of choice of mode is shifting from the shipper to the third party
irc
logistics provider, who often has multiple modes of transport and will use an
appropriate one. For example, a shipper may negotiate port to factory shipment
delivery of containers, with a third party logistics provider. The rates may be
C
attractive because the carrier has access to rail movement of containers. But if rail
services are not available (occasionally), the carrier may have the option of switching
d
to the more expensive road mode and still deliver as per the desired schedule. The
ite
occasional use of such options can be factored into cost computations by the carrier.
Such options may become increasingly common in a world where different transport
modes are increasingly available, containerisation is making the use of multiple
im
modes possible and where service norms are becoming tighter. For this analysis, size
of vessel (e.g. light commercial vehicle versus large truck) can also be considered to be
rL
different modes of transport, as far as cost and service computations go.

Fo
7.3 OPERATIONAL MANAGEMENT OF DISTRIBUTION

Some key elements of the operational management function in distribution are
material accounting and control, allocation, inventory control and routing and
scheduling. These are briefly discussed, in turn.
7.3.1 Material Accounting

The basic logic of inventory balances and accounting of material at multiple
locations forms the heart of Distribution Resource Planning (DRP) systems in many
different versions in the industry. The main features of DRP account for stocks of
y
items at different locations, and allow for different stocking units (SKUs at the retail
nl
level, and containers, pallets and packages at upstream levels). An issue of growing
importance is accuracy in inventory records (due to losses, perishability, pilferage
O
etc.).
Bar coded material storage and transfer units are the first step towards electronic
n
records and quick retrieval of information in a form compatible with computer
enabled management of activities. A subsequent step is to have radio frequency
io
enabled communication between bar code readers and a database for planning
applications (e.g. where a local handheld device communicates with a central
at
computer). A further step is an RFID enabled system, where there are readers that
ul
automatically pick up information through tags on items. As of now, tags are
economical and being used at the level of containers or pallets and it is anticipated
irc
that they will penetrate to the individual SKU level as costs come down. These
systems will make material accounting timely and accurate. RFID is discussed in
Section 9.5.
C
7.3.2 Allocation
d
ite
In a supply-constrained environment where there are a few manufacturing locations

and many demand locations, a basic distribution decision is the allocation problem.
This, while theoretically of some fundamental interest, gets resolved, often forcibly,
im
by linkages developed over time, and regional regulations, which do not permit too
many shipment choices (e.g. tax implications of inter state or inter country sales) and
rL
simply the availability of reliable transportation between various pairs of locations.

The resulting scenario is then often simplified to an extent where a heuristic or a
manual solution is good enough.
Fo
In situations where a formal model is required, the transportation (or trans-

shipment or allocation) model from linear optimisation encapsulates the basic single
period decision. This well-studied model, while sometimes not directly useful, is
nevertheless quite widespread in its applicability and forms the basis for more
complex decisions and even some heuristic techniques. An example is that of
locomotive assignment for railway operations on a large network, as part of the
internal supply chain of a transportation service provider. Another situation is a

consortium of oil companies trying to co-ordinate supplies from refineries to
stocking points. A more traditional example is that of a multi-plant manufacturer
supplying to several wholesalers in a state.
y
The basic version of this decision involves a single, homogeneous commodity; the
multi-commodity version of this has some complications, referred to briefly in
nl
Chapter 10.
O
7.3.3 Inventory Planning and Control
n
In what are called pure push systems, inventories are dispatched downstream as per a
io
central plan based on committed demands or forecasts. These are useful where there
are large economies of transport, and up to the level where demand is steady or
at
known well in advance. As we get closer to the downstream part of a supply chain, the
demands (of individual customers) becomes uncertain, mainly because of choices
ul
available to customers (variants as well as competition) and getting close to the
individual value establishment by the customer, which is difficult to predict very
irc
accurately. Conversely speaking, in upstream parts of a supply chain, uncertainty is
less because of aggregation or pooling of multiple random influences, which does
C
tend to make for more stable behaviour. In this setting, inventory planning is done,
which includes what are called lot sizing rules, to take advantage of scale economies.
The well-known EOQ (Economic Order Quantity) model is a repetitive, periodic
d
version. These are further simplified to take into account transport batch sizes and are
ite
often implemented in some periodic manner, for ease of planning and

implementation (e.g. shipments to some location once a week, some other locations
im
twice a week). This gives predictability and regularity to the entire system and allows
downstream players to place orders according to an announced plan.
In contrast to push systems, we have pull systems that are characterised either by
rL
re-order levels (and orders placed at varying times) or re-order times (with varying
order quantities). The main consideration here is that demand is random (not
Fo
entirely predictable) and the term used here is that of inventory control. Assuming
there is some knowledge of this random demand, in the best case, a demand
distribution, there are well developed theoretical bases for both types of systems.
Broadly speaking, the logic for order quantity is a modification of the EOQ to
account for unplanned stockouts or holding, but whose consequences can be
estimated in the long run. They result in (R, Q) policies, where R refers to a re-order
point and Q the fixed order quantity. The policy is to place an order of quantity Q
whenever the inventory level drops down to level R. This analysis depends quite
crucially on the lead-time of supply, because it is really the uncertainty of demand
during the lead time that one has to plan for, in a probabilistic manner.
y
The periodic review version of this results in a very well-known class of policies
called (S, s) policies, which have an ordering trigger(s)—where an order is placed
nl
provided the inventory level is below s, and an order-up-to level (S). Note that the
O
extent of the order is variable, but is at least (S-s), which captures economy of scale of
order fulfillment. An easily implementable approximation of such policies are two
bin policies, where an order (equal to the bin size) is placed when one bin is empty
n
and the other bin is used till the replenishment arrives. The one time (single period)
io
version of this problem is also referred to as the newsvendor problem, which
determines the base stock. This analysis has an elegant and intuitively appealing
at
structure to the optimum quantity which captures the demand characteristics and
cost characteristics in a neat way. ul
See Chapter 10 for more discussion in these models.
irc
7.3.4 Scheduling and Routing
C
In distribution logistics, scheduling of shipments is tied up with inventory planning,

since the frequency of ordering is determined by inventory and ordering economics.
d
As one goes up the supply chain, demand becomes more stable, and transport and
logistics costs are a bigger fraction of total costs. Scheduling at the strategic level helps
ite
in planning for requirements of transport assets and contracting with logistics

providers. At the next level, it allows for planning horizons to be shared between
im
supply chain players and makes for streamlined activities. Finally, at the lowest level,
it is tied up with the allocation decision of what goes where in a given time period
(like a day or a week).
rL
Routing decisions are considered in Chapter 10, and they form part of the core set
of decisions a transporter or a logistics service provider takes.
Fo
7.3.5 Service Measures and Performance Measures in

Distribution Systems
Performance measures have already been discussed in Chapter 4 in the general setting
of supply chain management. We discuss some specific issues arising in the design of
performance measures in the context of distribution systems.
In the design of any distribution system, the criteria for deciding on a good system
would mean cost considerations on the one hand, and others that reflect the service
goals of the system. In the area of distribution, service measures such as the
proportion of orders fulfilled on time (formalised by different types of fill rates) are
important, as also discussed in Chapter 6. To keep a check on inventory costs,
y
measures such as amounts of inventory that become obsolete are used. These
nl
indirectly indicate accuracy of forecasting—and therefore an understanding of the
market, and the design of a good order fulfillment system that supplies the right
O
items to the right location at the right time.
Warehouse managers would need to develop and use specific measures such as the
n
retrieval time, or turn around times of stock, in addition to cost measures (such as
io
turnover per unit of space).
Like all performance measures that attempt to capture the behaviour of complex
at
systems such as supply chains, the measures are useful up to a point, to draw attention
to various elements of cost and service. Very soon, systems adjust themselves to satisfy
ul
one measure (which is viewed as important by management), often at the expense of
other measures that actually contribute to the overall effectiveness. In that sense,
irc
measures themselves are used quite dynamically, thereby causing some distress to the
system in consideration. Despite these limitations, performance measures and their
C
understanding form a crucial role in modern supply chain management.

d
7.4 ITEMS OF DISTRIBUTION AND DELIVERY

ite
In a manner symmetric to the upstream interface, we look at various possibilities of

distributing and delivering goods and services.
im
7.4.1 Semi-Finished Goods

rL
In a supply chain, a firm can supply semi-finished goods to its customer. This means
that there is further processing to be done before the end customer is able to derive
Fo
value from the product. An important issue here is the synchronisation of deliveries
with the process requirement of the customer. Semi-finished goods are the supply
chain equivalent of WIP in the manufacturing activity. There are powerful reasons to
streamline the flow of such goods in the supply chain, in keeping with the lean
principles of Chapter 5.
Another important issue is quality and warranty liability on the part of the
upstream player. For example, an OEM would provide service quality through the
warranty or service guarantee provided by its customer to the final user.
y
7.4.2 Finished Goods
nl
If a firm is supplying finished goods to a market, it means that there is no subsequent
value addition through conversion of the product. There may still be a locational
O
transfer, stocking and final supply through further stages (all of which continue to be
inportant to customers in terms of utility derived from place, time and availability).
n
The issues here are more to do with information tracking and information sharing
io
among supply chain partners. Initiatives such as VMI (Vendor managed inventory) at
retailers have been tried to provide for directness of response. The retailer in this case
at
provides shelf space and it is the supplier who has to monitor availability, decide on
frequency of shipments to back-end store locations and replenish stocks.
ul
Bar coding and scanning (for robust and quick data capture), radio enabled and
irc
later full-fledged RFID technologies (which enable real time and automatic data
capture) are beginning to show potential for increased visibility in the supply chain.
One advantage of such technologies is that they provide a direct interfacing with
C
other planning tools such as an ERP system.

d
7.4.3 Services
ite
Service delivery is an important part of supply chain management, and although we

do not discuss it fully here, we point out some key features. Unlike tangible products,
im
service is difficult to provide ahead of time and store, so time synchronised supply is
very important. We do point out that anticipation of the type of service that may be
required can ease supply chain transactions considerably, and perhaps cut costs
rL
significantly. But generally, capacity utilisation is not as much of a goal here. Rather,
idle resources are often justified if they can be used to achieve acceptable levels of
Fo
service in peak periods.

A service provided by logistics players is customer side tracking of goods, which
allows for considerable reliability of planning on the downstream side. As much as
compensation for late delivery and claim settlement, it is information visibility that
customers seek in supply chains.
7.5 TRANSPORTATION, STORAGE AND WAREHOUSING

The distribution function tries to match the requirements of the customer in time
and in space. Transportation, including choice of mode, is discussed in greater detail
in the next chapter. While transportation would take care of the physical bridging
y
regarding this match, it is storage and warehousing that will allow for the time match
nl
to be made.
O
7.5.1 Transportation Mode Choice and Multi-Modal Transport
n
A supply chain perspective to transportation mode choice is the following. The end
to end requirement of movement is what needs to be considered. Most modes of
io
transport, except road, would not directly fulfil mine-to-factory or factory-to-factory
or factory-to-store needs. In that sense, except in a few cases where the economics
at
would permit pure road based movements, it is a combination of modes that is used.
ul
irc
7.1
Iron ore that moves from Karnataka to Japan and Korea moves by truck to rail
C
loading points, then by rail to unloading points, then by truck to barge loading
points, then by barge on the Mandovi-Zuari canal system in Goa to loading points
d
for ocean going ships at Marmagao or Panaji port (sometimes with staging of the
material at the port). So even in the movement prior to the ocean movement (the
ite
only realistic mode for that segment of movement), there are a variety of modes
(conveyors, trucks, rail, barges), each with its own role to play in the supply chain.
im
In this example, which in itself admits several variants in the flows, the bulk of
material is such that road (using trucks) is too inefficient in terms of managing such
rL
units of movements (10–12 tons per standard truck) over long distances. Rail is used
to bridge part of the gap, but because of port side infrastructure and also line capacity
considerations in the last lap, barges on the inland waterway system have emerged as
Fo
cost effective, environmentally acceptable (in comparison with hundreds of trucks

carrying iron ore through hebetated areas) and fuel efficient mode for that leg. But
since barge movement is not end to end, there are a number of bridging technologies
to handle the movements at interfaces. This example is analysed in detail in Example
8.4 in Chapter 8.
India has seen several shifts of transport mode in recent times. Cement from
Gujarat to South India has moved from rail to coastal shipping, often with
manufacturer stakes in jetties and port infrastructure. Iron and steel products from
steel plants have moved from rail to road because of smaller batches and more value
addition and innovations in handling products by road. The flexibility in batch size,
y
destination and timing of supply offered by road proved to be too much to ignore.
nl
Petroleum has moved to pipeline in a number of instances. With multi-modal
transport, there has been an emergence of third party logistics service providers who
O
sometimes take on the task of co-ordinating across modes.
n
7.5.2 Developments in Storage Technologies
io
From a practical point of view, storage during transport and at various stocking
at
locations is quite important, given the quality standards that customers expect of
final products. The most important of these developments is in cold-storage facilities
ul
and refrigerated movement for agro products and food products. These have led to
the development of cold chains, which allow for such products to be shipped across
irc
the world.
In India, professional warehousing infrastructure is now emerging. For example,
C
container freight stations and inland container depots for handling of containers are
coming up near all major ports.
d
ite
7.5.3 Warehouse Management

A host of design and operational issues arise in warehouse management. Some design
im
issues are to do with choice of material handling equipment, cross docking

arrangements and information system design for data capture and communication.
Some of the operational issues are deciding on stacking locations, order picking
rL
strategies and grouping of orders. A major area of development is to do with

information system integration with retailers who place orders with an increasing
Fo
frequency (and smaller batches) and manufacturers. Consolidation of orders for

delivery to a given location and sometimes combining locations for a transport
shipment (the milk-run) are made possible by systematic warehouse management
practices where order locations are known, where there is some automation in
material handling, where commercial documents can be quickly and accurately
prepared and where inspection and reception procedures are in place.
7.6 THE ROLE OF CONTRACTS IN DISTRIBUTION AND MARKETING

If we accept the role of multiple supply chain players who will stock products at
various levels (called echelons in the technical literature), and take active ordering
decisions, the following issues emerge. They are all a consequence of each player
y
having its own objective vis-à-vis the supply chain goal.
nl
Economists have long recognised that the introduction of multiple players each
with its own margin of profit would force decisions that are sub-optimal from the
O
supply chain’s point of view. A concrete example of this is the following. In the
presence of uncertain demand, a retailer would balance the costs of overstocking
n
(typically inventory holding costs) with the costs of under-stocking (the typically to
io
do with opportunity costs, including profit margins that are foregone). If this is done
by each player sequentially, the supply chain as a whole would order less than the
at
optimal quantity. This phenomenon is an example of what is called double
marginalisation. This is discussed in Chapter 10, in Section 10.5.2.
ul
7.6.1 Quantity Based Contracts
irc
Practically speaking, if supply chain players remain distinct entities, a way out of this
C
dilemma of double marginalisation would be for the supplier (upstream player) to

provide the retailer (downstream player) with some incentives. This can be done in
different ways, through quantity based contracts, as discussed below.
d
ite
Buy Back Contracts

Here, the manufacturer agrees to buy back unsold items at or close to cost. The
im
retailer still has an incentive to sell what he can, because the margin is attractive, but
the risks of unsold items are wholly or partly borne by the manufacturer.
rL
Revenue Sharing Contracts

In this type of contract, rather than an outright sale and defining a margin at the
interface, the manufacturer shares a fraction of all sales that the retailer makes, in
Fo
return for a lowered price of sale. This again increases the retailer’s ability to take risks
on the amount ordered/stocked and comes closer to the ‘optimal’ balance point.
Quantity Flexibility Contracts

In such contracts, there is an option for the retailer to change the final quantity
ordered after observing some of the demand. This is prevalent in the apparel industry,
based on early signals of fashionable trends.
In all these type of contracts, while they try to spread risks and assuage the problem
of double marginalisation, they do lead to certain information distortions with
regard to the ‘true’ demand at various stages. Often in such contractual settings, there
is a leader who would take certain decisions (e.g. announce a revenue sharing
mechanism) and a follower who would then act accordingly (e.g. decide on an order
y
quantity). To some extent, this behaviour can be anticipated and analysed in an
nl
appropriate multi-level optimisation setting. In game theoretic terms, these would
lead to leader-follower or Stackelberg games. Some aspects of pricing decisions are
O
taken up in Section 10.5.
n
7.6.2 Time Based Contracts
io
An example of a time based contract in a supply chain context is a performance based
at
contract, such as ‘pizzas in 30 minutes or money back’. The ‘optimal’ design of such
schemes may be a very subjective issue, dealing with many considerations of brand
ul
positioning and marketing effectiveness, but one thing that is certainly ‘objective’ and
useful is an estimate of the costs arising from such a scheme. Given a demand pattern,
irc
whose knowledge is anyway essential for any serious business, it may be possible to
compute such costs (at least on average) through analytical or simulation techniques.
C
More complex contracts can be envisaged, which take into account the time based
performance. JIT contracts, where there are penalties for late supply (because that has
implications on safety stocks and downstream activity delays of the customer) as well
d
as for early supply (which has implications on storage, perishability and double
ite
handling, for the customer). Such contracts then have implications on the optimum
time of order placement, among other decisions.
im
Supply chain managers would have to be responsible for questions of the following
kind. Customer X wants a service guarantee of delivery within 3 days of placing an
order (subject to a maximum of K units a week). Customer X is willing to pay a
rL
premium of Rs Y for this, but on the other side, demands a penalty P for every unit
that fails to meet this target. Should we agree to this proposal? We note that the
Fo
answer to this question has firm-wide implications on capacity management, because

of capacity that is already committed and sub-contracting options, apart from
marketing considerations. While a general answer to this question is not possible,
specific analysis of a given situation is perhaps achievable.
Bowersox, D. J., D. J. Closs, O. K. Helferich, (1986), Logistical Management: A Systems

Integration of Physical Distribution, Manufacturing Support, and Materials Procurement,
y
New York: Macmillan.
nl
O
n
Questions 1–6 refer to the description below. Femiliarity with the material in Section 10.3
may permit enhanced analysis of these issues.
io
Distribution of Fertilizer
at
Rashtriya Chemical Fertilizers (RCF) is considering the rationalisation of the distribution
of urea in Karnataka. Urea has to be moved from the factory in Bombay to five potential
ul
Rake Point Warehouses (RPWs) where it is required. The demand for urea in a particular
season at the nine Feeder Point Warehouses (FPWs) is given in the Table 1. The capacity at
irc
Bombay is adequate to satisfy the total demand. The cost (in Rs/ton) of distribution from
Bombay to the five RPWs is given in Table 3 and the cost of distribution from each RPW to
C
each FPW is given in Table 2.

Note: Usually the rail mode is used for distance above 300 km and the road movement is
used for distance less than 300 km. Hence, the costs between Bombay and the RPWs are
d
potential ones that the management is considering for the primary warehouses.
ite
Management wants minimise the total cost of distribution from the factory to RPWs and
then on to the FPWs but at the same time have only a few RPWs in order to have better
im
control.
(1) Provide the management with a trade-off between the number of RPWs and
rL
minimum total distribution costs.
(2) If the fixed costs of opening a warehouse are not the same, then provide the
management with a model to make the decision. How might fixed cost be estimated/
Fo
calculated?
(3) The management would also like to keep a percentage of the demand to provide for
fluctuations (from forecast) of demand at the FPWs. Develop the logic to allow
management to decide the additional stocking quantities. Show how this quantity
will decrease as the number of warehouses increases.
(4) The transport link between the following pairs of locations is not established and has
a cost associated with them: Gulbarga—Belgaum, Bellary—Raichur and Hospet—
Hassan. The company may not want to incur this cost at this time and would like to
do so only if some immediate benefits flow. Before contracting with transporters, the
company would like to know what sort of cost savings would accrue.
y
(5) Demand is placed once a month to the FPWs around the 25th of every month. In the
nl
peak season (March to September), they are placed once in 15 days. They are placed
at the RPWs after a gap three to four days and to the factory after another two to three
O
days. The lead times for supply are typically one week from RPWs to FPW and three
weeks from factory to RPW. FPWs are in more direct touch with dealers and final
n
demand and are apt to react vigorously to fluctuations. They placed orders
extrapolating both increases and decreases in demand for the month ahead as well i.e.
io
if the demand is 10 per cent above the forecasted value, the assumptions is that next
month too, the demand will be that much more than the forecasted value. Estimate
at
how long it takes for such a system to settle down (in terms of steady state pipeline
inventories).
ul
(6) RCF is also considering direct supplies to some FPWs (bypassing the RPWs). They
were warned that this would lead to much larger management of pipeline stocks.
irc
Therefore, RCF would also like to estimate the impact of its location policy on
pipeline stocks. Given the typical lead times, make assumptions about the fixed and
C
variable part of lead times (fixed part is the order processing loading of the rakes, etc.
and the variable part depends on the distance and location), estimate the pipeline
inventories in your distribution network solution.
d
ite
Table 1: Forecasted Requirements at Marketing Districts in Tons

im
Marketing Dist. Jan Feb Mar Apr May Jun Jul Aug
Gulbarga 32 26 24 35 57 69 75 52
rL
Belgaum 18 13 15 20 32 38 50 31
Bangalore 35 33 23 35 41 46 54 32
Mysore 18 17 18 20 23 27 32 19
Fo
Dharwar 12 13 8 11 15 20 26 16
Raichur 13 16 23 26 31 48 53 34
Bellary 10 10 4 7 10 16 14 10
Hassan 12 11 4 8 11 17 23 12
Chitradurga 8 7 2 4 9 15 15 9
Table 2: Cost Matrix (Rs/Ton)
Bangalore Hospet Raichur Gulbarga Bellary
Bangalore Nil 156 185 386 157

Raichur 113 103 Nil 156 108
y
Gulbarga 133 139 124 Nil 156
nl
Bellary 102 42 105 204 Nil
Hubli 130 83 145 176 108
O
Belgaum 139 120 162 180 138
Mysore 83 161 227 283 182
n
Hassan 102 139 201 270 191
io
Chitradurga 115 Not known Not known
at
Table 3: Cost Matrix (Rs/Ton)
Bombay to
ul
irc
Bangalore 320
Hospet 203
Raichur 203
C
Gulbarga 145
Bellary 230
d
ite
(7) What are the pros and cons of managing one’s own logistics vis-à-vis going in for third
Party Services? Further, India now has examples of organisations that offer logistic
services to other organisations, including their customers and sometimes even
im
competitors, apart from meeting their internal requiements. What is the benefit of
this?
rL
(8) Design an appropriate service measure for a warehouse storing multiple products.
How would you judge whether a storage policy in a warehouse is effective based on
some quantitative measure of performance?
Fo
y
CHAPTER
nl
8
Transportation, Storage and
Warehousing
O
n
CHAPTER OUTLINE
io
Introduction
at
8.1 Transportation Mode Choice
8.2 Transport Operator Decisions ul
8.2.1 Fleet Management
8.2.2 Size of Vehicle
irc
8.3 Trucking Sector in India
8.4 Rail Transport
8.5 Container Operations
C
8.6 Air Transport

8.7 Water Transport
d
8.8 Transport Networks

8.8.1 Hub and Spoke Networks
ite
8.9 Storage and Warehousing

8.10 Concluding Examples and Remarks
im
References
Exercises
rL
Fo
Transportation, Storage and Warehousing 193
INTRODUCTION
M
any activities in the logis- specific technological and managerial
tical sphere can be im- complexities per se, which are beyond
y
proved if viewed from a the scope of this chapter. But we point
supply chain viewpoint. We look at out many advantages of looking at
nl
some such sectors; in particular, trans- these operations through a supply
portation, storage and warehousing. chain lens.
O
Of course, each of these sectors has
n
io
at
ul
irc
C
d
ite
im
rL
Fo
INTRODUCTION TO TRANSPORTATION
Traditionally transport operations were initiated and managed by either a seller who
wished to access new markets for sale, or a buyer looking for new sources of supply to
y
fulfill a certain need. The former was a predominant driver of freight transport
nl
activity, once technologies developed that facilitated larger volumes of production
than could be locally consumed. Over some period of time, specialist transport firms,
O
for example, trucking companies, emerged, which were hired for achieving this
spatial transfer of commodities. With international trade becoming more pervasive,
the commercial maritime sector developed into a complex multi-stage operation
n
because of the specialised technology involved, the need to transfer material across
io
modes of transport (at the land sea interface) and the commercial requirements of
inter-country trade. Ancillary players such as freight forwarders, port terminal
at
operators, customs agents, insurers and others formed part of a complex chain of
entities that moved cargo across oceans.
ul
With this complexity in operation, specialised firms that manage a range of
irc
activities around transport have emerged, called Third Party Logistics (TPL) service
providers. These have further evolved in functions, to include IT integration with
partner firms, to be labelled fourth party logistics firms, but we concentrate on
C
describing the main transport activity and the major decisions involved in this sector,
in the first part of this chapter.
d
ite
8.1 TRANSPORTATION MODE CHOICE

We now discuss transport mode choice detail from two points of view. We have
im
already seen it from a shipper’s point of view in the supply chain context in Section
6.4.4. We will see this from the point of view of asset management of the transporter,
rL
in this chapter.
In many parts of the world, transport (especially road transport) is by dedicated
vehicles, which would typically return empty to the original point. This has got some
Fo
advantages,
1. The increased reliability of the transport service, since the vehicle does not
have to wait for a potentially uncertain return load,
2. The reliability to customise the transport mode for the commodity and
thereby reduce handling losses and quality defects during transport.
In some cases, the choice of type of vehicles and mode of operation is forced
because of dedicated vehicle owned by the shipper. Another driving factor is when a
commodity requires special handling (e.g. petroleum) or makes it difficult for the
vehicles to be used for other commodities (e.g. coal using bottom discharge wagons).
y
In India, the option of using transport in a one-way mode is available in various ways
nl
1. Road, because of the presence of well-developed, decentralised transport
markets in which independent truckers participate.
O
2. Rail, where Indian Railways allow point to point booking of wagons or rakes
and manages the overall flow of empty wagons on a system-wide basis.
n
This system is also available in international sea or air, especially container
io
movement.
This is possible only with standardised vehicles. This has certain advantages
at
1. Lower cost of operation because of better vehicle utilisation
ul
2. Bigger pool of vehicles available and better contingency planning in case of
disruptions.
irc
In supply oriented or supply constrained markets, standardisation of vehicles (or
in general, any product or service) is very important for interchangeability and
C
effective offering.
d
8.1
ite
The case of Rajshree Cement in the book looks at this issue in an integrated manner.
Here, asset management of the railways (in this case locomotive) is examined from
im
the utilisation as well as the service effectiveness point of view. A similar set of
arguments applies in the case of dedicated rolling stock for certain closed circuit
rL
movements in coal mine and iron ore shifting.
8.2 TRANSPORT OPERATOR DECISIONS

Fo
The following form a set of (these are stated roughly in the order they are made, in
practice) decisions that a transport operator needs to make.
➣ Locations to serve (markets in which to operate)
➣ Network design (that is, which pairs or sets of locations to serve, and route
structure)
➣ Route capacity planning (often has a large infrastructural element, with

public, or public-private financing)
➣ Vehicle capacity and number (fleet planning)
➣ Vehicle schedules (timetabled services and provisions for on-demand or
y
charter services)
nl
➣ Crew planning (in keeping with vehicle schedules)
➣ Terminal facilities planning (can be infrastructural, and increasingly in
O
private control for dedicated needs)
➣ Maintenance plans (for vehicles and terminal facilities)
n
➣ Day to day management of operations (includes pricing of services)
io
at
8.2.1 Fleet Management
We discuss an issue that is uniquely relevant in India where cost considerations in
ul
transport are very important. Consider general purpose trucks or railway wagons.
irc
These can be used to transport a number of commodities. There is a natural tendency
to maximise their utilisation across these possibilities in order to cut costs. This leads
to empty trucks or wagons being made a part of an open system where they are either
C
assigned to, or find their way to nearby locations where they are demanded. This is in
contrast to the model of operations where dedicated vehicles would return empty to
d
the origination location.

ite
In trucking, we would have two options of fleet management; (a) dedicated fleet for
one customer, and movement to and from pre-decided locations, on an agreed upon
frequency with minor variations through the year or (b) on-demand services to a large
im
number of customers and meeting of multiple types of demands to several locations.

In barge services and ocean sea transport, we have two types of operations (a) liner
rL
services which run to a specified schedule among many ports of call and (b) tramp
services that are as per the demand of customers.
Fo
In air cargo services, most traffic is by scheduled services, but charter services could
be used, for occasional or high value cargo.
In almost all modes, the dominant mode of long distance passenger transport is the
scheduled or liner mode of operation. Charter services for small groups of specialised
customers (high end business and location specific tourism) are the only exceptions
to this. The only comparable mode to the tramp mode is the taxi cab operation in
most cities. Long distance passenger operations have an inherent symmetry of

demand in the medium run and it makes sense to concentrate on scheduled services.
Fleet management for scheduled services is easier to plan. Some issues relate to
route structuring (choosing originations, destinations and points en-route for
y
different routes) and frequency of services. This last one is a competitive element in
transport strategy as choices are often made based on the service offered.
nl
O
8.2.2 Size of Vehicle
The unit of movement in transport has a significant impact on supply chain
n
effectiveness. To some extent, this is parallel to batch size in manufacturing. Like in
io
manufacturing, there is a movement towards small batch movement, in keeping with
greater flexibility, responsiveness and just-in-time logic. Small vehicles, especially in
at
road (for example, the light commercial vehicle on Indian roads), and regional
aircraft for feeder services are examples in this direction. Mini rakes in railway
ul
operations are an example, since they are often viable in sections where line capacity is not
a constraint. One of the factors pushing systems towards larger batches (apart from
irc
lower unit cost) is the reduced set up or capacity consumption per unit of movement.
There is an opposing pressure towards scale efficient technologies in transport.
C
Examples of these are the super-tankers carrying more than 50,000 tonnes of cargo
on ocean going routes and large commercial jet aircraft. In such cases, the unit costs
d
are considerably lower and provide a significant competitive advantage. A limiting

factor could be the facilities in the entire network that can accept large vehicles (e.g.
ite
runways that accommodate large bodied aircraft).

The pros and cons of these options are well known, but need to be detailed in each
im
environment. A solution is to combine the benefits of many vehicle types (but using
efficient material transfer technologies!), such as using large vehicles for primary
rL
transport to warehouses and using smaller vehicles for secondary transport. If a

warehousing stage is beneficial from commercial reasons, this is a good compromise.
Combining large ships with barge movements into ports or into canal shore locations
Fo
is another example of this.
8.3 TRUCKING SECTOR IN INDIA

Since there are well-developed, decentralised transport markets in India, the option
exists of general purpose vehicles being hired for one way travel. This does increase
vehicle utilisation and brings down costs. However, reliability of vehicle supply is
affected as the requirement is now met by a return flow of loaded commodities
(which may not be synchronised in time) or by a market driven flow of empty
vehicles (which could occasionally have mismatches because of information delays or
competitive forces).
y
A challenge in India is to facilitate the growth of large scale national truckers, given
nl
the regulatory regime and the business environment. The cost structure favours small
O
operators, self driven vehicles or small fleet owners. As of now, large operators in turn
sub-contract operations to smaller operators, which is cost effective, but has reached
its limits in terms of tracking, flexibility, service provisioning and overall
n
effectiveness.
io
8.4 RAIL TRANSPORT
at
Freight movement by rail is through wagons or containers (which are discussed more
ul
generally in the next section). Wagons are generally formed into movement units
called rakes, which have a standard size depending on the terrain and the type of
irc
locomotive used. Each unit movement of a rake is a train that has to be scheduled and
handled through the railway system. For the last many years, Indian Railways
C
operates mainly rake loads, thereby catering to movement of bulk commodities. The
advent of containerisation has made it possible to carry mixed traffic through
aggregation into rake loads of containers by operation such as Container
d
Corporation of India and others.

ite
As discussed in Section 8.2, wagons could be general purpose or special purpose in

nature. Rail wagons for specialised commodities generally move in specified circuits
im
of operation. Special purpose wagons are usually justified when there are significant
volumes of traffic over a period of time. Such demands sometimes justify closed-
circuit operations of wagons.
rL
The other mode of movement of commodities is through general purpose wagons.

This is justified when commodities have seasonal demand and vary from location to
Fo
location at different times. Food grain is a good example of this type of movement.
Indian Railways made a strategic decision to move the large fraction of its traffic in
unit loads called rakes (collection of wagons formed as a train used over large circuits
of movement in between train inspection and other safe running formalities). This
led to considerable gains in wagon turnaround and asset utilisation. Today, wagon
loads are accepted in principle, but almost all traffic that is meaningful is offered in
rake loads. Containers form a convenient way of allowing the carrier (railways) to
carry rake loads and traffic to be conveniently built up, e.g. by Container
Corporation of India and other operators who consolidate loads and form trains.
The overall asset management from the railways point of view is in terms of track
y
and terminal capacity, locomotive management and wagon management. The overall
management of these resources assumes that subsequent resources like crew are
nl
essentially available on demand, and need not be planned at this level.
O
8.5 CONTAINER OPERATIONS
n
The container box has had a revolutionary impact in the freight transport sector and
io
today, it is impossible to think of movement of value-added commodities across
countries, without containerisation. This technology, now standardised in design
at
(especially in dimensions), has made it possible for safe carriage of goods across
modes [land (road or rail), water (inland, coastal or ocean going) and even air] in a
ul
seamless manner.
For both domestic and international operations, container movement is an
irc
increasingly viable option. In India, traditional rail containers like Concor, also
operate road container services, to improve flexibility. Other players have been
C
allowed to offer container based rail traffic to the Indian Railways and this has
thrown up some interesting issues in asset planning for these operations. In
particular, new players have had to face difficulties in procuring flat railway wagons
d
for carriage and also attain access to the rail network through appropriately situated
ite
container freight station with railway sidings. The latter issue is partly resolved
through leasing and partnership arrangements with Concor.
im
Road based container operators need a fleet of trucks that can carry containers (in
20 TEU or 40 TEU sizes—which are the standard ones). The major operational
problem then is to plan the return flow of containers. The dynamic positioning of
rL
containers is a complex problem in the face of uncertain demand. A basic choice to be

made is whether to wait at a location to meet a known or anticipated demand at a
Fo
later date or to move empty to a location where there is already committed demand.
Technically, dynamic programming is a technique for such decisions.
8.6 AIR TRANSPORT

Cargo moved by air is time sensitive, high value and needs tracking and careful
handling. A number of operational decisions have to be made in addition to the
above-mentioned transport planning decisions. We mention only two: (a) dynamic

pricing and capacity management to exploit elasticity of demand with respect to price
and service levels and (b) stacking and storage decisions on the aircraft and at the
terminal for load balancing as well as quick retrieval when required.
y
In (a), capacity is allotted to different segments of the traffic, so as to maximise
expected revenue and in (b), prices adjust themselves dynamically to perceived
nl
demand and what bookings are already done and what sort of cancellation pattern
O
there is. These are technical problems that are quite challenging, and form an
important part of ongoing research.
A supply chain issue of direct relevance to air transport decisions is that economies
n
of scale preclude the regional dispersal of air services to the extent normally required
io
and therefore, it is inevitable to have other modes to complement air transport, to
finally achieve door to door service. Providing for this supply chain requirement is a
at
major concern of new airports, new hubs and regions that plan to have air transport
infrastructure. ul
irc
8.7 WATER TRANSPORT
Water transport is cheap when it can be done, since the cost of movement is generally
C
in a natural channel (low capital cost of acquisition of right of way and construction),
and fuel efficiency is very high. From a supply chain point of view, the major
d
disadvantage is that water transport can rarely meet end-to-end needs of customers
and there is a need for a change of mode of transport at either end of the water
ite
movement. Also, water movement tends to be slower than competing modes. But the
cost advantages of handling large volumes are quite significant.
im
Large ocean liners are transport supply chain drivers as their schedules at major
ports determine the timings of all the other legs of movement. They have restricted
rL
access to ports and need supporting infrastructure to load and evacuate material, such
as barge movement, port terminals and hinterland connectivity.
A small sector in India, but a significant one in some other countries, is inland
Fo
water transport, on rivers and canals. This sector can merge with ocean going sea
traffic in a few cases, such as operations on the Great Lakes in the U.S./Canada.
Otherwise, barge operations in inland waterways are too small to be cost effective in
long distance ocean movements. See example 3 at the end of this chapter to
understand how inland waterways can form an important part of a logistics supply
chain.
8.8 TRANSPORT NETWORKS

A conceptual design issue that arises in transport is the possibility of aggregation and
dis-aggregation of traffic over the end-to-end demand satisfaction. Two extreme
structures that are used worldwide are the hub and spoke system and the direct
y
connectivity network options. These extremes and variants in between, can be found
nl
in rail, air, water and all forms of road transport.
O
8.8.1 Hub and Spoke Networks
n
The network design problem is of specific importance in the air freight sector since
the cost of assets and operations are huge and any inefficiency in planning gets
io
magnified greatly. The operational economics are as follows: larger aircraft are
at
cheaper to fly (per seat mile at full load) but are costlier to purchase as compared to
smaller aircraft. The demand dynamics are as follows: direct flights are faster and
ul
more reliable as compared to connecting flights and hence more in demand.
Consider a demand distribution for a typical network (in tons of cargo)
irc
From/To Mumbai Pune Goa Delhi Lucknow Chandigarh
C
Mumbai 0 20 30 100 10 15
Pune 20 0 15 15 0 10
Goa 30 15 0 20 5 10
d
Delhi 100 15 20 0 20 30
ite
Lucknow 10 0 5 20 0 10
Chandigarh 15 10 10 30 10 0
im
Aircraft of the hold capacities of 40 ton and 100 ton are available.
Two contrasting network designs are:
rL
The highly aggregated hub and spoke network

➣ A few airports are chosen as hub airports. Demand originating from nearby
Fo
airports is aggregated at these hub airports.

➣ This demand is then transported to their final destination (if it is a spoke of
the same hub) by small aircraft or to another hub by large aircraft.
➣ Fewer flights are required totally. For a network of n nodes only n–1 flights
are required (assuming single frequency is enough to satisfy the demand).
➣ On the hub-hub routes a large amount of traffic is generated due to

aggregation and the economies of scale offered by the larger aircraft can be
utilised.
For the transport of most forms of packaged goods this approach is adopted since
y
it provides wider network connectivity using fewer trips. Moreover, unlike
passengers, with packages it does not matter what routing they take provided they
nl
arrive on time. The downside is that since the smooth functioning of such a network
O
requires that many connections be made, in case of contingencies such as delays, etc.,
the entire operation is affected badly.
In the freight context, the hub and spoke model is characterised by the cross-
n
docking methodology adopted successfully by big retailers like Wal-Mart. There are
io
multiple suppliers (for different commodities) and multiple demand points
(individual stores). Instead of each supplier supplying separately to each store, they
at
supply the goods meant for all the stores to a centralised location, much like the hub
explained earlier. Here all the goods destined for one store (from various suppliers)
ul
are loaded on to one truck directly from the suppliers’ trucks without storage. The
timing of arrival of the suppliers’ trucks and the departure of the stores’ trucks must
irc
be well synchronised. If there are n suppliers and m stores (n+m+1 node network,
including cross-docking station), n+m truck trips are required to the central location.
C
A possible hub and spoke network for our example is shown below
d
Chandigarh
ite
Delhi
im
Lucknow
rL
Fo
Mumbai Pune
Goa
Mumbai 0 1*100 1*100 2*100 0 0

Pune 1*100 0 0 0 0 0
Goa 1*100 0 0 0 0 0
y
Delhi 2*100 0 0 0 1*100 1*100
Lucknow 0 0 5 1*100 0 0
nl
Chandigarh 0 10 10 1*100 0 0
O
The dispersed point-to-point network
➣ Each demand is met by a separate service.
n
➣ A larger number of flights have to be operated, for an n node network
io
typically n2 flights need to be operated.
at
Point-to-point networks are more suited to specialised services such as delivery of
perishable goods where time is of the essence. Direct connections offer the fastest way
ul
to get from A to B. Also the reliability of these networks is better since the operations
on various sectors are more isolated and hence problems in one sector do not
irc
transform into network wide problems. Regular high volume traffic between specific
nodes may also warrant the use of direct connections.
C
A point-to-point network for our example is shown below

d
ite
im
rL
Fo
The point-to-point solution to the sample demands shown above is as follows
Mumbai 0 1*40 1*40 1*100 1*40 1*40
y
Pune 1*40 0 1*40 1*40 0 1*40
Goa 1*40 1*40 0 1*40 1*40 1*40
nl
Delhi 1*100 1*40 1*40 0 1*40 1*40
O
Lucknow 1*40 0 1*40 1*40 0 1*40
Chandigarh 1*40 1*40 1*40 1*40 1*40 0
n
Most operators settle for a combination of the two solutions where they have hubs
io
that aggregate traffic and several focus cities that get a lot of point-to-point
connections to hubs and other focus cities.
at
8.9 STORAGE AND WAREHOUSING ul
The last decade has seen many developments in storage and warehousing
irc
technologies. One area is in information technology, to allow quick tracking of
locations, SKUs or even specific items. Another area is in the technology of storage in
C
special conditions, such as temperature control. These allow an increased time frame
of product life and therefore access to markets that could not otherwise be serviced.
The impact of this is significant in outbound agri-logistics (sea-food, fruits and
d
vegetables, flowers, meat, etc.).

ite
An example of storage technologies synergising supply chain operations is a typical

warehouse attached to an air-cargo hub, where unit loads are stored before and after
im
the flight segment of movement. Complete synchronisation of arrivals/departure of

loads and flight movements is neither possible nor even desirable, one needs locations
to de-couple these activities when required. An efficient storage system, such as a fully
rL
automated or partly automated storage facility with quick storage and retrieval is
essential. Such facilities require design of a two or even three dimensional storage
Fo
space, order picking procedures (which have a host of options ranging from
completely dedicated storage areas to randomly assigned areas as two extremes) and
technologies for item movement. The trend in these technologies is towards
traceability and responsiveness, rather than cost minimisation, per se, since the costs
of cargo not meeting demands in time, or not making transport connections would
often justify one time investments in the technology for storage and retrieval.
8.10 CONCLUDING EXAMPLES AND REMARKS

We conclude with some illustrations of how transport, storage and processing can all
be viewed as part of a common framework of value addition in a supply chain that
ultimately tries to provide location and time value of products and services to
y
customers.
nl
O
8.2
Ready Mix Cement
n
In the last few years, we see the emergence of cement concrete being moved in
io
vehicles that also perform the function of mixing in the required proportions. Such
technologies allow for flexibility of operation, since the timing of the activity can be
at
synchronised with the requirement, without having (a) facilities at the construction
ul
site and (b) stacks of material on the construction site, instead having a common
facility to serve many sites over a period of time.
irc
8.3
C
Railway Mail Service on Indian Railways

d
In earlier days, overnight trains used to carry mail in mail vans, coaches equipped
ite
with sorting facilities so that mail could be accepted till the time of train departure
and then sorted en route and ready for dis-aggregated dispatch at the destination (e.g.
sorted into locality wise bundles at the destination city/district). Over time, demand
im
has grown such that a large number of buckets (destinations) are required.
Automated sorting technologies have emerged which are used in large cities with a
rL
large volume of traffic. Finally, the opportunity cost of space on trains has become so
high, that sorting facilities on trains are now not provided and mail bags are carried as
such. This represents a change in the mix of operations, storage and transport in the
Fo
mail supply chain.

The relationship between transport and supply chain management is two way. On
the one hand, transport is a significant value addition activity in the supply chain,
leading to place utility for the customer (providing products and services at the
locations where they want them) and on the other hand, principles of supply chain
management can be applied to the transportation activity itself. A striking example is
the application of the principles of lean manufacturing (including standardisation,

waste minimisation, streamlined flow etc.) in airline operations, such as those for
Southwest airlines (one of the longest running profitable airlines in the world,
consistently).
y
nl
8.4
O
Inland Water Transport in Goa
The tidal riverine system in Goa, comprising the Mandovi and Zuari rivers, the
n
Cumberjua canal and the linkage with Mormugao and Panaji ports forms more than
io
90 per cent of the commercially viable freight inland water movement in the country.
Almost all of this traffic is exported and inland waterways in Goa form an integral
at
part of the competitiveness of the mineral ore export industry. In recent years, higher
grade iron ore from mines in Karnataka is also brought to Goa, blended with the
ul
lower grade Goa ore and moved by barge to ocean going vessels for export
destinations. Hundred per cent of this export traffic is handled by barges on the
irc
Mandovi and Zuari and although the distances involved are small, the inland water
mode of transport is vital to the functioning of the whole activity. This is from two
C
points of view: overall logistics costs and environmental acceptability.

The point to note is that the entire movement of iron ore by export is by barge in
the Goa sector, even though historically there was rail movement of iron ore right
d
upto the port on the earlier metre gauge railway system. When this was dismantled,
ite
barge movement emerged as the primary mode. Apart from the lack of a proper
unloading interface at the port, a bottleneck was the line capacity on the broad guage
im
rail line section leading to Goa. This has led to the emergence of a dedicated set of
barges and loading/unloading infrastructure at several loading points on the
Mandovi and Zuari and at Mormugao port.
rL
The analysis also illustrates and emphasises the supply chain element in transport
planning. The interfaces with other modes and an origin-destination view of the flow
Fo
are essential for a mode to be an effective part of the movement of a commodity. In

this case, the inland water mode, together with its interfacing costs, offers a cost
competitive way of moving material to the next step in the supply chain (the ocean
going vessel that carries bulk cargo for export).
Mormugao port is specifically designed to handle iron ore for export through
barge shipments in three different ways: by stacking on the ground and loading on to
ships via conveyors, by transferring ore from barges to ships anchored at points called
mooring dolphins at the port and finally, by trans-shippers or own fleet of ships
docked at anchorage in deep water locations at Panaji port or Mormugao port limits.
Lack of loads on the return trips for the barges is an area of concern since the lead
y
times are short and there is considerable cleaning, etc required before different
nl
commodities can be handled by the barges.
Figure 8.1 gives a flow chart of the iron ore movement by IWT in Goa for 2003–
O
04.
n
Figure 8.1
io
at
ul
irc
C
d
ite
im
rL
Fo
➣ All figures are in million tons (mt)

➣ Exports at Mormugao Port Trust (MPT) were 22.6 mt, at Panaji were 8.1 mt
➣ Total Goa iron ore production was 22.1 mt. All were exported. 60 per cent
were from the northern mines and 40 per cent from the southern mines
➣ About 1 mt of Karnataka ore was blended with Goan ore for exports. The
remaining 7.6 mt were exported directly, mostly through Panaji
➣ Ore from Mandovi loading points moved to MPT, both via the sea route
from the Mandovi mouth and the midway Cumbarjua canal route
y
Figure 8.2 gives a flow chart of the iron ore movement by IWT in Goa for 2002-03.
nl
O
Figure 8.2
n
io
at
ul
irc
C
d
ite
im
rL
➣ All figures are in million tons (mt)

➣ Exports at Mormugao Port Trust (MPT) were 18.7 mt, at Panaji were 5.0 mt
Fo
➣ Total Goa iron ore production was 19.0 mt. All were exported. 60 per cent
were from the northern mines and 40 per cent from the southern mines
➣ About 1.7 mt of Karnataka ore was blended with Goan ore for exports. The
remaining 3.0 mt were exported directly, mostly through Panaji
➣ Ore from Mandovi loading points moved to MPT, both via the sea route
from the Mandovi mouth and the midway Cumbarjua canal route
Bowersox, D. J., P. J. Calabro, G. D. Wagenheim, (1981), Introduction to Transportation.

New York: Macmillan Publication.
y
Coyle, J. J., E. J. Bardi, R. A. Novack, (2000), Transportation, Cincinnati: South-Western
nl
College Publication.
Gittell, Jody Hoffer, (2003), The Southwest Airlines Way: Using the Power of Relationships to
O
Achieve High Performance, McGraw-Hill Companies.
Case references : Transportation issues are dealt with in the cases Western Oil Limited (A),
n
Laxmi Transformers and Rajshree Cement. Third party logistics issues are the main
io
concern in the cases Air Freight Limited and Seth Dhaniram (C & FA). Vehicle
utilisation and fleet planning issues are relevant in Concor and Rajshree Cement.
at
Containerisation is dealt with explicitly in Concor.
ul
irc
1. Provide details for a typical (a) air cargo operator, (b) barge operator on inland
C
waterways (on a river/canal system), and (c) operator who is planning container
services by rail (refers to Section 8.2 in chapter).
2. Determine the trucking requirements for iron ore to rail siding movements of ore,
d
given the following parameters: average distance (pit head to rail depot): 20 km,
ite
average speed of trucks, 10 kmph, loading time: 12 minutes per truck, unloading
time on to wagons 10 minutes, to ground stack locations, 3 minutes. Amount carried
per truck, 12 tons. Hours of operation, 10 hours a day. Requirement: two million
im
tons a year.
3. In India, truck rates are published and available through local transport exchanges.
rL
Consider a simplified two node network (Mumbai-Delhi). The demands for loaded
movement for a particular carrier specialising in industrial products is 35 trucks a
week from Mumbai to Delhi and 30 trucks a week from Delhi to Mumbai. In a given
Fo
season, these demands can be taken as representative of the overall demand. (a) Since
transport rates are competitive, would you expect Mumbai-Delhi rates to be higher
than Delhi Mumbai rates? (b) What is the overall fleet requirement for this
transporter (you can assume four days of travel time each way, but you would have to
make additional assumption of variability of demand over time and also vehicle
maintenance requirements)?
4. From secondary sources, identify directions of movement of (a) coal from producing
regions to thermal power plants, (b) petroleum products, and (c) cement, in India.
Estimate the time taken for loading, movement and unloading for any one such cycle
and thereby estimate the pipeline inventory of the commodity and its value.
5. A Concor terminal in a city has three loading lines (sidings) and has estimated that its
y
internal requirements vary between 30 and 90 rakes a month, uniformly over a
nl
month. If loading/unloading can be done during daytime hours and take anywhere
from four to six hours, does it have enough capacity to offer other players? Design the
O
terms of a reasonable contract, in terms of service levels to a shipper interested in
utilising this terminal.
n
6. A container company wants to qualitatively understand cost trade-offs in its
operations. Assume a container costs Rs K per day to hold at some location. Extend
io
the Delhi-Mumbai trucking problem with the same data and assumptions. Compare
the options of owning a fleet of L1 versus a fleet of L2 (L2 > L1), but with less empty
at
running. What are minimum/maximum values of L1 or L2 for viable running?
7. Gather data about the economics of such an operation and compare it with the option
ul
of static facilities for storage and mixing of cement and the transport of the material
required for this.
irc
C
d
ite
im
rL
Fo
Fo
rL
im
ite
d
C
irc
ul
at
io
n
O
nl
y
y
CHAPTER
nl
9
Role of Information Technology
in SCM
O
n
CHAPTER OUTLINE
io
Introduction
at
9.1 Basic Requirements of Supply Chain Management
9.2 Elements of a Modern IT System ul
9.3 A Typical Architecture of an IT System
9.4 Elements of the Application Layer
irc
9.4.1 Transaction Engine (Core Applications)
9.4.2 Extended Application Layer
9.4.3 Integration Layer
C
9.4.4 Business Intelligence Layer

9.4.5 Portal Technologies
d
9.4.6 Dashboarding Technologies

9.5 Current Practice
ite
9.5.1 Design of Information Technology for Supply Chain

9.5.2 Adoption of Information Technology
im
9.5.3 Deployment and Usage

9.4.4 Costs
9.6 The Promise of Information Technology
rL
9.6.1 Radio Frequency Identification Technology (RFID)

9.6.2 Service-Oriented Architectures
9.6.3 Service Providers, Outsourced IT Services
Fo
Summary
Exercises
Role of Information Technology in SCM 213
INTRODUCTION*
F
rom its early beginnings as a 3. Watch for vital events and trigger
set of automated electronic alerts to appropriate people. For
y
procedures for recording example, a non-receipt or a likely
transactions and managing data, Infor- delayed receipt of an important
nl
mation Technology has become perva- scheduled shipment can be noti-
sive in all walks of an enterprise today. fied to the factory manager.
O
The term ‘Information technology’ is of
4. Provide an up-to-date status of
a recent origin. In its basic form, it
key operational parameters, for
gives the users the ability to record
example daily sales figures for an
business transactions, to review or
n
in-launch product, or machine
change them later, to save the data re-
utilisation figures.
io
lated to the transactions, to review or
change it later, and transmit the infor- 5. Analyse the data to extract com-
mation in form of printed reports or mercially useful trends or pat-
at
online enquiries. In addition to these terns that are otherwise hard to
basic tasks, it does many more things detect. For example, customer in-
for an enterprise:
1. Connect internal and external
ul teractions, product co-influence
factors in buyer behaviour, trans-
port efficiency etc.
irc
people for transaction collabora-
tion, effectively giving the people 6. Predict the trends for decision
an ability to see the same data support, for example the move-
ment of the stock or forward
C
and take decisions through online

communication. For example, the build-up of inventory at a particu-
Engineering Department may col- lar location.
laborate with the suppliers for
d
The key function of ‘Information’

new designs by sharing informa- technology will always be to manufac-
ite
tion on materials, drawings, mature and distribute ‘information’ to be

terials, processes etc. used by the businesses. The exact role
2. Enable automated workflows for of Information Technology in the Sup-
im
business processes across de- ply Chain management is determined

partments and locations. For ex- by its usability; i.e. it depends on how
ample, automated process can the supply chain managers want to use
it, and how usable they find it.
rL
be initiated for the materials to

move from the factories to the
warehouses and to the distributors.
Fo
* The authors thank Milind S. Padalkar for his extensive contribution to this chapter.
9.1 BASIC REQUIREMENTS OF SUPPLY CHAIN MANAGEMENT

To support supply chains, an IT system must perform certain basic tasks.
Two fundamental tasks supporting supply chain management are: (a) Information
exchange and reporting and (b) information alerts. At a more advanced level Supply
y
Chain Management will require: (c) data and trends analysis, (d) forecasting
nl
(e) scenario modelling, planning, decision-making.
O
Basic Requirements of a Supply Chain
n
(a) Information Exchange and Reporting
io
An organisation may have several requirements of reporting. Apart from providing
at
information of different types to different people, it will be necessary to exchange
information and data with entities outside the organisation in a timely and an
ul
effective manner. Examples include information flows between logistics service
providers, transporters, distributors, warehouses, and related third party operators.
irc
When integrating the information from the outside agencies, the issues of data
homogenisation and integrity need to be kept in mind.
C
(b) Information Alerts

d
Information alerts need to be specially designed and implemented. The alerts are
implemented on the computer systems through a set of memory-resident
ite
programmes—called resident tasks—that act as watchdogs for the transactions taking

place and the data streaming in and out of the system. For example, specific high-
im
value transactions can be detected as soon as they occur, and the alerts can be sent out
to the defined receivers.
rL
To enable information alerts, the IT systems need to be equipped with an online

data capture facility at various points of the enterprise supply chain. In many cases,
the alerts are based on data aggregations. For example, the total capacity of the trucks
Fo
reporting to a warehouse may fall short of the planned shipments. For such alerts, the
IT systems require data aggregation/consolidation programmes to be running
frequently or constantly.
Advanced Requirements of a Supply Chain
Analysis of Supply Chain Data

This includes the development of appropriate algorithms and analytical products to
y
analyse the supply chain data. These products are called ‘Analytics’. Several leading
nl
software companies such as SAP, Oracle, Cognos, SAS, Informatica etc. supply
specialised analytics products.
O
Analysing supply chain data often involves a process whereby supply chain
managers formulate hypotheses and test them on the data. Hence, the analytics
n
products have a superior capability to flexibly manipulate and view the data in
different ways. We note that choices of management in many areas would have
io
significant impact on the way data is managed and what analysis is done. For
example, an organisation changing its overhead cost allocation scheme from standard
at
costing to an activity based costing would require major changes to data handling and
analysis. ul
Forecasting
irc
Based on the analysis of the data and its aggregates, it is possible to extract transaction
patterns. These can be used to forecast key supply chain trends to support the
C
decisions and forward planning of supply chains. The forecasts can also be
represented in various forms such as tables, charts, simulated transitions from current
d
to future state etc. The graphical presentation can generally enhance the usability of
ite
the forecasts by integrating a comparative analysis of the forecasts versus present or

against other estimates.
Supply chain forecasts must assume a continuation of the present policies. If the
im
supply chain design or the policies change, the forecasts will be affected.
rL
Scenario Modelling, Planning and Decision-Making

The supply chain involves a large number of coordinated or un-coordinated actions
by many players. Any proposed changes in the supply chain policies cannot be pre-
Fo
tested in real life, because such tests would necessarily be expensive, time consuming,
geographically impractical and disruptive. Information Technology can create a
powerful scenario modelling capability to enable the supply chain managers to
understand the component behaviours and to evaluate the systemic impacts arising
from proposed policy changes. A good understanding of the interactions between
supply chain stages and the behaviour of the data allows organisations to model the
different scenarios.
Scenario modelling goes a step beyond forecasting by conducting a ‘what-if ’

simulation by changing one or more of the design rules. For example, inventory
forecasts can be created after an analysis of the demand patterns, assuming no
changes to the replenishment strategies. But the organisation may want to introduce
different replenishment strategies for different demand patterns.
y
Because changes in the supply chain design cannot be studied through physical
nl
means, a scenario modelling capability is a powerful platform for analysing changes.
O
Such a platform can also evaluate how the various applications at various levels will
work with each other.
A scenario modelling platform can also assist in planning the supply chain
n
variables from time to time, thereby acting as a key decision support system.
io
at
9.2 ELEMENTS OF A MODERN IT SYSTEM
Modern IT systems include many components such as the data, programmes, printed
ul
reports, screen images, network connections etc. ‘IT architecture’ describes how these
components are tied together. The objective of an architectural design for a given
irc
application is to ensure optimal levels of Usability, Scalability, Interoperability, and
Flexibility.
C
Most IT systems are generally designed around client-server architectures, in

which multiple users can use an application simultaneously. The applications and the
d
data reside on separate computers called ‘servers’. The servers are connected in a
ite
network with the user computers. The user accesses the system through a local PC
(also known as a desktop, or a workstation)—called as a ‘client’. The client places
specified requests on the ‘server’ which fulfills them. In their physical forms, the
im
clients and the servers are computers running programmes, and having a designated
function.
rL
In practice many users will access different applications simultaneously. Hence the
IT architectural designs keep four factors in mind. These are: (1) efficiency (2)
security (3) error-free operation, and (4) load balancing.
Fo
To ensure these objectives, the IT architecture as separate layers which communicate

with each other. Each layer communicates with the next layer in the architecture and
not with the others. The 3-Tier (also 3-layer) architectures are most common in
designing the client-server systems. For scalability, the three layers are generally kept
on separate server machines. The three tiers are described as below:
1. Presentation Layer: Also called User Interface layer. This layer includes all
programmes, display formats, objects and techniques to present the
information to the users. The information may be presented as screen images,
printed reports, messages on personal devices, etc. For example, the display of
inventory status on a desktop will be handled within the presentation layer.
y
2. Application Layer: Also called Logic layer, or Rules layer. This layer includes
nl
all programmes which perform the business processing. The extraction of the
O
data, the logic to manipulate it, and the selection of the data sub-set for the
purpose of presentation will be handled in this layer. Continuing with the
preceding example, the Application layer will access the inventory data,
n
process it according to defined rules, and forward the sub-set of the data to
io
the presentation layer.
3. Data Layer: This layer includes the physical and logical data stores along with
at
the logic to receive requests and send data to the Application layer.
ul
The middle layer may be further divided into multiple tiers—resulting into n-tier
architectures. To prevent unauthorised accesses between the layers, the architectures
irc
may include firewalls—essentially computers to allow or prevent network traffic
between the two layers.
C
Supply chain managers may need to access and analyse data coming from several
different sources. Web technologies allow many players in the supply chain to
collaborate and exchange data over the internet. The operating conditions of web
d
applications differ widely. A range of technologies, standards, protocols and software

ite
object libraries are available to support the variability in designing the web-based IT
systems. These are briefly described below:
im
Web browsers (such as Microsoft Internet Explorer©) belong to the presentation

layer. Web browsers decode and present the data sent by the application layer. Other
presentation technologies include Java Server pages (JSP) or Active Server Pages
rL
(ASP). The data is usually exchanged by using standards such as XML (eXtensible
Markup Language).
Fo
The Application Layer uses several technologies and standards. J2EE or .NET are
frameworks for connectivity to various systems. Servelets or Java Beans refer to the
programme objects to perform functions within the Application layer. The entire set
of such programmes is called Application Server.
Database server keeps the data. The Application Server exchanges data with the
database server by using standards such as JDBC or ODBC.
Depending on the application size, the application server and database servers may
be hosted on the same computer or on separate computers.
A careful mix of the technologies and standards can result into an optimal
performance of the IT system.
y
nl
9.3 A TYPICAL ARCHITECTURE OF AN IT SYSTEM
O
An IT system is described through a collection of its components. A pictorial
representation of the various IT system elements and how they interact is given
below.
n
io
at
ul
irc
Presentation
Portals, JSP, ASP, web browsers etc. layer
C
d
Initegration
layer Analytics, reporting, intelligence
ite
Application
im
Extended applications layer

rL
Transaction engine
Fo
Data layer
Transaction data
9.4 ELEMENTS OF THE APPLICATION LAYER

The application layer encompasses the core transaction engine, i.e. the rules,
programmes, algorithm logic, workflow decision triggers etc. It is the heart of any IT
system. Hence a further understanding of its elements is necessary. These are
y
described below:
nl
O
9.4.1 Transaction Engine (Core Applications)
Any IT system will need a transaction engine, i.e. a set of core applications to process
n
transactions and capture data. The transaction engine provides complete support for
all operations in the different business processes such as order-to-cash cycles, design-
io
to-manufacture processes, other supporting business processes etc. The users interact
at
with these applications in course of the regular business operations.
Depending on the type of the IT infrastructure, these applications could be of two
types:
ul
irc
9.4.1.1 Integrated Enterprise Resource Planning (ERP) Applications
Applications such as SAP or Oracle provide an integrated set of applications to
C
enterprises for all or most of the operations. Within an ERP package, there are several
smaller applications performing specific sets of functionality (such as materials
management, accounts receivables, plant maintenance etc.). However they are tightly
d
coupled, and share the same database, user interfaces and reporting standards. By
ite
‘Integrated’, we mean that the various business processes are connected, automated
and work in tandem. For example, a new sales order can automatically trigger a shop
im
order if the finished product inventory is not sufficient to fulfill the order.
‘Integrated’ also means that these applications come in as one package that cannot be
dissembled into its component applications or the data. They also cannot be used for
rL
stand-alone functionalities.
9.4.1.2 Loosely Coupled Applications

Fo
Companies often have many discrete applications that are essentially designed as
stand-alone. These applications operate on their own data bases/structures, and
interact with each other only for defined transactions. For example, a new sales order
will not automatically create a shop order or a purchase order on the suppliers, but it
will leave an indication, usually a data flag that can be read by the shop order
application when it is run. In that sense, the applications are said to be loosely
coupled. A transaction engine built on such loosely coupled applications will

generally not enable some of the features talked above, i.e. scenario modelling, trend
analysis, forecasting etc. It will also not support an on-line, real-time analysis,
information alerts and other features dependent on the quick availability of data.
y
nl
9.4.2 Extended Application Layer
The Transaction Engine serves the basic users, i.e. the people directly engaged in
O
transactions such as purchase orders, sales orders, payments etc. These transactions
create records in form of data which accumulates over a period of time. Such
n
historical data is useful for analysing the trends and understanding how the different
io
entities in the supply chain perform. For example
1. A history of material receipts from a particular supplier will indicate the quality
at
of shipments, and price/delivery performance.
2. Corporate finance may require an application for treasury management i.e.
ul
management of free cash.
3. Sales department may want a set of ad-hoc reports to link the salesman’s
irc
performance with his incentive calculations.
Such applications are not core to the business. They are required for extended
C
business purposes and operate on the derived forms of the basic transaction data.
Such applications collectively form the Extended Application Layer.
d
Generally, the Transaction Engine alone cannot meet the requirements of supply
ite
chain managers. These are met by the Extended Applications Layer. When this layer
is integrated with the Transaction Engine, it becomes effective in managing the
supply chain. It is easy to see that a loosely-coupled application structure will create
im
difficulties in the integration with the Extended Applications and will lead to a loss of
effectiveness.
rL
9.4.3 Integration Layer

Fo
Since the extended layer is actually a set of independent applications, another set of
applications is required to connect the Extended Applications to the Transaction
Engine. Collectively, such applications are called ‘Integration Layer’. The integration
layer employs several different methods for integration. These include: (a)
integration at a data level (b) integration at a message level, and (c) integration at an
application level. A detailed description of these models is out of scope of this

chapter, however for a supply chain practitioner, they can be understood as follows:
(a) Under data integration, the applications exchange only data with each other.
Generally this type of integration is simple and the quickest to set up and
y
maintain. However, it is the slowest form of integration, and does not
nl
support fast collaboration between the applications.
(b) Under message-based integration, the applications communicate with each
O
other via coded messages, typically through short data packets. The
participating applications have an in-built logic to decode the messages and
take appropriate actions. The messages are managed by programmes called
n
Message Brokers and Message-Oriented-Middlewares (MOM). IBM, SAP,
io
Oracle and Microsoft are some of the companies supplying these products.
Generally this type of integration is more expensive, and most effective in
at
supporting pair-wise application integration. They are less effective when an
application needs to communicate with several other applications using
ul
broadcast or publish-and-notify schemes.
irc
(c) Under process integration, the participating applications enter into a
common protocol to collaborate in executing a business process. This type of
integration is supported by softwares called web servers. This is the most
C
expensive form of integration to set up and manage. Process integration

becomes necessary to support the complex, real-time analytical requirements
d
in the supply chain.

ite
Apart from providing integration within an organisation, this layer also provides
an ability to exchange data from third party sources. For example, retail trade data
im
supplied by market research agencies.

rL
9.4.4 Business Intelligence Layer

Key decision makers often need specially prepared data to take complex decisions.
For example, a marketing manager may want to decide the price of a product after
Fo
looking at the cost of production, the relative brand appeal of the product, the
purchasing power of the consumer, the prices of competing products, the distributor
commissions etc. or a supply chain manager may decide the warehouse inventory and
the re-order variables based on the demand pattern, cost and time taken to transport
the product, the service levels etc.
For such decisions, the decision maker typically cannot use the basic transaction
data. He needs data computed or derived from the transaction data. He also needs to
view this data in conjunction with the other data that may come from outside the
enterprise. For conducting any analysis, he typically needs a large volume of various
kinds of data to support his decision-making needs.
y
Further, the Transaction Engine and the Extended Applications layer may not be
nl
able to fulfill some of the supports needed for decision making. To illustrate, consider
O
the following example:
A company makes shaving blades in various styles, and sells it in different markets
at different prices. It may run promotion campaigns from time to time and may
n
respond to competitor moves by price cuts or other strategies. Further, it may believe
io
that its products are preferred by a certain kind of a customer. The decision-makers
want to know whether the price set for this product is optimal for the company, and
at
whether it is able to fulfill the demand in the various markets.
ul
Such a situation cannot be handled by the Transaction Engine. The Extended
Application layer is often insufficient to address requirements stated above.
irc
Business Intelligence layer includes technologies to set up data warehouses
containing large amounts of data extracted and transformed to support special needs
C
as above. Business Intelligence also includes special programmes to analyse and view
the data in a flexible manner. Data warehouses are designed by using the concept of a
‘multi-dimensional cube’ which can support fast online analysis. A multi-
d
dimensional cube is a collection of data around a few central key elements

ite
(dimensions) such as product, customer, account, or supplier. For the above example,
consider the following data ‘cube’:
im
Pcube: {Product, geography, customer demographics, competition price and market

share, market size estimates, sales history, price realised, promotional campaigns}
rL
The way to read this: For each product, its sales volume and price history, in each
geography, split by the customer types, against the market size estimates from time-
to-time, against the competitor prices and volumes, under the promotional
Fo
campaigns launched by the company. This becomes a 9-dimensional data warehouse.

The basic transaction data is processed by Extract-Transform-Load (ETL)
programmes to shape it for fitting into the data warehouse design.
Once the warehouse is populated with all the data classified under these
dimensions, it becomes possible to analyse the product behaviour on the various
dimensions. This is called ‘slicing and dicing of the data’. The analysis is conducted
by a class of fast softwares called ‘Analytics’. Analytics use a methodology called On-
line Analytical Programming, or OLAP in short. Using these tools, it is possible to
y
address the ‘what-if ’ questions (what will be the incremental sales, if a promotional
nl
campaign is launched in a specific geography?). These techniques also allow setting
up and testing hypotheses.
O
Data warehouse can be used to carry out trend analysis and scenario modelling
usually required in operating a supply chain. Hence it becomes a useful part of the
n
infrastructure to validate the health of the supply chain from time to time.
io
Third-party data (such as retail and consumer spending data or demographics
data supplied by market research firms like AC Nielsen) is available at a price—
at
however, most companies will not be able to use it unless they have an ability to
integrate it into their systems and data architecture. Data warehouse can be a place to
ul
accommodate and harmonise such third party data.
Data warehouses are expensive to set up and operate. Hence most mid-sized firms
irc
do not have them. For such firms, object-oriented technologies could provide an
answer. These technologies are able to encapsulate the third-party data in its native
C
form with an agent software that can answer specific pre-defined queries to the
querying application.
d
ite
9.4.5 Portal Technologies

An enterprise will have many users, decision-makers or trading partners who will be
im
marginal in their IT usage. These marginal users do not need all parts of data or the
IT system functionality. For example, the Corporate Treasury Manager may only
access the system to compute the cash flows and surpluses. A supplier may only want
rL
to know the status of his deliveries and payments.

Enterprise information portals are frameworks for integrating information from
Fo
multiple sources and personalising it for the specific user. Portals give the users a
single point of entry into the IT systems—usually through a web-based user
interface. The user accesses information from various systems through specific
‘portlets’ (programmes to extract and display the data). The main purpose of the
portals is to access and view the information, hence the processing ability is limited.
IBM, Oracle, SAP, Vignette, BEA are some of the companies providing portal
technologies.
With portals, the marginal users get a controlled but limited and secure access to
the IT system. The marginal users can be within the same organisation or outside the
organisation. They can be located anywhere, and will be able to access the system
simply via internet. Portal technologies also massage the information, so that
different people are able to look at it differently. For example, a materials manager
y
and a distribution manager may look at the same data in different ways, as per their
nl
needs.
O
9.4.6 Dashboarding Technologies
n
Analytics is often used to compare two or more sets of data to understand underlying
io
phenomena or trends. Presentation of such comparisons in graphical form is useful.
Dashboarding technologies allow painting, charting and frequent refreshing of the
at
vital information and making it visible to the various users.
9.5 CURRENT PRACTICE

ul
irc
9.5.1 Design of Information Technology for Supply Chain
C
The key function of Information technology is to manufacture and distribute

information to be used by the businesses. Many companies are unable to use
d
Information Technology effectively for managing their supply chains. There are
ite
several hurdles to be overcome before IT can be made effective. Leading global

corporations such as Hewlett Packard, Procter and Gamble or Alpha Laval have
implemented several strategies to address this area. For using IT effectively, the
im
enterprises need to take a strategic view of supply chain management, and go beyond
the technology into several organisational areas. Such a view will include four factors
rL
given below:
(a) The Role of Supply Chain Managers—From Operators to Change

Fo
Managers
Supply Chain managers must play a principal role in determining exactly what
‘information’ is to be ‘manufactured’ for their use. The pace of change in the product
innovation and its supply chain environment has accelerated considerably over the
last two decades. Accordingly, today’s supply chain managers are required to be
change-oriented (experimental) in the way they manage the supply chain. While
being experimental, they must be careful that each new initiative creates an
incremental value for the enterprise. All such initiatives will require accurate and
timely data from the supply chain operations and an ability to perform data analysis.
The Information Technology today has the capacity to provide such data and its
y
analysis. If the supply chain managers have an insufficient understanding of what IT
can provide, it becomes difficult for them to visualise how they can—given the right
nl
information—improve the supply chain processes and their control over them.
O
Hence, in addition to being change-oriented, they need to be familiar with what the
Information Technology can do.
n
(b) The Role of IT Organisation—From Service Providers to
Partners/Consultants
io
IT organisation accordingly must accept the burden of educating the supply chain
at
managers about how and where the information technology can create an advantage.
Business users have traditionally viewed the IT organisation as slow and change-
ul
resistant. This is partly due to the fact that the IT has been historically structured as a
corporate department providing information to the corporate users.
irc
As the supply chain managers get experimentally active, the IT organisation will
need to get more business-oriented, consultative and initiatives-driven. They will
C
need to view themselves as engaged in a proactive partnership with the businesses

rather than providing a service-on-demand. They should develop an improved level
d
of understanding of the business environment faced by the functional users such as

supply chain managers. They should also seek to obtain a greater access and leverage
ite
to the business users, many of whom will be situated away from the corporate
headquarters. Leading global companies like General Electric or Procter and Gamble
im
have implemented the role of a full-time IT representative to the business, to support

the IT:Business partnership.
rL
(c) Investments in IT Infrastructure
Need for Building IT Assets

Fo
A robust IT infrastructure is necessary to take advantage of the features commented

in the beginning of this section. Corporate executives may not completely
understand that such investment can be rather large. The case for such capital
investment needs to be made by the supply chain managers working in partnership
with the IT organisation. However, once in place, the advantages of such an IT
infrastructure are significant and not easily obtainable by any other means. For
example, a capability to perform trend analysis will require the following features in
IT systems:
(a) Well designed data model with multiple relevant dimensions (Enterprise
Data Dictionary)
y
(b) Capability to synchronise and harmonise data from outside the enterprise
nl
(Data Management and Synchronisation)
(c) A single enterprise-wide business rules engine having an ability to capture
O
and store transactions and store high quality data in large volumes
(Transaction System, Application or ERP)
n
(d) An organisation of such data into a form that supports powerful analysis
io
(Data Warehouse or Data Mart)
(e) A capability for rapid search, pattern recognition, analysis and visual
at
presentation of findings (Search Algorithms, OLAP procedures,
Dashboarding/charting tools) ul
(f ) Ability by decision makers who may be located anywhere, to have an
irc
integrated view of selected data and conduct ad-hoc comparative analysis
based on their functional responsibilities (web-based Portals)
C
Enterprises not having such a backbone will be unable to analyse or predict the
trends in their businesses in any scientific manner.
d
Such an investment is perhaps not needed for smaller manufacturing businesses,

since they may find equal utility in ad-hoc spreadsheet-based analyses of data.
ite
However, as these enterprises grow, the need for a well-designed infrastructure will
grow rapidly.
im
Limitations of Existing IT Systems

The design of the existing IT systems may impose restrictions on the ability to meet
rL
the demands of the supply chain managers. Extra investment will be required to
remove such limitations. These restrictions principally arise from the departmental
Fo
design of the IT systems. This is explained below:
Departmental Nature of Application Design

The core applications are generally structured in departmental terms, i.e. each
application supports business processes used within a department. For example, the
warehouse applications will only process all transactions for a warehouse; inventory
applications will be concerned only with the movement and storage of the materials
in the inventory, etc.
In this architecture, any single business process that spans multiple departments
will be supported by a sequence of transactions occurring in multiple discrete
y
applications responsible for the departments involved. Such a sequence cannot be
coordinated or automated unless the applications are integrated.
nl
A departmental design creates three major hurdles in designing the IT systems for
O
supply chain.
(i) Lack of Orchestration Ability
n
By definition, supply chain must span multiple departments acting in concert for the
io
various business processes. When the applications are loosely coupled, the supply
chain process orchestration will be disjoint and poorly controlled. For example, a
at
goods receipt note in a warehouse receiving application will not automatically be
considered for a picklist operation, unless the receiving, the order matching, and the
ul
cargo planning functionalities are orchestrated across the various applications.
Careful attention needs to be given to ensure that such applications are able to work
irc
in a coordinated manner, if they are to play a role in the supply chain management.
C
(ii) Lack of a Uniform Data Model

Since a supply chain spans multiple departments, it is necessary to have a
standardised data nomenclature across the applications. If the data model differs
d
across the applications, it becomes hard to manage the supply chain. A heterogeneous
ite
data model reduces the manager’s ability to compare, control and validate the health
of the supply chain. For example, the Engineering Bill of Materials (BOM) may
im
differ from Production BOM in many companies. It is not a problem for the
respective department to manage their work—but it becomes difficult to manage the
extended supply chain. To address this problem, IT organisation needs to employ
rL
methodologies and tools such as Master Data Management to harmonise the data
model and ensure a uniform treatment of the data.
Fo
The supply chain management requires a fair amount of third-party data, i.e. data
which is neither owned nor controlled by the enterprise. Getting this data from
outside trading partners is difficult because the partner will generally resist sharing
this information. This problem is out of scope for IT, and requires negotiations and
trust-building initiatives. However, even where such data is available (through
Vendor-managed-inventory initiatives with retail stores; or EDI-based collaboration
with suppliers), such data is not in a form or definition consistent with the enterprise
data model. Before such data can be added to the enterprise database—for further
analysis by the supply chain manager—it needs to be harmonised and synchronised
with the enterprise data model. Portal technologies or object-oriented technologies
could partially address the latter problem. However, doing this requires special
y
technology skills not easily available to the IT organisation in a manufacturing
nl
company.
O
(iii) Problem of Absent Data
Under the departmental structure of applications, the data controlled by each
n
application is the one directly required by the specific transactions in that
department. Supply chain managers are often concerned with the data relating to the
io
overall process. Such data does not fall within any single departmental boundary.
at
This creates an important limitation for managing supply chains. The data required
to manage the supply chain does not exist in any application, and must be computed
ul
from multiple data points that may be controlled/owned by different applications.
Thus, a new set of programmes and data structures become necessary to address the
irc
needs of supply chain managers. Such programmes generally do not exist in the
Transaction Engine—especially the loosely coupled core applications. The following
example will illustrate this problem.
C
A perishable product is made in a factory, dispatched to a warehouse, and from

there it is dispatched to the retail store. In this movement, the finished goods
d
inventory application will create a dispatch note listing the item code, quantity
ite
dispatched and the date (optionally, time) of dispatch. The warehouse receiving
application will record the receipt of the item, the quantity and the date (optionally,
time) of receipt and will allocate the shelf location. The logistics application will
im
match the inventory to a sales order, and generate a pick list noting the item code,
quantity, date (time), carton/crate number and the truck details. On arrival the retail
rL
store will accept the shipment, sign the warehouse dispatch note and proceed to make
appropriate entries into its own store applications.
Fo
Suppose, because of the deteriorating road conditions, the supply chain manager
of the product is now concerned with the time-to-dispatch and wants to keep it to a
minimum. It will be clear that such time-to-dispatch has not been captured by any
application in this sequence. It is not known when the item was actually loaded on
the truck and when the truck actually reached the customer. The time of customer
receipt may be captured in the retail store application, but that data is no longer
available to the supply chain manager, since the store is not controlled by the
company. It could have been captured by the transporter who delivered the item to
the store, but many transporters have no access to online IT applications, and instead
use manual, paper-based transactions. Hence they may simply not have this data.
There is no requirement that the transporters return the paper logs to the company.
In summary, the supply chain manager has no way of knowing the actual time it has
y
taken to dispatch the material, and its variability. Depending on the exact design of
nl
the systems, it may not even be possible to compute this value.
O
The example given above is not hypothetical. Similar limitations exist if the supply
chain manager wants to study other control variables that span multiple supply chain
stages such as transport cost/unit.
n
To overcome this limitation, an extended supply chain application with a separate
io
data structure must be designed on the top of the Transaction Engine.
at
(d) Organisation Culture and Structure—Enabling New Ideas,
Methods and Technologies ul
The power of IT depends on the hands in which it is placed. Many manufacturing
companies may persist with their view of IT as merely a record-keeping system. This
irc
may have worked well when they were smaller. But with continuing growth, the
operations cannot be managed without increasing sophistication of the IT systems.
C
As the enterprises implement a strong and active supply chain function, they need to
create an IT organisation working in consultative partnership, and invest into a
d
robust IT infrastructure. The organisational culture becomes important to sustain

these initiatives and to continue to create an advantage. The top management could
ite
implement several initiatives to ensure that the supply chain management becomes a
powerful instrument of competitive advantage.
im
➣ Supply chain organisation structure in which strategy formulation/

implementation is recognised as a key responsibility, different from
rL
operations responsibility.
➣ IT organisation structure should recognise architecture strategy and design as
a key responsibility, different from IT operations
Fo
➣ Selecting and staffing right individuals for the roles

➣ Implement the Key Result Areas (KRAs) and the review processes for the key
individuals
➣ Ensure accurate information for the performance management of the key
individuals
➣ Key supply chain leaders and individuals should report to higher levels of
organisation to ensure visibility and support for the supply chain initiatives
from the top management.
y
9.5.2 Adoption of Information Technology
nl
In general, for any IT system to succeed, it must be seen as an organisation-wide
change programme and not merely as a technology implementation project. IT
O
system is a replica of the various organisational processes, operations and behaviours.
In the context of supply chain management, such processes are orchestrated across
n
the departmental boundaries. A mature enterprise experiences many levels of
io
disagreements and non-standard processes between its departments. Such lack of
standardisation leads to value destruction for the company. Information Technology
at
provides a platform for the executives to obtain consensus towards a standardised set
of processes. Defining and implementing such processes becomes the core of a
ul
change management programme, with Information Technology providing a
backdrop.
irc
The key elements of a successful implementation are: (a) educated users, (b)
proactive IT organisation acting as change facilitators, (c) robust IT assets, and (d)
C
top executives playing a positive role as change managers. The leadership of the
respective functions needs to engage in positive negotiations, and employ initiatives
to promote a consensus-building environment.
d
Choice of technology is crucial for the implementability. Technology comes in

ite
different forms/features at widely varying costs. The difference in cost to support a

feature can be staggeringly high. For example, online, real-time reporting can cost as
im
much as 20X–50X more than if the same reports are acceptable with a 24-hour delay.
With constant evolution of the computing technology, newer technologies may be
very expensive to implement, even if the cost of acquisition is low. When choosing a
rL
technology, due attention must be paid to the features, utility, usability,

implementability, maintainability and cost. It is useful to evaluate the technologies as
Fo
to their life cycle costs rather than one-time purchase costs.

Due to lack of in-house skills, it often becomes necessary to engage third-party
service providers to implement the information technology. Such engagements need
to be managed carefully especially if their work involves technology or architecture
decisions.
9.5.3 Deployment and Usage

New release of Information Technology generally changes the operating environment
for the users, and is therefore disruptive. The training (or re-training) needs of the
users are almost always underestimated and under-budgeted. Whenever a version of
y
IT is released for use, it causes a spike in demand for support for which the IT
nl
organisation needs to prepare itself. Increasingly, such releases affect a large number
of users who are spread out geographically—sometimes internationally. Web-based
O
training programmes can ease the cost and effort of preparing the users for a new release.
Once the users become familiar with the new operating features, the initial spike
n
for support dies down. However, with growing familiarity, the users start expecting
more features and functionality. Continued effective usage is possible if the IT
io
organisation provides support for such demands.
at
9.5.4 Costs ul
Information Technology is not cheap. Whether used well or not, it costs about the
irc
same to set up and operate. Since its usability is largely governed by factors outside
the immediate technology domain, the high-costs-poor-usability syndrome often
gets it low marks in user popularity. IT costs come from (a) design effort
C
(b) technology acquisition, (c) maintenance of the IT infrastructure and (d) training
costs. IT organisations must share the responsibility to educate the company
d
executives about the costs and to manage the user expectations.

ite
In the context of supply chains, the benefits of IT will come from an overall
improvement in the supply chain efficiency. However, most enterprises lack the
mechanism of computing the cost benefits and linking them to the improved
im
Information Technology. Without such benefit estimates, the investment costs may
appear large and the proposals may be turned down by the corporate executives. A
rL
project costing approach with inclusion of long-term benefits from improvements in

supply chain may be a solution to address this issue.
Fo
9.6 THE PROMISE OF INFORMATION TECHNOLOGY

The Information Technology continues to evolve by adding more powerful
technologies, and applications. These can have a significant impact on the design and
the operation of the supply chains. We will discuss three such trends:
9.6.1 Radio Frequency Identification Technology (RFID)

A material in the supply chain must have a unique identification to ensure control
over its custody and movement. A common way to do this is to fix a barcode label on
the material listing its item code or SKU code. Barcodes are black-and-white vertical
y
strips printed on label, which represent the item code using some industry-standard
nl
format. Barcodes can be read by hand-held optical scanners which convert the
scanned image into the alphanumeric digital item code. The scanning operation
O
eliminates the need to enter data manually and speeds up the operation significantly.
Barcode labels are very cheap and can be printed on a standard printer.
n
Figure 9.1: Example of a Barcode Label
io
at
ul
irc
0 123456 789005
However, the scanning of the barcodes needs human operators. Hence it cannot be
C
speeded up beyond a certain limit. The number of human operators is directly

proportional to the material volumes to be handled. Hence it is difficult to handle
sudden fluctuations in material volumes.
d
Radio Frequency Identification (RFID) is a new technology to scan the materials

ite
without any human intervention. The technology is built around a small, paper-thin
electronic microchip and antenna circuit called RFID tag. When exposed to specific
im
Figure 9.2: Example of RFID Tags Showing the Microchip at the Centre
rL
Fo
radio frequencies, the circuit in the tag transmits a unique digital code identifying its
presence. This code is read by radio receivers which transmit it digitally to a computer
system. RFID overcomes the limitations of barcode technology listed above.
RFID can have a material impact on the supply chain, since it can record the
movement of materials without any human intervention. The data enters the systems
y
automatically as soon as the material movement occurs. This has many advantages:
nl
1. Information about the movement of materials is available real-time, and not
O
delayed as is the case with most IT systems.
2. Theft of sensitive/expensive materials is controlled, thereby reducing the
n
shrinkage cost.
io
3. Improved inventory accuracy leads to more efficient supply chain policies.
RFID tags come in three types:
at
1. Passive RFID tags do not have own power source, but use the induced power
ul
from the scanning radio wave to transmit the identification code. Passive tags
have the shortest transmission range and need to be close to the scanning
irc
reader. Depending on the radio frequency used, the effective range varies
from 10 cm to a few metres. Passive tags may cost in the range of US$0.07–
US$0.15.
C
2. Active tags have an in-built battery source that powers the microchip and also
the transmitter. This allows them to transmit over much longer ranges–up to
d
500 metres. Active tags can cost US$2.00 or more.

ite
3. Semi-passive (battery-assisted) tags use in-built power for the microchip, but
do not have a transmitter. The signal transmission works like a passive tag.
im
Semi-passive tags are faster and give higher reader accuracy. Semi-passive tags
cost around US$1.00 to US$1.50.
Barcode costs are negligible—around US$0.002. In contrast, RFID tags cost
rL
considerably more than the barcodes. The tag costs are expected to decline as the
technology matures. Even so, the industry analysts do not expect the costs to fall
Fo
below US$0.05. Supply chain managers should conduct a careful cost-benefit

analysis of the technology before they consider its adoption.
9.6.2 Service-Oriented Architectures

Service Oriented Architecture – SOA in short, is a way of designing the IT systems so
that they operate as a collection of services. Each service consists of its own
programmes, code, data, workflows and presentation layers and performs a specific
set of tasks upon request. The requestor can be a person or another service. The
services can exchange data amongst themselves, or execute a common process in a
coordinated manner.
y
By its definition, SOA is independent of the underlying technologies. Hence, IT
systems designed under different technologies can work with each other
nl
(interoperability) as long as they are designed using the SOA principles.
O
This feature is most useful from the supply chain manager’s point of view. There
will be several independent or loosely-coupled players in any given supply chain.
Each of these players may be using its own IT systems and data. For a supply chain
n
manager, many initiatives may span the supply chain and may require a coordination
io
of the different IT systems for exchanging data or for process execution. If these IT
systems are architected as silo-designs, then coordinating them becomes an
at
extraordinarily complex task. A re-design of the IT systems using SOA framework
can greatly assist the supply chain management.
ul
irc
9.6.3 Service Providers, Outsourced IT Services
Over the past five to six years, a set of independent service providers have emerged in
C
the field of IT. These providers are increasingly providing several types of services
which can be used for addressing the supply chain management requirements that
d
cannot be met by the internal IT departments. Examples of such services include:

ite
1. Data Collection, Cleaning, Archiving

2. Data Preparation, Analytics, Presentation
im
3. Application Service providers offering specialty applications on a pay-per-use

basis
4. Turn-key IT systems installation and operation—typically for 3PLs.
rL
Fo
In summary, Information Technology offers compelling application in the field of supply

chain management. In the hands of a thinking and experimental supply chain manager, it
can become an instrument of unique competitive advantage for the enterprise. IT enables
many functional features that cannot be obtained by any other means. However, there are
several considerations before IT can play an effective role in supply chain management.
As organisations start focusing their attention on managing extended supply chains, and
as the Information Technology continues to evolve, the scope for using IT in supply chain
will continue to widen.
Case references : An example of IT issues in the context of the retailing end of supply chains
y
is found in the analysis of the case of Food World (B).
nl
O
n
1. What are the basic information requirements of Supply Chain Management? Give
examples for a maker of consumer products.
io
2. What are the advanced information requirements of Supply Chain Management?
at
Why are these called advanced requirements? How is the supply chain requiring
advanced requirements different from the one that is satisfied with the basic
requirements? Discuss in the context of the consumer products maker.
ul
3. What are the elements of a modern IT system? How does the 3-tier system design
irc
contribute to a more effective operation of the supply chain? Which goals of the
supply chain management will be harder to achieve without such a design?
4. What is the utility of the Business Intelligence Layer and Portal Technologies? Discuss
C
giving concrete examples in the context of the maker of consumer products.

5. How should the supply chain manager augment his role to make the most effective
d
use of IT? Contrast the goals and methods employed by a traditional operations
manager vs. the supply chain manager. What changes may be required in terms of
ite
complementary support from IT organisations and the top management?

6. What would be the elements and features of an IT backbone that is ‘ready-for-supply-
im
chain’? What limitations may be expected in implementing such a backbone?

7. What hurdles may be expected in deploying and using a ‘ready-for-Supply-Chain’ IT
rL
backbone? What methods would you suggest in addressing such hurdles?

8. How would you estimate the total cost of implementing the changes to an IT
backbone? What cost elements would you include? How would you estimate them?
Fo
How would you assess the benefits flowing from such an improved IT backbone?
9. What are some of the new technologies and developments in Information
Technology? How would you assess their utility in the context of the goals of supply
chain?
y
CHAPTER
10
nl
Quantitative Tools for Supply
Chain Management
O
n
CHAPTER OUTLINE
io
Introduction 10.3.3 Solution Techniques for LP
at
10.1 Forecasting and the Role of Sensitivity
Analysis and Duality
10.1.1 Method of Moving Averages ul 10.3.4 Integer Constrained LP
10.1.2 Method of Exponential
Smoothing 10.4 Routing Models
irc
10.1.3 Trend 10.5 Pricing Decisions
10.1.4 Seasonality 10.5.1 Transfer Prices and Supply
Chain Entities
10.1.5 Regression
C
10.5.2 Double Marginalisation

10.1.6 Forecast Error
10.5.3 Elasticity
10.1.7 Forecasting of Multiple Products
10.5.4 Pricing Contracts in SCM
d
10.1.8 Point Estimates and Distributions

References
10.1.9 Qualitative Aspects of Forecasts
ite
Exercises
10.1.10 Collaborative Planning,
Forecasting and Replenishment
(CPFR)
im
10.2 Management of Inventories in

Supply Chains
10.2.1 Inventory Planning
rL
10.2.2 Inventory Control

10.2.2.1 Cost models
10.2.2.2 Service-level Models
Fo
10.2.3 Positioning of Inventory

10.2.4 Management of Perishable
Inventory
10.2.5 Management of Obsolescence
10.3 Linear Programming or Linear Optimisation
10.3.1 Network Flow Models
10.3.2 General Linear Programming
Quantitative Tools for Supply Chain Management 237
INTRODUCTION
I
s Supply Chain Management the supply chain, and where this inven-
(SCM) an art or a science? Arg- tory should be strategically located.
y
uably, since it is still an evolv- In this chapter, we present tools for
ing field, SCM can be viewed as more
nl
helping managers (a) match demand
of an art than a science. However, a with supply (Forecasting), (b) deter-
number of tools are available to help mine where to hold inventory and the
O
the manager arrive at a more scientific quantity to be held (Management of In-
basis for decision-making, many of ventories in the Supply Chain), (c) de-
them drawn from the Operations Re- termine production quantities and how
search/Management Science (OR/MS)
n
to transport them (Linear Programming
discipline. In this chapter we present and Network Models), (d) evaluate
io
some quantitative tools that find ready pricing decisions.
application in SCM.
These tools are well studied in the
at
Consider first the issue of matching academic world and professional prac-
supply with demand. In the ideal world, tice, and can be used effectively in
there would be no variation in cus- many situations. This chapter cannot
ul
tomer demand, the production pro- do justice to all of them in detail, but
cesses would be extremely reliable, we do provide an orientation to classify
irc
and products could be delivered as models and decide on the scope of
soon as they are demanded. Since this their applicability. To use them to full
is a situation that will almost never be effect would require further study of
C
realised in practice, organisations in the associated technique. In all cases,

the supply chain will have to estimate software support is available in a vari-
what their customers are demanding ety of combinations, and part of the
and determine how this demand can
d
supply chain manager’s function may

be met. A related decision is con- be to judiciously plan and use such
ite
cerned with where to source the raw software solutions. We provide a num-
material or components required so as ber of small exercises which will help in
to meet demand and how to arrive at appreciating the scope of the model-
fair pricing contracts. The manager of
im
ling techniques that we present and

the supply chain will also have to de- the likely data requirements and level
termine how much inventory to carry in of detail that the model would require.
rL
Fo
10.1 FORECASTING
As noted earlier, if lead times in operations were zero, then there would be no need to
forecast demand. This points to a very important direction in supply chain
improvement, that of reducing the lead time. Apart from obvious benefits like
y
reducing pipeline inventory (lower lead times mean shorter pipelines and so less
nl
inventory and working capital to attain the same throughput), there are less apparent,
but equally important benefits in the area of reducing uncertainties. In other words,
O
the smaller the lead time, the smaller is the need for safety stock to achieve given
service levels. One way of explaining this would be that smaller lead times lead to
n
smaller forecasting errors and so less safety stock requirements. That said, since
supply chain operations will certainly involve some lead time, it is necessary to
io
forecast various quantities, especially demand. Forecasting also plays a key role in the
sales and operations planning process.
at
We consider the forecasting process for a single product, although we will
ul
subsequently point to some issues in forecasting simultaneously for multiple, related
products. For a single item, the standard forecasting framework represents time in
irc
discrete periods (e.g. week 1, week 2, etc.) and a time-indexed series of demands is
maintained. This allows us to extrapolate demands in future time periods as per the
requirement. The model is as follows:
C
Let X1, X2, …, Xt be the past demands up to time period t, which is the current
time. We shall use these values to generate forecasts for time periods t+1,…, t+k.
d
There are two standard ways of estimating the base value, given past data that is
ite
subject to some randomness. These are (a) the moving average model and (b) the
exponential smoothing model. With these, the trend, if any, and seasonality, if any, is
im
superimposed, to get a point forecast for future time periods. The intention is to
generate a forecast, Z, for the next time period, given the Xi values.
rL
10.1.1 Method of Moving Averages

In the moving average model, a parameter m is chosen and the base value is estimated
Fo
as Z’ = (1/m) [Xt + Xt–1+ ……+ Xt–m+1]. The value is thus taken to be the average of
the previous m values. The set of values considered therefore is a moving bunch of m
values over which the average is taken, hence the name. We see that fluctuations in
the data in the distant past are eventually immaterial. The less the value of m, the
more weightage we give to recent data, i.e. we believe that the demand data is more
volatile.
10.1.2 Method of Exponential Smoothing

In this method, a parameter a is chosen between zero and 1 and the updated forecast
for the base value is a weighted mean of the previous forecast and the most recent value.
i.e. Zt+1 = a Zt + (1– a) Xt. Applying this recursively, we get Zt+1 = a Zt + (1– a) Z t– 1
y
+ (1– a)2Xt–1 … so each term in the time series is weighted by an exponent of the
nl
term a (the smoothing constant). This explains the term exponential smoothing.
O
10.1.3 Trend
The consideration of trend in the demand data is easily incorporated. For example, a
n
linear grown rate is accounted for by making the forecasts Zt+k = Zt + ka, where a is
io
the growth rate. This growth rate itself is estimated from the past data (e.g. the most
recent estimate of the growth rate is Xt – Xt–1 or we may wish to take the smoothed
at
value of this growth rate over the past several time periods), by using the moving
average logic or the exponential smoothing logic. The latter method, using double
ul
exponential smoothing, to estimate the trend as well as the base value, is called Holt’s
method.
irc
10.1.4 Seasonality
C
If we believe that the demand data exhibits seasonality (e.g. daily data exhibits
d
seasonality with respect to day of the week, or monthly data exhibits seasonality over
the year), we can bring in seasonality factors quite easily. The full blown version of
ite
this, where exponential smoothing is used to estimate seasonality factors as well, is

called Winter’s method. For this, we require at least three full seasons’ worth of data.
im
10.1.5 Regression
rL
Another approach to forecasting is to use regression, a tool commonly used in

economics and statistics. To apply regression in a forecasting context, time is assumed
Fo
to be an independent variable, and demand is assumed to be the dependent variable.

The dependence of demand over time is typically presented as a linear equation:
Dt = a + bt,
where a and b are constants to be determined from the given data. The usual way is
to use the least squares criterion for determining the parameters a and b.
Regression models are applicable if we have other data that can be assumed to
(largely) influence what we are interested in. For example, we may believe that tractor
sales are influenced by agricultural prices in the previous season. A regression model
with prices in the previous season as the independent variable and tractor sales as the
dependent variable can be calibrated and used.
y
The R2 measure of fit of a regression model tells us how accurate the model is likely
nl
to be.
O
Although the terms dependent and independent variables are used, actually there
is no restriction of causality and it could just be that the two sets of data are related in
some way that is accurate enough for our purpose. The model is easily extended
n
where the quantity of interest ‘depends’ on not one, but several independent
io
variables. With an assumption on the nature of influence (e.g. linear), the
appropriate co-efficients can be calibrated using previously gathered data.
at
Returning to time series models, we note that if the data is inherently volatile, then
ul
forecast error is bound to be high. All we can ensure through our techniques is that
the forecast error is in some way not too far from the ‘true’ error.
irc
If data are auto-correlated (i.e. depend on previous values in the series), apart from
depending on time, there are additional methods that can be used. Box and Jenkins
C
ARIMA (Autoregressive Integrated Moving Average) model is an example of such a

model. We note that the term Moving Average here refers to a more subtle notion
than in Section 10.1.1 and the reader is referred to a text in Statistics or Time series
d
analysis.
ite
10.1.6 Forecast Error

im
Measures for determining forecast errors (such as Mean Absolute Deviation) are
useful to estimate the efficacy of forecasts and the resulting contingency plans. A
rL
truism is that forecasts are always wrong, but it is then important to estimate how
wrong they have been in the past (when using a given technique) and perhaps refine
the forecast technique. It is also true that at least some inventories in supply chains
Fo
are a consequence of not having information (as a ‘substitute’ for inventories) and to
that extent, forecast errors capture the inventory related costs of inaccurate
information.
We remark that by and large, forecast errors are about as much as the inherent
volatility in demand (they can certainly be no better than the demand variability and
a good forecasting method will see to it that they are not much worse than that). So
the only good way to reduce forecast errors is to manage demand variability itself.
This can be done, to some extent by customer relations, information sharing,
product design, and other marketing initiatives, some of which are beyond the scope
of this chapter. From a supply chain point of view, information sharing upstream can
reduce variability to some extent. If customers are aware of the accurate position of
y
dispatches and production quantities, they can plan purchases and will not react in a
nl
volatile manner (resorting to contingency buying, for example). This can actually be
a valuable smoothing impact on supply chains.
O
10.1.7 Forecasting of Multiple Products
n
io
We briefly look at some issues in forecasting of multiple products. Consider, for
example, an automobile dealer, who deals with several different variants of a certain
at
model. Forecasting the demand of each of these variants may make very little sense
except for a few of the models of the bread and butter variety. Many models may sell
ul
in ones and twos over several time periods, depending on the tastes of customers.
Rather than forecasting such demand, a demand sensing procedure based on a
irc
behavioural model of the customer’s preference may be more useful.
An example of a behavioural model is the multinomial logit model.
C
10.1.8 Point Estimates and Distributions

d
A single estimate of demand may be enough for aggregate planning of quantities

ite
ahead of time, or for capacity planning. But for inventory planning, we need an
estimate of the variability in demand as well. This is most usually done in the
im
stationary case, using data over a period where the underlying distribution of the
demand is the same. Stationarity actually means more than this, but this is enough
for our purpose. We emphasise that stationary demand does not mean constant
rL
demand, but rather that the demand distribution (probabilistic behaviour) remains
the same from one time period to the next. In this case, we can use statistical
techniques of fitting distributions to map the past data to derive a demand
Fo
distribution. The simplest way is to de-seasonalise and remove the effects of trend and
then plot a histogram of the demand data. This will give a variability which we can
capture using a discrete (or sometimes continuous) distribution.
Sometimes, we have reason to believe that the demand data is normally distributed
(e.g. when it arises as the sum of several independent and identically distributed
choices of various individuals, when by the Central Limit Theorem, the resulting sum
has the well known bell shaped distribution). In this case, all we need to do is to
estimate the mean and variance of this distribution and the entire probabilistic
behaviour is then available for further decision making (e.g. safety stocks). This is
because the normal distribution is fully characterised by two parameters (usually the
y
mean and variance, which is nothing but the square of the standard deviation).
nl
We remark that the mean and variance exist for all meaningful distributions. For
O
example, a Poisson distribution (capturing completely random occurrences drawn
from a large pool) is characterised by a single parameter l and the mean and standard
deviation are both equal to l.
n
io
10.1.9 Qualitative Aspects of Forecasts
at
In business, the qualitative understanding of the market is as important as the formal
techniques for analysing demand. Competitive elements are important, qualitative
ul
shifts in customer preferences can be estimated well before a model can detect it and
finally, a firm’s own initiatives can influence the demand (e.g. through advertising or
irc
promotions). These aspects should obviously be taken into account when moderating
forecasts that are generated by a model.
C
10.1.10 Collaborative Planning, Forecasting and

d
Replenishment (CPFR)
ite
In supply chains, information sharing is a critical issue. Forecasting is an activity that

generates crucial information on the basis of which plans are made, orders placed,
im
dispatches made, stocking decisions taken etc. It makes sense to try to see if this can
be done collaboratively with supply chain partners. CPFR is an attempt to achieve
this.
rL
We remark that each supply chain player can very well forecast independently, but
several decisions made by one player as part of a plan can affect the resulting flow of
Fo
information (e.g. orders) and this is what should be shared where appropriate. An
example is the following. A manufacturer decides to move to rail based container
movement to cut transport costs. This involves batching of dispatches to some
locations.
This creates a lumpy demand for the supplier, which if seen in isolation at one
time, seems like a big swing in the demand. This is really a policy driven decision, and
not a real uncertainty in demand. With this, we observe that a lot of ‘randomness’ in
demand vanishes and what remains to be estimated is the true fluctuation in the
primary demand (and supply) which the supply chain has to really grapple with.
CPFR is now part of a business level set of standards in forecasting, in which major
y
retailers and suppliers are participating, so as to improve the state of the art.
nl
10.2 MANAGEMENT OF INVENTORIES IN SUPPLY CHAINS
O
The management of inventories in supply chains has two broad aspects. One is the
planning of quantities at different stages based on aggregate demands, capacities and
n
overall economies of production—this is the inventory planning decision. The
io
second is the inventory control issue of responding in the short run to demand and
supply uncertainties and the operation of safety stocks. We discuss both of them in turn.
at
10.2.1 Inventory Planning ul
The single stage inventory planning problem to minimise costs of production and
irc
stocking is exemplified in the well-known Economic Order Quantity (EOQ) or
Economic Batch Size (EBS) concept. Even this fundamental concept captures some
C
ideas central to supply chain management, where the attempt is to meet a given
(steady) demand at minimum cost considering the basic technology on hand—which
considers either the batch production cost or the batch ordering cost. It is interesting
d
to note that both these elements of cost are coming down due to technological
ite
advances.
1. Batching constraints are coming down in manufacturing because of flexible,
im
quick change technologies in many processes.

2. Batch ordering costs are coming down because of developments in IT.
rL
For example, a shipment of goods may be consumed uniformly over time at a

certain rate, for which we charge an average inventory holding cost, but equally true
is that it takes a uniform production rate a certain time to build up that shipment. In
Fo
a traditional system, since it was another person (the supplier) who was bearing that
cost, it did not factor in the EOQ computation, but the supply chain view is that the
cost is anyhow borne and reducing it is of the first order of importance, with the next
task being that of splitting the accrued gains.
10.2.2 Inventory Control

Inventory control refers to the management of inventory under uncertainties in
supply or in demand, or both. Two commonly used ideas in inventory control are the
following:
y
1. Fix the size of an order or a shipment, but allow some flexibility in the timing
nl
of the order. This is often called a reorder point system with a fixed reorder
quantity or (R, Q) policy. Here R is the inventory level which triggers off a
O
reorder, which is always a certain fixed amount Q. These parameters R and Q
can be chosen ‘optimally’ based on cost consideration in an expected sense.
See the next section for an illustration of a commonly used variant of this
n
policy, the two bin policy.
io
2. Another commonly used idea is that of the reorder level, order up to level.
This works as follows. The ordering cycle is fixed such that orders can be
at
placed only at certain fixed intervals. The amount of the order, however, can
vary. A general form of this is the (S, s) policy, where an order is placed to get
ul
the inventory level back up to S whenever the inventory in the review period
(those intervals when orders can actually be placed) falls below s. Note that
irc
since we survey the system only at certain periods, the inventory at that time
could be below s. In fact, in most such systems, a certain percentage of stock
out is anticipated and accepted as a via media solution from a cost
C
perspective. The value of s is chosen so as to guard against stock out in the

lead time and the value S is chosen keeping in mind the economies of
ordering and holding inventory.
d
ite
10.2.2.1 Cost Models

The basic idea in planning for inventory under uncertainties is the following. First,
im
analyse the consequences of falling short by one unit of inventory (this includes the
cost of express shipments in the case of committed demand, the cost of losing
customer goodwill, the opportunity loss of the margin of sale assuming that the
rL
customer will not return for that particular sale), etc. Call this cost Cu (the cost of
under-stocking). Next, analyse the consequences of having one unit of excess
inventory (this includes inventory holding costs, the cost of capital, the cost of
Fo
storage, and perhaps even the possibility of spoilage, obsolescence, losses etc.). Call
this cost Co (the cost of overstocking). We emphasise at this point that there could be
some subjectivity in both these costs, but it is still useful to estimate them and use
them as guidelines for making the decision of how much to stock.
We can now see that the amount to stock will go up as Cu increases (it is more
attractive to stock as the cost of falling short is high) and will go down as Co increases
(it is expensive to wind up with excess stock). The precise trade-off requires an
estimate of the demand, which is uncertain. The standard approach is to capture this
uncertainty through a demand distribution (using the histogram derived from past
y
data, for example) and position the inventory at a level Q* = F-1(Cu / (Cu + Co)). This
nl
formula is of considerable theoretical importance and is the basis for cost optimal
inventory control in an uncertain environment. It seeks to minimise the expected (or
O
average) costs of operating in this environment. This Model is referred to as the “news
vendor” model in the inventory control literature. Note that in such a policy, one
n
expects to fall short some of the time and have excess some of the time (as per the
io
random demand), but the optimal thing to do (with all the assumptions of cost and
demand behaviour) in the long run is to stick to this quantity of stock at the
at
beginning of the period under consideration.
The same argument can be used to derive an ‘optimal’ policy for some similar
ul
situations. See example below.
irc
10.1
C
What is the optimal time of ordering under uncertain lead time conditions, where
there is a cost of material arriving early and a cost of arriving late?
d
Suppose trucks arrive at a loading point after t days and with probabilities as given
below in Table 10.1 below. This summarises the uncertain supply condition of trucks
ite
that arrive between two and 10 days after placing an order.

im
Table10. 1
rL
Days (t) 2 3 4 5 6 7 8 9 10
p(t) .01 .02 .3 .25 .2 .15 .05 .01 .01
Cumulative .01 .03 .33 .58 .78 .93 .98 .99 1.0
Fo
probability F(t)
If the desired time of shipment is time t* and if the truck arrives before t*, then a
cost Rs per day (cost due to early arrival of trucks) is incurred. This cost is basically
waiting charges which are incurred due to keeping the trucks idle and could
correspond to demurrage charges by the vehicle supplier (commonly charged for
ships or railway rakes) or one’s own fleet utilisation charge.
This cost can be calculated in many ways as below.

(1) From depreciation cost: Suppose truck cost is Rs 10,00,000 and life of truck
is 18 years, then C1 = 10,00,000/18 ´ 365 = Rs 152 per day.
(2) From the profit earned: Suppose profit earned from a truck per year is Rs
y
6,00,000 then, C2 = 6,00,000/365 = Rs 1,644 per day.
nl
If truck arrives after time t* then some cost Rs C2 per day (cost due to late arrival
of trucks) is incurred. This cost depends upon the type of commodity, inventory
O
holding cost, penalty charged due to late delivery and future market loss due to poor
customer relationship etc.
n
The intention is to place an order at the optimum time before one needs it so as to
io
minimise the total expected cost, which in this case is given below as
t* 10
at
Total expected cost = C1 å ( t * - t )p(t) + C2 å ( t - t * )p(t)
t=2 t=t *
For finding optimal t* ul
C1
irc
F(t*) = , let C1 = 60 and C2 = 50, F(t*) = .05455
C1 + C2
Expected cost for t* = 5 is E.C5 = 62.9 rupees per day
C
Check that the expected cost for t* = 4, C4 = 86.6 rupees per day and the expected
cost for t* = 6, E.C6 = 66.7 rupees per day, and that t* = 5 is optimal. A schematic
d
C1
view in Fig. 10.3 plots the critical ratio against the cumulative distribution
ite
C1 + C2
function and derives the optimal time. The solution in this case is to place an order
five days before one needs the vehicle.
im
This idea leads to several natural and useful extensions. One is the inventory
control of a non-perishable item over multiple periods, where this quantity Q* is
rL
then viewed as an order-up-to level. In the absence of any fixed costs for ordering, this
leads to a simple, implementable policy: in every period, examine the inventory and
order enough to push the level back up to Q*. Note that this could lead to some
Fo
instances of small sized orders.

How would one approach this question when there is a fixed cost C for every order
that you place? This can be viewed as an extension of the EOQ idea, but in the
presence of uncertain demand. If we assume a demand rate, akin to the EOQ
situation, we can derive a reorder point, re-order quantity policy, which says that
when the inventory level hits a level R, place an order for a quantity Q. This will
Figure 10.1
0.9
y
0.8
nl
0.7
Cumulative probability
O
0.6
0.5
n
0.4
io
0.3
at
0.2
0.1 ul
0
irc
1 2 3 4 5 6 7 8 9 10
Days
C
prevent the placement of very frequent, small orders (keeping in mind the ordering
cost C). The exact formula for this optimum policy is a bit involved. The key point is
d
that it is the uncertainty of demand during the lead time that is of importance.
ite
A simpler version is the two bin policy, where R = Q, where there are two bins each
of capacity Q, and the policy is simply to place an order of one bin-worth when the
bin number 1 becomes empty. Intuitively, bin number 2 contains the stock that will
im
be used till the ordered material actually arrives.

Finally, the multi-period, discrete time-bucket version of this is the (s, S) policy,
rL
mentioned in Sec. 10.2.2.1. Note that this periodic review and replenishment policy
is non-trivial even for the zero fixed cost case.
Fo
10.2.2.2 Service-Level Models

Cost minimisation provides one rationale for inventory control. Another transparent
and customer oriented method in the presence of uncertain demand is service level
based inventory planning. Here, if the demand pattern is known, say as a histogram,
we peg an inventory level so that we fall short of meeting the demand at most a per
cent of the time. Then, 1– a is the service level. If the demand is normally
distributed, say with mean µ and standard deviation s, then holding an inventory of
m + 2s, for example, will ensure that we are short no more than 5 per cent of the time,
which may be acceptable.
y
10.2.3 Positioning of Inventory
nl
In a multi-stage system, should one have inventory at upstream stages or downstream
stages? If the same item is stocked (without any major material transformation but
O
involving operations like repacking and local customisation), there is a case for
holding it as close as possible to the customer points. Especially if the customers are
n
willing to wait a little (some of the time) and trans-shipment across locations is
io
possible at a cost, this is an attractive option.
When some value addition is in fact done at a certain stage, then retaining the
at
inventory at that stage allows for flexibility in meeting final demands precisely. In this
case, a centralised stock of inventory becomes attractive.
ul
irc
10.2.4 Management of Perishable Inventory
Almost all inventories are actually perishable, truly speaking, either because of
C
physical deterioration of products or because of product obsolescence. Some models

of inventory management do need to be significantly enhanced to handle this aspect.
In particular, the costs associated with holding inventory at the end of a planning
d
period may need to be modified and this results in smaller ordering quantities.
ite
Paradoxically, organisations would like to penalise themselves for not meeting

demand as well, pushing the system towards larger inventories. So it turns out that all
im
costs are higher, and to that extent, operating policies may not be too different. But
there is sometimes an impact on optimal stocking strategies.
rL
10.2.5 Management of Obsolescence

Fo
Planned obsolescence is now a business strategy in many firms. Such initiatives are
taken in conjunction with the planned introduction of new products. This is done to
continuously respond to changing customer needs and to compete based on new
features. The complication here is to take into account the end of horizon of the
product life cycle in inventory terms. Since customers are unlikely to be satisfied with
older variants when newer ones are available in the market, the attempt is to
minimise the presence of multiple variants at the same time.
In inventory terms, this leads to a tapered ordering policy, where the order up to
levels (and thereby the average inventory held at any time) is gradually brought down
as the time of product withdrawal nears. Two possible complicating impacts of this
are as follows:
y
➣ Service levels may suffer since the firm is operating with less inventory for
nl
some time. This would be harmful at a time when the firm is relying on an
increased market presence through a new product in the near future.
O
➣ However, customers may postpone purchases in anticipation of new products
that are already announced.
n
This phenomenon is quite prevalent in white goods, electronics and related
products. It is increasingly relevant in the software industry as well.
io
Simulation: Models such as the one above, can become analytically very complicated,
at
and for a one time understanding of the impact of various factors, simulation is a
useful technique. Broadly speaking, this would need the following steps (for the
ul
scenario above).
irc
(a) Demand patterns are captured from past data and using an appropriate
histogram, we should be able to generate a series of appropriate random
demands.
C
(b) The cost criterion and other service measures of interest need to be fixed.
(c) Equations relating the inventories and other parameters of interest need to be
d
developed. For example: Inventory in the beginning of period (n) =

ite
Inventory in the beginning of period (n –1) + Orders received in period (n –

1) – Demand in period (n –1).
im
(d) These quantities are related to the costs, e.g. positive inventories at the
beginning of a period indicate holding cost and negative inventories would
indicate shortage or penalties.
rL
(e) The simulation is done over a large number of time intervals to get a
statistically valid picture of behaviour.
Fo
The advantage of such an approach is that a number of practical considerations

(ordering costs, capacity constraints etc.) can be easily incorporated into the analysis.
Correspondingly, various performance measures of interest, including service levels,
can be quantified and assessed. Careful planning of numerical experiments can then
give a good idea of the impact of different parameters as well as the effect of different
policies.
Tools like Pro-model, Arena, Quest and several general-purpose simulation tools
can be employed for such an analysis, apart from building one’s own simulation
y
numerical procedures. Simulation is useful in assessing facility investments,
complex operating policies and many other decisions, before they are actually
nl
implemented.
O
A related technique is that of system dynamics, which emphasises certain
quantitative features of system stability, asymptotic behaviour and the time lags that
may occur in a system because of implementing some policies. Tools like iThink can
n
be used for this purpose.
io
at
10.3 LINEAR PROGRAMMING OR LINEAR OPTIMISATION
The framework of linear optimisation, popularly called Linear Programming (LP)
ul
when dealing with a finite set of variables and constraints, is a powerful and appealing
irc
one. We do not attempt a comprehensive review of this topic, but illustrate its use
with a number of examples.
Even if ‘nature’ is inherently non-linear, many practical conditions can indeed be
C
modelled as linear relationship between entities. One example is the inventory

balance over time mentioned elsewhere in the context of multi-period inventory
d
management. A similar condition arises in the inventory balance across locations. For
ite
example, for a given product, the amount received at a warehouse in a week plus the
initial inventory equals the departures to various stocking points in a week plus the
inventory on hand at the end of week. A particularly simple form of this is the so-
im
called transportation problem–perhaps more accurately called the trans-shipment

problem.
rL
Exercise: Empty containers emerge in a nation wide logistic operator’s network at

the end of every day and they have to be re-balanced as per the requirements on hand
at various locations. How is this to be done in a cost optimal manner if the following
Fo
are the locations and quantities?

Requirement Matrix Cost Matrix
A B C D E F G H A B C D E F G H
A — 25 24 40 29 35 27 12 A 0 200 450 450 150 250 300 600
B 18 — 50 30 12 45 5 10 B 200 0 300 180 500 450 230 480
y
C 35 18 — 28 35 15 10 36 C 450 300 0 320 350 150 150 300
nl
D 13 25 30 — 10 15 13 15 D 450 180 320 0 250 250 300 600
E 14 33 15 41 — 17 16 15 E 150 500 350 250 0 350 430 200
O
F 26 10 28 17 30 — 43 15 F 250 450 150 250 350 0 280 160
G 11 14 5 21 10 21 — 40 G 300 230 250 300 430 280 0 400
n
H 15 10 15 14 8 12 20 — H 600 480 300 600 200 160 400 0
io
For example, the entry 45 in the cell BF indicates that 45 wagons per week are
at
required to be moved from B to F and the entry 450 in the BF cell of the cost matrix
indicates the transportation cost of moving a empty wagon from B to F.
ul
First identify the locations where there is a surplus of wagons every week and
irc
locations where there is a deficit and then a plan to balance these requirements at
minimum cost.
From a mathematical point of view, a special case that is well-studied by students is
C
the assignment problem where there are an equal number of supply and demand
points and each shipment or assignment is of one unit. Although this may have some
d
applications, in practice the standard formulation of this problem is not that useful.
ite
The transportation problem on the other hand, is quite widely applicable, and in
fact, occurs as an underlying structure in more strategic decisions such as plant/
warehouse locations.
im
10.3.1 Network Flow Models

rL
A really useful analytical class of LP formulations, which is applicable in very many

contexts, is that of network flows. This class of formulation is exemplified by the
Fo
minimum cost flow problem, which is as follows:

Transportation network consists of n nodes. There are three types of nodes. T is the
trans-shipment node. S is the source node and D is the demand node. ij is a link
between node i and node j.
cij is the Cost of unit flow in link ij
Fij Flow on link ij

Cij Capacity of link ij
Si Supply from the source node i
Di Demand at the destination node i
y
cij * Fij Cost of net flow on link ij
nl
Thus total cost of flow over all nodes å å c ij * Fij
O
i j
Objective is to minimise this cost
n
Min å å c ij * Fij
io
i j
Net flow at each node is zero if it is a trans-shipment node
at
å Fij - å F ji = 0 "i if i ÎT
i j ul
Net flow at each node is equal to supply if it is a source node
irc
å F ji - å Fij = Di "i if i Î D
i j
C
Net flow at each node is equal to demand if a destination node
å Fij - å F ji = Si "i if i Î S
d
i j
ite
Flow on each link should satisfy the capacity constraint

0 £ Fij £ Cij
im
This formulation permits, among other things, the modelling of

➣ The shortest path decision in directed networks
rL
➣ The trans-shipment/assignment problem

➣ The longest path (critical path) determination in activity networks in project
Fo
management
➣ The max flow problem
➣ The tanker scheduling problem (a simple version of the fleet utilisation
problem in a transport network)
and many others.
If more than one commodity is flowing is the network, the problem is

conceptually similar, but generally much harder to solve, computationally speaking.
This multi-commodity version of this problem requires some specialised tools.
Example: A Network Flow Model of Iron Ore Movement
y
(From Section 8.10)
nl
The multiple paths of iron ore in the example in Section 8.10 (Example 3) allow a
network flow characterisation. The nodes of the basic network are (a) supply nodes
O
where iron ore is produced, (b) trans-shipment nodes, where ore is loaded to a mode
of transport or transferred from one node to another and (c) terminus nodes, where
n
ore is finally put on to ocean going vessels. More than one node could correspond to
io
a single physical location if there are multiple ways of handling material. In
particular, the unloading node representing Mormugao port is in four parts
at
depending on how the material is handled; by the land based mechanised bulk ore
handling system, by mooring dolphins and trans-shippers, by manual unloading or
by rail. ul
The flows (yearly movement of iron ore) on various arcs of the network in two
irc
successive years are shown schematically in Figures 1 and 2 in Section 8.10,
Example 3. To derive more detailed schedule related information, the flow network
can be replicated to include time by considering either a month or a fortnight as a
C
time bucket. The capacities at certain nodes are significantly different in different
time buckets, e.g. Panaji port during monsoon. A single instance of the network flow
d
model would include the capacity and cost constraints on different arcs and would
answer questions like:
ite
(a) What are the expected flows through Panaji port and Mormugao port, given
the constraints on loading, unloading and movement (including the
im
movement through the Cumberjao canal)?

(b) What is the impact of increasing capacity of the facilities at Mormugao port?
rL
(c) What will be the impact of increased facilities of iron ore handling through
rail at Mormugao port, through the tippler unloading system?
Fo
Modelling Capacities
Capacities on different segments of the supply chain translate naturally to capacities
on arcs of the network shown in Figures 1 and 2 in Section 8.10, Example 3.
Depending on the granularity of the model (e.g. weekly flows or yearly flows), the
capacity is calculated based on specific operating conditions. It is convenient to take
a standard barge operating size for looking at the throughput in tonnes through
different segments and modes. Interestingly, with an increase in traffic in
2003–2004, barge owners are considering investing in barges of larger capacity (upto
2,000 tons), as it is economically viable, the draft permits it and the traffic can sustain it.
For example, maximum yearly flows at the Panaji port are calculated based on the
availability of the port for eight months of the year and some maximum number of
y
barges per week, considering the loading options available. The capacity of the river
arc segments is approximately governed by the draft available and the tide conditions.
nl
Actually there is a complex relationship between the size of the barges that are
O
operated and the achievable throughput, because for larger barges, there are monsoon
related restrictions as well as restrictions on moving in the Cumbarjao canal.
n
Modelling Costs
io
Operating costs (in our case, costs of moving one ton over the relevant link) are
approximated in by a linear function of overall flow (assuming a certain barge fleet
at
adopted by the group of operators). Individual flows on the network clearly incur
different costs, depending on the size of the barge, but at the large levels of traffic, we
ul
can assume that the operating ranges of costs on any segment will vary linearly with
the total traffic on that segment. Costs are then mainly a function of barge size
irc
(already accounted for in the vessel mix) and time cost (including paying for crew
time). Other costs, such as diesel and vessel maintenance costs are assumed to be
C
same, no matter what mode of operation is chosen.
Cost of Expanding Capacity

d
From an infrastructure point of view, capacities on many links are flexible. More
ite
capacity can be added to some links by actions such as increasing the number of
trans-shippers, dredging of the canal, night navigation aids, etc. One way to handle
im
these options in a simplified model is to model costs as non-linear functions of the

flow (simple polynomial functions which represent convex, increasing costs with
respect to flow). This is particularly simple to do when examining incremental
rL
scenarios where traffic in some arcs is guaranteed to be in a certain range.
Analysis: The formal model yields a network of modest size (the base network is 14
Fo
nodes and 14 arcs and can be extended to include more detailed options), and one
which is flexible in its use. For example, the model can quickly estimate the flows on
different parts of the river canal system at different times of the year and can also
identify bottlenecks over a longer time frame. Observing the changes in traffic from
2002–2003 to 2003–2004, we see that the bulk of the increased movement has been
by the following two paths:
(1) Karnataka mines to Tinaighat and other stations before the ghat—road to the
Mandovi loading points—Panaji port (mid stream grab loading)
(2) Karnataka mines to Tinaighat and other stations before the ghat—Road to
the Mandovi loading points—Cumbarjua canal—Mormugao port (mid
y
stream grab loading and trans-shipper).
nl
Current bottlenecks in the network (correspond to saturated arcs in our flow
model) are (a) grab loading at Panaji (b) trans-shipper at Panaji (c) berth loading at
O
Mormugao, (d) rail movement from Karnataka through the ghats to Sanvordem and
(e) movement on the Cumberjao canal. With these as capacitated arcs, it can be seen
that the only increased flow of material that can be handled is from the southern Goa
n
mines through the Zuari river and through Mormugao port.
io
The model shows that increased traffic from the Karnataka mines can be achieved
at
by a variety of options, depending on the costs of increasing capacity on various links.
The capacity expansion options corresponding to the five bottleneck arcs correspond
ul
to (a) and (b) : Increasing accessibility of Panaji port during the monsoon—this is
quite an expensive option as it involves an analysis and feasibility study to mitigate
irc
the effect of a long standing oceanographic phenomenon, viz. the Aguada sand bar,
which makes navigation difficult in the monsoon months; (c) increasing mechanised
handling facilities at Mormugao, which is possible although there are some space
C
constraints; (d) doubling of the railway line between Londa-Castle Rock-Sanvordem,

which is quite an expensive option and (e) dredging and improving navigability of
d
the Cumbarjao canal, which is possible up to a small increase. These options can be
ite
modelled by appropriate cost functions on arcs, still yielding a tractable and

insightful flow model.
If the network is extended to include (a) the road options from Tinaighat to Zuari
im
loading points and (b) direct rail from Karnataka (Hospet/Bellary) to Mormugao
Port and (c) a tippler unloader node at the port that can handle rail traffic, then other
rL
expansion options can be explored. This is easily done in the model.

These options can be explored using linear and non-linear (convex) minimum cost
Fo
flow models on the appropriate network. The base network has 14 nodes and arcs (a
more compact formulation is possible, but the model that we use leaves open the
possibility of selectively increasing the capacity of different parts of the physical
network) and the extended network has three more arcs in the network. As usual in a
minimum cost flow model, there are several options of including the costs and
capacity constraints and we have chosen a simple variant where capacity constraints
are imposed only on arcs and not on nodes. The optimal-flow model (solved in
LINGO) has flow balance constraints and capacity constraints on individual arcs and
a linear or non-linear objective function.
10.3.2 General Linear Programming
y
The framework of LP is used in many other decisions as well. We discuss two of them
nl
here, from the manufacturing environment.
O
1. The production planning problem: This problem balances the cost of production
in various periods (allowing for variable production costs) vis-á-vis the costs of
n
inventory holding, capacity sub-contracting and cost of backlogging (or cost of
delayed satisfaction of demand).
io
We can see a few qualitative ways in which the overall cost function would
at
behave. If we follow the requirement every month (subject to capacity constraints),
we incur the lowest possible inventory costs. This is called the lot-for-lot solution
ul
in MRP/ERP systems. But this may incur a larger production cost, if costs are
different in different time periods (e.g. because of sub-contracting costs or seasonal
irc
production costs, including procurement of raw materials). The alternative of
advance production to take advantage of lower costs in some periods leads to higher
inventory costs. The balance between these factors is the one captured by the linear
C
programme.
d
ite
10.2
Assume that there is a production capacity of B units available each month. If the cost
im
of inventory is Rs h per unit per month, production costs are ci in month i, and
quantities required are di in month i, then we look for the lowest cost, feasible
rL
production plan. This is obtained by solving the LP on the variables xi (the amounts
produced in period i), as below.
Fo
Min Si [hIi + ci xi] (total cost of inventory and production)

where I0 = initial inventory,
Ii = Ii –1 + xi – di, for all i = 1, …, n (the inventory balance condition)
Ii >= 0 all i (this captures the no backlogging condition)
and B >= xi >= 0 all i.
2. The product mix problem: This is a very important and useful formulation in
aggregate planning where a facility manufactures multiple products. It has been
successfully implemented in petroleum refineries, to name one specific sector. Here,
particular care needs to be paid to material balance equations which could involve
conversion co-efficients, including yield of processes.
y
nl
Such formulations typically involve a certain number of constraints (say m) and a
much larger number of activity variables (say n). The LP formulation makes it quite
O
explicit that there are many potential bottleneck resources in a system and a small set
of them could be operative bottleneck resources at any cost optimal operating point.
An interesting fact is that, generally speaking, very few of the activities would be at a
n
non-zero value at the optimal operating policy. In the above instance, only m of the
io
activity variables would be non-zero.
at
Also, only some of the resources would be binding or bottleneck resources. These
would have non-zero impact on the objective function, if relaxed. The section below
ul
points to the fact that further useful information is available through the techniques
of duality and sensitivity analysis.
irc
10.3.3 Solution Techniques for LP and the Role of Sensitivity
C
Analysis and Duality

Linear programming is a popular and well established tool for analysis and decision-
d
making. The solver utility in Microsoft Excel allows reasonable sized formulations
ite
and solution. Several modelling languages and solver engines provide more detailed
formulations. This includes Lindo/Lingo, Gams (popular in the process industry and
for formulating macro-economic models) and Cplex, a most powerful general
im
purpose LP solver which is also used in Ilog and other resource planning tools.
Intimately related with the effective solution of LPs is the fact that there is a well
rL
established sensitivity analysis that is possible. This arises from duality theory, which
provides an elegant, economic analysis to the issue of certifying optimality,
identification of bottleneck (binding) constraints and the impact of perturbation of
Fo
certain parameters (especially resource availability or demand requirements) on the

final solution to an LP. This analysis provides a rich basis and motivation for problem
solving using formal tools such as LP.
Solution of LPs
There are two main techniques for solving LPs, extreme point iterative methods
(referred to as the simplex method, because a simplex is a full dimensional
polyhedron) and interior point methods (which are now getting to be part of
y
commercial LP software). Both have comparable practical performance but differ
nl
widely in the underlying mathematics and also some theoretical properties (especially
regarding convergence to the optimal solution).
O
10.3.4 Integer Constrained LP (Also Called Integer
n
Programming – IP)
io
These are optimisation problems where there is a linear relationship between
variables in the constraint and objective function with the added condition that the
at
variables are constrained to be integer. The most common use of this is in modelling
Yes/No decisions using {1, 0} variables. This is of use in several practical problems, for
ul
example facility location, scheduling problems and the production planning problem
with fixed costs or batch constraints. A formulation in which some variables are
irc
restricted to be integer and some are from a continuous domain is called a Mixed
Integer Linear Programme (MIP).
C
10.3
d
ite
Distribution Network Design

A firm has to decide on the location of plants to meet their customers demand
im
located in m cities. There are n potential sites for the plants. Let Fi be the annual fixed
cost of opening a plant at location i, Ki be the capacity of the plant at at location i, Dj
be the customer demand at city j and cij the unit transportation cost from location i to
rL
city j. The aim is to design the network to meet all the demand. This is obtained by
solving the MIP on the variables xij (shipped quantity from plant located at i to city j)
Fo
and yi (1 if plant is open at site i, 0 otherwise)
Min Si Fiyi + Si,j cij xij

s.t. Si xij > = Dj for all j Constraint set 1
Sj xij < = yi Ki for all i Constraint set 2
xij > = 0, yi in {0,1}
Here, the objective function captures the total cost over an operating horizon,
including the fixed cost of operation and the variable cost of transport. There are
choices here, one being the consideration of costs over a finite horizon (say five years)
over which the plant is assumed to operate and therefore the expected transport costs
over that time. The other option is to annualise the fixed cost using some method of
y
capital cost depreciation, and consider annual costs. The data regarding costs and the
nl
demand estimates affect the model outcome significantly and estimating these
satisfactorily is a big part of successful modelling.
O
Constraint set 1 in the above formulation indicates that all demand is met by
supply from various locations and constraint set 2 indicates that supply is not made
n
from a facility unless it is declared as open. Check that values of yi being zero or one
io
does give the required meaning in each case. Here again, there are multiple ways of
modelling such a constraint. This is a formulation which is adequate for moderate
at
sized decisions.
10.4
ul
irc
Consider an example where customers are located in four cities and there are four
C
potential sites for the plants. The model inputs are given in the following table.
Plant Customer location Capacity Fixed cost

d
location C1 C2 C3 C4
ite
P1 16 12 13 17 40 200
P2 14 19 15 13 50 300
im
P3 18 11 13 16 70 450
P4 15 8 10 12 80 550
Demand 29 37 40 24
rL
The cost optimal solution in this case is to select locations P2 and P4. In the same
example, if the fixed cost of facility P2 is 1000 units, the model selects locations P3
and P4 as the best solution.
Fo
In general, the more the fixed cost of facilities, the fewer of them are declared open
in the optimal cost solution.
We only note that IP, in its most generic form, is difficult to solve, but there are
some classes of IPs (arising from network flow formulations, mainly) that are quite
tractable. Most LP solvers also include the option of including integer restriction on
variables, but these may not be useful for large unstructured problems. Such
problems would require specially customised and constructed algorithms.
10.4 ROUTING MODELS
y
Routing is a concept that is useful in operations planning of transport. A term used in
nl
this context is that of a milk-run, derived from the regular (daily) multi-point
delivery of largely standardised products—such as milk from a dairy to several retail
O
dispersing points. This concept is in fact very widely applicable in many settings such
as courier operations, delivery of soft drinks, collection of crates/bottles of soft
n
drinks, newspaper deliveries and several retailing applications. The advantages are:
io
➣ Cost minimisation of meeting the demands of a large set of demands, where
each individual demand may not meet a full vehicle load (called FTL—Full
at
Truck Load, in retail and short haul transport by road).
ul
➣ Frequent delivery cycles because of the above consolidation and as a result of
which retailers do not have to stock too much—since they are assured of
irc
supplies ever so often.
A decision that has to be made in a milk run is the order (sequence) of destination
C
visited. A single vehicle version of this decision is referred to in the Operations

Research literature as the Travelling Salesman Problem (TSP) whereas the multi-
vehicle version is known as the Vehicle Routing Problem (VRP). The TSP is a
d
classical combinatorial optimisation problem (of selecting the least cost option
ite
among n options when visiting n cities). This problem is not difficult to solve for
small n and has a number of heuristics (rule-of-thumb) procedure for medium and
im
large n, but is known to be difficult to solve to optimality. Note that it is possible to

get a solution by examining all options, which would involve a finite (but could be
very large) number of computations. Of more importance in SCM is that both the
rL
TSP and the VRP admit reasonably good solutions that are approximately optimal.
There are several variants of this that arise in practice which can make the problem
Fo
difficult. Of particular interest in express delivery services and courier operations is

that there are time windows for collection and delivery. In personnel scheduling,
factors like detours incurred by each entity could be of interest. Heuristic procedures
need to be designed to take care of such concerns in a reasonable way.
An argument working against the milk run idea is the following. When
commodities are not time sensitive and are needed in large quantities (e.g. bulk
commodities), dedicated location shipments maximise vehicle utilisations and reduce

costs (remember that a milk truck gets progressively empty as its van progresses, so
the average utilisation of the vehicle may be close to half its capacity).
Clarke Wright Heuristic:
y
➣ Calculate the savings resulting from combining location i and j into a single
nl
route for every location pair (i, j) and rank the savings in decreasing order.
O
➣ Pick the link with largest saving and include in a route if it satisfies the
constraints. If the constraints are violated, move to next link in the list.
n
➣ Repeat till the list gets exhausted.
io
Schematic View:
at
ul
irc
C
d
ite
im
Most multi-stage facility planning problems start with this basic design issue.
Extend this schematic diagram realistically to a real supply chain [e.g. cement] and
rL
evaluate different options.
10.5 PRICING DECISIONS

Fo
Pricing of products and services is clearly an important business decision driven by

marketing considerations in a firm, and the prevailing economic scenario. We discuss
a few aspects that provide a supply chain focus to this rather complex issue.
10.5.1 Transfer Prices and Supply Chain Entities

If we view a supply chain as an entity, then what determines its competitiveness is the
value generated in its entirety vis-á-vis what the end customer is really willing to pay
for consumption of the product or service. In this view, all transactions between
y
supply chain partners, including selling prices and purchasing prices between these
nl
entities, are all internal to the supply chain and do not affect the final profitability.
However, these internal prices do affect some key decisions of the various supply
O
chain players and to that extent, they do affect competitiveness.
A highly vertically integrated organisation would use a mechanism of transfer
n
prices between its constituent managerial units to judge the financial performance of
these units. It is quite difficult to know if these prices are ‘fair’ measures that can be
io
used for this purpose. They are sometimes quite artificial and to that extent, may
at
foster some inefficiencies in the system.
10.5
ul
irc
Indian Railways is a highly integrated organisation which manufactures coaches and
locomotives, runs catering units, hospitals and schools, as well as its ‘core’ business of
C
train operations. The internal pricing of locomotives is somewhat open to debate. It

is based on a cost oriented system of accounting. There are no regular competing
d
suppliers of locomotives of similar types, nor any regular customers, other than
ite
internal one, for the products. Therefore, there is no good market signal as to
prevailing norms in this sector. Prices are therefore not subjected to proper signals
that push the system to efficiency.
im
On the other hand, some services like catering are now undergoing a change in two
ways. One is the creation of a separate entity with its own managerial control. This
rL
brings in an element of transparency in the costs of services and may in itself lead to
some improvements. The second is the introduction of competitors, if not in the
same route, at least on similar routes elsewhere. Because of this, there are some
Fo
pressures to improve and/or standardise the offering of products/services.
10.5.2 Double Marginalisation

In the presence of uncertainty, a firm takes a position on quantities to be transacted or
stocked depending on the margin of profit made for every unit sold. Against this,
there is the cost of holding and/or the cost of salvage/disposal. Therefore the prices
transacted between supply chain entities, although internal to the chain, do affect
stocking quantities and therefore the actual availability to final customers.
Economists have alerted us to this phenomenon of double marginalisation, which is
y
now interpreted in supply chain terms. The basic impact is that whenever the overall
margin in a supply chain is split into two because of two supply chain entities
nl
controlling two parts of the supply chain activity, the chain as a whole understocks
O
products relative to the ‘optimum’ value.
n
10.5.3 Elasticity
io
It is well-known that demand for a product depends on the price at which it is
offered. If this aspect is to be formally captured in decision-making, a useful concept
at
is that of price elasticity of demand. A practical way of capturing this impact is
through price-demand curves, where demand Q is expressed as a (decreasing)
ul
function of price P such as Q = a – bP (for some positive constants a and b).
irc
Economists define elasticity in terms of relative changes in demand with respect to
prices. Such ratios are useful for computing or predicting equilibrium outcomes, but for
decision making related to pricing, equations such as the above, are used more often.
C
In practice, the demand for a product is also determined by the value proposition
it offers. If we consider S as a quality parameter (e.g. the hours of life of a light-bulb),
d
we can write demand Q as a function of P and S as Q = a – bP + cS, for some positive

ite
constant values a, b and c.

Finally, if we include cross-elasticities, which reflect the impact of competing
im
products, we can write Q = a – bP + cS + dP¢ – eS¢, where P¢ and S¢ are the price and
quality parameters of the competing product(s). The values of these constants are to
be estimated by statistical methods such as regression or other means. Such data is
rL
difficult to obtain and such models are more useful conceptually, to understand how
pricing and product/service quality levels affect demand, revenues and overall supply
chain performance.
Fo
10.5.4 Pricing Contracts in SCM

Contracts provide one mechanism for conveying price and other signals to different
entities in the supply chain, as we have seen in Chapters 6 and 7. In simple
transactions, we can consider quantity discounts and other pricing initiatives as part
of a single cost minimisation objective. When we look at longer term transactions

spanning more than one period, other types of contracts and pricing mechanisms can
be designed. We look at two possibilities.
1. Buy-back arrangements to counter the effects of double marginalisation.
Here a manufacturer agrees to take back unsold items at the end of a period.
y
This should encourage a retailer to order the ‘optimal’ value since part of the
nl
risk is now covered.
O
2. Service level based contracts: Supply chain analysis inherently bring focus to
the interface between two decision making units in a linked series of
activities. The timing of deliveries is one aspect of this. Late deliveries are
n
known to cause shortage and penalties to the system. The JIT concept
io
highlights the pitfalls of early delivery and ‘Just-in-case’ safety stocks. In a
similar vein, the Taguchi methodology in quality management highlights the
at
costs in any deviations from a given target. When this is formalised for a
logistic service provider, what emerges is a service level contract.
ul
irc
C
Hillier, F. S., G. J. Lieberman, (2005), Introduction to Operations Research, McGraw-Hill.

Nahmias, S., (2005), Production and Operations Analysis, McGraw-Hill.
d
Taha, H. A., (2003), Operations Research: An Introduction, Prentice Hall.

ite
Winston, W., (2003), Operations Research: Applications and Algorithms, Duxbury Press.
Case references : Models and techniques can be profitably used in almost all cases in the
im
book. Specifically, Linear Programming and mixed integer variants can be used in Farm
Aid Tractors Limited, Inventory theory in Bayer Crop Science, Fleet planning models in
Laxmi Transformers, Simulation in Western Oil Limited (A) and many others.
rL
Fo
1. If there is a qualitative shift in the market, and the data shifts from a base value K to
a base value K+G, how long will the moving average method take to discover this?
Remember that the data you would typically have is of the form [base value + e],
where e is some fluctuating term representing the randomness (usually assumed to be
a zero mean normal random variable with some variance).
2. For a concrete example, consider the series 4, 4, 5, 4, 5, 3, 4, 5, 5, 4, 6, 5, 5, 4, 5 ...

and make a guess as to when the market shifted from a base value of 4 to a base value
of 5. Assume some reasonable value of m, such as 3 or 4.
3. For a look at how the method works on extreme cases (e.g. without randomness),
consider how the moving average method works with the time series 4, 4, 4, 4, 5, 5, 5,
y
5, 5, 5 ... (we note that the human will discern that the market has shifted to a base
nl
value 5 in the fifth period, but only past fact!). How long will the moving average
method take to discover this (as a function of m)?
O
4. Apply the exponential smoothing method to the same sample data as in the moving
average case, for different parameters a = 0.7, 0.8, 0.9. Here, the larger a is, the more
n
weightage is given to the most recent data (similar to the case of small value of m in
the moving average method). In this sense, the two methods are quite similar. The
io
exponential smoothing method requires you to maintain the time series method in a
compact form, only the most recent data and the previous forecasted value. The
at
moving average method requires you to maintain the m most recent values. When
dealing with forecasts for hundreds or thousands of items, this may be an issue.
ul
5. Look at data given below and try to generate forecasts for each quarter in year 5.
Assume constant seasonality factors.
irc
Year Quarter Sales Year Quarter Sales
1 I 98 3 I 138
C
II 106 II 130
III 109 III 147
d
IV 133 IV 141
2 I 130 4 I 144
ite
II 116 II 142
III 133 III 165
im
IV 116 IV 173
6. Daily demand for an item is observed to be as in the table below, for the last 30 days.
rL
This includes items actually sold as well as unfulfilled demand.

25 38 26 31 21 46 29 19 35 39 24 21
17 42 46 19 50 40 43 34 31 51 36 32
Fo
18 29 22 21 24 39
For this data, plot the histogram. What is your estimate for the probability that the
demand in week 31 is less than 40? Using a normal approximation (for which you
would have to compute the mean and standard deviation of this sample), compute
this probability and compare with the empirical distribution. The empirical
distribution is a tabulation of thr values of Prob{X <= t} for different values of t, from
the observed data and their histogram. The normal distribution is also tabulated in
references as a table for observed value of t.
7. The traditional EOQ model considers inventory costs on the consumption side (the
saw tooth curve of inventory versus time). Extend this idea to include the inventory
costs on the supply side as well.
y
8. All other things being the same, what is the relationship between the lead time of
nl
supply L (days) and the order quantity Q (units)? Do you expect it to be increasing,
decreasing or neither? Linear or non-linear?
O
9. When ordering costs are zero, show that it is economical to order in every period (this
amounts to s=S)—even if it is a small size order. Such a system is called an order-up-
n
to system.
io
10. As before, convince yourself that if review is continuous, and replenishment was
instantaneous (lead time is zero), then a simple, trivial policy is adequate. Now assess
at
what would be the effect on R and/or Q if the lead time increases. Try to see if you can
deduce the nature of the cost curve with respect to R (for a fixed Q), which could arise
ul
because of transport constraints (say a truck load). Separately, argue what would
happen if the fixed cost for ordering is zero.
irc
11. If the demand at a warehouse is the aggregated demand from several retailers and is
observed to be as given in Section 1, over 30 weeks. How much inventory should you
hold at the warehouse to ensure that you (almost) never fall short?
C
12. A stockist of a food item orders daily. Items ordered today can last three days before
they have to be scrapped or returned. The demand is somewhat uncertain and follows
d
the histogram as derived from Exercise 6 above. Profit in selling an item is Rs 5 and
cost of the product is Rs 12.
ite
1. What is an approximate optimal stock level at the beginning of each day?

2. If there is a salvage value of Rs 2 for each unsold item, what is the stocking policy?
im
3. If the perishability of the item is now considered as two days (because of customer
perceived attitude to ‘fresh’ products), how would the ordering strategy change?
rL
These are difficult issues to analyse, especially if ordering cannot be done too often.
13. A retailer faces a weekly demand for a food item with the given data histogram
capturing the demand distribution for 16 weeks. What should the ordering strategy
Fo
be if the retailer has the knowledge that the product is to be withdrawn in four weeks
time? You can assume a nominal salvage value for returned, unsold items.
108 116 118 124 96 119 96 102 112 102 92
91 117 111 94 98
14. If capacities are not constant throughout of the year then the quantity of B is replaced
by the relevant capacity Bi in each month. This could be because of planned plant
maintenance or vendor capacity fluctuations that are known in advance. Try

modelling the constraint that the inventory at the end of each month cannot exceed
some quantity K (because of storage requirements).
Try applying this logic to the situation in the case of Farm Aid Tractors Limited.
y
We note that if there are batching constraints or fixed costs (which again push one
towards batches), the problem is more complex and is in the realm of integer
nl
programming. There are also multi-product versions of this production planning
problem where some group of products share resources.
O
15. Design a set of routes for collection of goods from the following eight locations where
the daily demands are given and the distance of each location from the other
n
(including the central depot) are as given below. Capacity of the vehicle is 10 tons.
io
at
ul
irc
C
d
ite
Try to design a general purpose procedure yourself and then compare it with the
im
following well-known procedure called the Clarke Wright heuristic for VRP.
16. Any manufacturing supply chain with (say) one stage of value addition or processing
rL
can be located close to the source of raw materials or close to the demand centres, or
something in between. For example, cement plants can be located close to limestone
deposit (raw material source) or consumption centres (e.g. urban agglomerations).
Fo
What are the pros and cons of each option? The same issue arises in the location of
power plants close to coal mines (pit head generation of electricity) versus location
close to consumption centres (urban areas).
17. Discuss whether the setting up of a supply chain in such an environment (e.g.
corporatisation of some part of the activities or privatisation and the introduction of
competition), actually helps the end customer effectiveness.
18. For the end demand that is random with demand distribution as in the histogram as
derived from Exercise 6 above, compare the cases of:
1. An integrated supply chain that has a raw material purchase price of Rs 4 and an
overall selling price of Rs 10.
y
2. A two stage supply chain that has an intermediate transfer selling price of Rs 6.
nl
19. Analyse a two period decision of a retailer with the following data.
O
n
P1: Transfer price
P0: Cost P2: Selling price
Manufacturer Retailer
io
(Determined by (Determined by
the market) P3: Buy back price the market)
for unsold items
at
Demand characteristic: 10 (low) Probability = 0.25
ul
12 (medium) Probability = 0.5
15 (high) Probability = 0.25
irc
Here, we make the assumption that the manufacturer announces the transfer price
C
P1 and the buy back price P3 first, and then the retailer makes the ordering/stocking
decision.
20. An iron ore transporter needs to move ore from a mine site to a barge location 50 km
d
away. The requirement is a steady ore of 500 tons a day. The material is moved in 10
ite
trucks moving at an average speed of 20 km/hr and allowing 30 minutes each for
loading and unloading. If the transit time is not fixed but varies between 2.5 to 4 hrs
with some distribution, estimate the number or trucks that the transporter should
im
deploy. For this, you would first have to determine a suitable objective of the supply
chain (inventory cost, contingency cost of storage of ore in case of transport delays at
either end, etc.). You may need additional data, which you can assume for the purpose
rL
of this exercise. After that, formalise this arrangement as a contract between shipper
and carrier.
Fo
Fo
rL
im
ite
d
C
Cases
irc
ul
at
io
n
O
nl
y
Fo
rL
im
ite
d
C
irc
ul
at
io
n
O
nl
y
CASE 1
y
nl
The issue of mode choice for movement of both raw materials and finished products is the
O
focus in this case. The location of the plant has already been decided based on various
considerations, such as the availability of natural gas as a fuel and access to new sponge iron
n
plants in Western India. However, the market for the final product of Laxmi Transformers,
DRI (sponge iron) was a growing one in India and there was scope for focusing on some
io
selected final markets. Logistics costs could form a significant factor in this consideration.
Apart from mode choice, distribution network design, investments for railway siding, barge
at
rentals and scheduling of ships could be explored for a complete logistics plan.
ul
irc
LAXMI TRANSFORMERS
C
Direct reduced iron (DRI), commonly called sponge iron, was to be produced by Laxmi
Transformers (LT) at its new Rs 500 crore Alibag (Maharashtra) plant starting in February
1991. DRI was an intermediate product in the steel making process and essentially served as
d
a substitute for scrap iron. DRI was used by both mini steel plants and large integrated steel
plants. It could also be used by ferrous foundries. Maharashtra and Gujarat were the two
ite
most industrialised states in India and had a number of mini steel plants and ferrous
foundries which were potential sponge iron buyers. All the integrated steel plants were
im
located in the eastern part of the country except the SAIL plant which was at Bhadravati in
Karnataka.
The advantages of using sponge iron were as follows:
rL
1. It was a substitute for ferrous scrap which was currently imported in large
quantities, costing the nation a large amount of foreign currency.
2. Undesirable elements which may be found in scrap such as chromium, tin, nickel
Fo
etc. were absent. This gave a more consistent and reliable end-product quality.
Prepared by G. Raghuram and Dilip Mathew. Case material of the Indian Institute of Management,
Ahmedabad, is prepared as a basis for class discussion. Cases are not designed to present illustrations of
either correct or incorrect handling of administrative problems. © 1990 by the Indian Institute of
Management, Ahmedabad.
3. It was easier to handle than scrap.
Location
The two primary reasons for locating the project at Alibag were the availability of natural gas
and nearness to sponge iron markets. Natural gas from the Bombay High offshore field was
y
brought to Uran by a pipeline, about 25 km due south of Bombay. Industries located in areas
nl
near Uran would get gas at what was called “landfall prices” which were substantially less
than what inland customers would pay. The landfall price of gas was about Rs 2500 per 1000
O
cubic meters. About 300 cubic meters of gas were needed per tonne of sponge iron. The
project site was on the seashore, with a view to having access to sea transport, both during
construction and operations.
n
There were a number of different processes that could be used for sponge iron production,
io
some being coal based and some being gas based. In consultation with the engineering
advisers of the project, Technocrats India Ltd., LT decided to adopt the HyL III process for
at
sponge iron production, under license from the Mexican firm HyLsa which had patented it.
This process used reformed natural gas to convert iron ore to sponge iron.
ul
The plant which had a rated capacity of 500,000 tonnes per annum, would require about
1.24 tonnes of pellets per tonne of DRI and 0.31 tonnes of lump ore per tonne of DRI,
irc
making for a feed mix proportion of approximately 80 : 20. A problem with the Alibag plant
was that the iron ore and pellets required were not available locally.
C
Transportation
LT’s newly appointed Manager, Logistics decided that his first task would be to review the
d
proposed transportation strategies of his company for the inward movement of raw materials
ite
and the outward movement of his company’s product.
Raw Materials
im
He first met the Manager, Raw Material Procurement (RMP) and then the Manager
(Marketing) to get their views. Manager (RMP) was responsible for identifying potential
reliable suppliers of the principal raw materials, namely, lump iron ore and iron pellets. He
rL
found that the Manager (RMP) had already made a thorough examination of the production
situation and entered into contracts with three mines that would supply lump ore. The
Manager (RMP) had also entered into a long term contract with Kudremukh Iron Ore
Fo
Company Limited (KIOCL), Mangalore, for supply of iron pellets which were essentially
produced by agglomerating fine particles of iron ore. Even though this was an export-
oriented unit whose rupee prices should ordinarily fluctuate according to the value of the
rupee with respect to the US dollar, LT was able to negotiate a special arrangement
guaranteeing its requirements at the rate of Rs 600 per tonne FOB. This rate would be
reviewed annually. The Manager (RMP) was able to clinch this deal because he negotiated at
a time when KIOCL was going through a very troublesome period financially.
Case 1: Laxmi Transformers 273
Based on considerations such as quality of ore and possible supply and handling
constraints, LT had decided to procure lump ore in the following proportions from the mines
mentioned below:
Percentage Price (Rs/tonne)
Daitari (Orissa) 40 250
y
Banspani (Orissa) 40 250
nl
Goa 20 330
O
These sources were not substitutable, since they had been identified according to certain
required chemical properties.
n
Finished Goods
io
The Manager (Marketing) was in the process of trying to line up customers for LT’s DRI.
While he had not yet been able to get any firm commitments, he had been able to make some
at
estimates of the size of the market. A major difficulty he faced in this task was that while
there were a few sponge iron producers in the country already, none of them could be called
ul
major producers, because, for various reasons, they were producing at only a fraction of their
capacity. The Manager, Marketing’s estimates of the market potential is shown in Exhibit 1.
irc
Inbound Mode Choice
C
LT was now faced with the problem of deciding the best way of transporting raw materials to
the Alibag plant and DRI from the plant to market centres. There were three possible modes
that could be used – rail road and sea – or combinations of these.
d
Rail
ite
The nearest railhead to Alibag was at Pen about 15 km away. The Railway Board in Delhi had
categorically stated that there was no prospect of building a line to Alibag, but suggested that
im
LT would be able to get a line constructed at its own expense at an approximate cost of about
Rs 1 crore per km (including handling and storage infrastructure). If other industries which
were coming up in the area (a cement bagging plant, a gas based fertiliser plant, and a few
rL
others in the drawing board stage) wanted to use the track, LT would be able to share the
construction cost.
In the case of rail transport, freight rates were determined by the Indian Railway
Fo
Conference Association (IRCA) which published a book of tariffs. Exhibit 2 gives relevant
tariffs. While iron ore was classified as Category 110 for purposes of tariff determination,
there was some confusion on the classification of DRI. While IRCA stated that sponge iron
came under Category 150, Central Railway’s commercial staff maintained that DRI was a
different product and would be charged under Category 210.
Sea
A jetty capable of handling four barges simultaneously with an unloading rate of 2000
tonnes per hour was to be built at the plant site in Alibag. Giant Western Shipping had said
that it would be able to provide two ships, one of 65000 deadweight tonnes (dwt) and the
other of 35000 dwt. LT would be expected to pay market rates whichever ship it decided to
y
employ, as the shipping industry was riding on the crest of a wave. Going rates and details of
nl
ship operating costs and times are given in Exhibit 3.
High-tonnage ocean-going ships like the ones being considered, could not enter Alibag’s
O
minor port due to draft restrictions. Hence, cargo had to be loaded/unloaded in deep water
into/from barges which could use the jetty. Each barge could be used for one round trip in a
day, due to the tidal variations. The current plan was to charter five barges of 1,000 tonnes
n
payload capacity for this task. (The limits for unloading were determined by the operating
rate of the cranes on the ship. A total of 10000 tonnes per day was possible, but then 10
io
barges would be required. Five would be at the ship side when the other five were at the jetty.)
The charter rates for these barges were Rs 300 per tonne per month. Deep water operations
at
at Alibag were not possible for 120 days of the year due to the monsoon.
Road
ul
Road transport rates were expected to be 50 paise per tonne for a kilometre, though, due to
irc
the new Motor Vehicles Act and consequent strict imposition of no overloading, the rates
could possibly go as high as 70 paise per tonne kilometre.
C
Exhibit 4 gives a distance matrix between the various mines, Pen station, ports and market
centres.
d
Considerations for Outbound Movement

ite
In examining the options LT had regarding outbound movement, the following were some of
the key issues that had to be borne in mind:
im
1. Since the ships that bring iron ore would be returning empty (being on a time
charter), sponge iron to markets in the eastern parts of India could be transported by
ship. This would involve additional time due to loading and unloading, but not due
rL
to travel time.
2. Scheduling the ships for various trips could become an issue.
3. If LT opened a stockyard for redistribution, it would cost about Rs 1 lakh per month
Fo
to maintain the stockyard.

4. Inventory carrying costs would be significant.
5. If trucks were to be used for outbound movement, two issues would dominate.
Would it be possible to get enough 10 tonne trucks to carry the quantities
demanded? Would the plant be able to handle such a large number of trucks?
6. If a rail-siding were to be laid from Alibag to Pen, inbound rail transport would also
be facilitated. Regarding outbound movement, only the major consumers could
handle direct rail movement. If a siding was not considered, and rail movement was
still planned from/to Pen, then road movement between Pen and Alibag would take
place at a cost of Rs 30 per tonne, including handling.
7. Each additional handling of DRI resulted in a loss of 1%. The average selling price
of DRI was expected to be Rs 4000 per tonne.
y
8. The above considerations could even influence the market choice for LT, since
nl
demand was expected to exceed supply.
O
The Manager (Logistics) was keen on developing a transportation plan keeping all these
considerations in mind.
n
Exhibit 1: Market Centre wise Demand Estimate for Sponge Iron (10,000 tpa)
io
No. Market Centre 1990/91 1995/96 2000/01
at
1 Delhi 8.8025 17.605 35.21
2 Hissar 1.0375
ul 2.075 4.15
3 Ambala 0.885 1.77 3.54
irc
4 Ludhiana 3.8475 7.695 15.39
5 Jaipur 2.475 4.95 9.9
6 Lucknow 3.4325 6.865 13.73
C
7 Muzzafarnagar 1.7375 3.475 6.95

8 Tatanagar 2.0275 4.055 8.11
d
9 Ranchi 1.175 2.35 4.7

ite
10 Barajamda 0.0525 0.105 0.21

11 Calcutta 10.705 21.41 42.82
im
12 Rajkot 0.0525 0.105 0.21

13 Ahmedabad 0.845 1.69 3.38
14 Baroda 1.3075 2.615 5.23
rL
15 Raipur 1.7775 3.555 7.11

16 Gwalior 0.565 1.13 2.26
Fo
17 Indore 1.9125 3.825 7.65

18 Jabalpur 0.1825 0.365 0.73
19 Bombay 9.37 18.74 37.48
20 Nagpur 2.485 4.97 9.94
21 Aurangabad 0.35 0.7 1.4
(Continued)
(Continued)
No. Market Centre 1990/91 1995/96 2000/01
22 Kolhapur 0.6775 1.355 2.71

23 Waltair 0.8075 1.615 3.23
y
24 Secunderabad 1.16 2.32 4.64
nl
25 Kothagudam 2.44 4.88 9.76
26 Bangalore 2.65 5.3 10.6
O
27 Hospet 0.5575 1.115 2.23
28 Bhadravati 1.48 2.96 5.92
n
29 Calicut 1.275 2.55 5.1
30 Madras 2.1175 4.235 8.47
io
31 Tiruchirapalli 0.245 0.49 0.98
at
68.435 136.87 273.74
32 Bhilai 20 25 30
33
34
Bokaro
Rourkela
ul
20
9
30
11
35
15
irc
35 Durgapur 8 9 11
36 Burnpur 5 5 6
37 TISCO, Jamshedpur 10 15 20
C
38 VISL, Bhadravati 2.5 3.5 5

Total 142.935 235.37 395.74
d
ite
Exhibit 2: IRCA Tariff (Rupees per Quintal)

im
Category 110 150 210 110 150 210

Km Rs Ps Rs Ps Rs Ps Km Rs Ps Rs Ps Rs Ps
rL
1-100 4.25 5.44 7.24 341-350 10.74 14.27 19.58

101-105 4.54 5.82 7.77 351-360 10.98 14.59 20.03
106-110 4.66 5.99 7.99 361-370 11.21 14.91 20.46
Fo
111-115 4.79 6.17 8.21 371-380 11.46 15.23 20.91

116-120 4.91 6.32 8.45 381-390 11.67 15.55 21.36
121-125 5.02 6.48 8.68 391-400 11.91 15.87 21.80
126-130 5.14 6.65 8.91 401-410 12.59 16.78 23.06
131-135 5.26 6.81 9.14 411-420 12.84 17.12 23.52
(Continued)
(Continued)
Category 110 150 210 110 150 210

136-140 5.38 6.98 9.37 421-430 13.08 17.45 23.99
y
141-145 5.49 7.14 9.59 431-440 13.33 17.78 24.46
nl
146-150 5.61 7.29 9.84 441-450 13.55 18.11 24.92
151-155 5.74 7.47 10.06 451-460 13,80 18.44 25.38
O
156-160 5.86 7.64 10.28 461-170 14.05 18.77 25.84
161-165 5.98 7.79 10.52 471-480 14.30 19.10 26.30
166-170 6.11 7.97 10.74 481-490 14.54 19.44 26.76
n
171-175 6.22 8.12 10.97 491-500 14.78 19.77 27.24
io
176-180 6.34 8.30 11.20 501-510 15.66 20.96 28.87
181-185 6.47 8.45 11.44 511-120 15.91 21.28 29.33
at
186-190 6.59 8.61 11.66 521-530 16.13 21.59 29.77
191-195 6.71 8.79 11.90 531-540
ul 16.38 21.91 30.22
196-200 6.82 8.94 12.12 541-550 16.59 22.23 30.67
201-205 7.15 9.39 12.73 551-560 16.84 22.53 31.11
irc
206-210 7.28 9.57 12.98 561-570 10.07 22.86 31.56
211-215 7.40 9.73 13.21 571-580 17.31 23.18 31.99
C
216-220 7.53 9.91 13.45 581-590 17.53 23.50 32.45

221-225 7.65 10.07 13.68 591-600 17.77 23.81 32.90
d
226-230 7.78 10.24 13.92 601-610 18.59 24.93 34.44

231-235 7.91 10.41 14.15 611-620 18.84 25.26 34.90
ite
236-240 8.04 10.58 14.40 621-630 19.07 25.58 35.36

241-245 8.15 10.74 14.64 631-640 19.32 25.91 35.82
im
246-250 8.28 10.92 14.86 641-650 19.56 26.25 36.29

251-260 8.51 11.22 15.30 651-660 19.81 26.57 36.73
rL
261-270 8.73 11.54 15.75 661-670 20.04 26.90 37.20

271-280 8.97 11.86 16.19 671-680 20.28 27.23 37.64
281-290 9.19 12.18 16.63 681-690 20.52 27.54 38.11
Fo
291-300 9.42 12.48 17.06 691-700 20.77 27.90 38.57

301-310 9.81 12.99 17.79 701-710 21.08 28.34 39.19
311-320 10.05 13.32 18.23 711-720 21.34 28.67 39.65
321-330 10.27 13.64 18.67 721-730 21.57 28.99 40.12
331-340 10.51 13.95 19.12 731-740 21.81 29.32 40.58
(Continued)
(Continued)
Category 110 150 210 110 150 210

741-750 22.05 29.65 41.04 1651-1675 43.60 59.02 82.13
y
751-760 22.30 29.97 41.50 1676-1700 44.04 59.61 82.95
nl
761-770 22.54 30.31 41.97 1701-1725 44.65 60.43 84.13
771-780 22.78 30.65 42.44 1726-1750 45.10 61.04 84.96
O
781-790 23.03 30.98 42.89 1751-1775 45.52 61.62 85.76
791-800 23.26 31.31 43.36 1776-1800 45.95 62.21 86.57
801-825 23.98 32.26 44.69 1801-1825 46.35 62.74 87.34
n
826-850 24.57 33.09 45.85 1826-1850 46.77 63.34 88.16
io
851-875 25.18 33.92 47.01 1851-1875 47.21 63.91 88.97
876-900 25.80 34.76 48.18 1876-1900 47.65 64.50 89.81
at
901-925 26.52 35.71 49.53 1901-1925 47.95 64.92 90.42
926-950 27.11 36.55 50.69 ul 1926-1950 48.38 65.54 91.23
951-975 27.72 37.38 51.86 1951-1975 48.81 66.12 92.07
irc
976-1000 28.33 38.22 53.02 1976-2000 49.24 66.69 92.88
1001- 1025 28.96 39.06 54.22 2001-2025 49.38 66.90 93.15
1026- 1050 29.49 39.78 55.21 2026-2050 49.52 67.10 93.42
C
1051- 1075 29.99 40.46 56.18 2051-2075 49.91 67.60 94.14

1076- 1100 30.51 41.17 57.16 2076-2100 50.28 68.10 94.85
d
1101-1125 31.13 42.03 58.37 2101-2125 50.64 68.61 95.55

ite
1126-1150 31.66 42.75 59.36 2126-2150 51.01 69.13 96.27

1151-1175 32.17 43.44 60.35 2151-2175 51.39 69.62 96.98
1176-1200 32.69 44.13 61.32 2176-2200 51.75 70.13 97.67
im
1201-1225 33.83 45.70 63.50 2201-2225 52.12 70.64 98.39

1226-1250 34.38 46.42 64.50 2226-2250 52.50 71.15 99.11
rL
1251-1275 34.89 47.15 65.49 2251-2275 52.88 71.64 99.81

1276-1300 35.40 47.84 66.50 2276-2300 53.24 72.15 100.53
1301-1325 36.06 48.73 67.75 2301-2325 53.61 72.66 101.22
Fo
1326-1350 36.59 49.46 68.75 2326-2350 53.98 73.16 101.93

1351-1375 37.12 50.17 69.75 2351-2375 54.35 73.66 102.63
1376-1400 37.64 50.89 70.75 2376-2400 54.72 74.16 103.35
1401-1425 38.64 52.23 72.64 2401-2450 55.45 75.18 104.75
1426-1450 39.17 52.97 73.66 2451-2500 56.21 76.20 106.18
(Continued)
(Continued)
Category 110 150 210 110 150 210

1451-1475 39.71 53.69 74.68 2501-2550 56.84 77.07 107.40
y
1476-1500 40.23 54.42 75.69 2551-2600 57.49 77.95 108.62
nl
1501-1525 40.88 55.28 76.90 2601-2650 58.14 78.83 109.85
1526-1558 41.30 55.88 77.73 2651-2700 58.77 79.69 111.07
O
1551-1575 41.73 56.45 78.54 2701-2750 59.41 80.56 112.31
1576-1600 42.13 57.04 79.34 2751-2800 60.04 81.46 113.53
1601-1625 42.75 57.84 80.50 2801-2850 60.69 82.33 114.75
n
1626-1650 43.18 58.44 81.31 2851-2900 61.32 83.19 115.98
io
at
Exhibit 3: Shipping Charges and Times
Ship size (dwt)

ul 35000 65000
irc
Payload (tonnes) 34000 64000
Standing charges
(US $ per day) 9000 15000
C
(Rs per day) 153000 255000

One way voyage fuel costs to/from Alibag (Rs)
d
Mormugao
Loaded 120000 125000
ite
Ballast 105000 110000

Mangalore
im
Loaded 180000 187500

Ballast 157500 165000
Paradip
rL
Loaded 840000 875000

Ballast 735000 770000
Calcutta
Fo
Loaded 960000 1000000

Ballast 840000 880000
Fuel costs at anchor (Rs per day) 15000 17000
(Continued)
(Continued)
Port dues (Rs per entry)
Alibag 300000 412500

Mormugao 336000 462000
Mangalore 330000 453750
y
Paradip 380000 522500
nl
Calcutta 420000 577500
Unloading/loading charges (Rs per tonne)
O
Alibag 15
Mormugao 50
Mangalore FOBT
n
Paradip 40
io
Calcutta 45
One way average travel times to/from Alibag (days)
at
Mormugao 1.0 1.0
Mangalore 1.5 1.5
Paradip
Calcutta
ul 7.0
8.0
7.5
8.5
irc
Average waiting/loading period at ports (days)
Alibag (unloading) 8.0 11.0

C
Mormugao 4.5 5.5

Mangalore 3.0 3.5
Paradip 5.0 6.0
d
Calcutta 6.0 7.0

ite
Exhibit 4: Distance Matrix (kilometres)

im
Inbound road and rail distances between mines, ports and the plant
Goa Mangalore Daitari Banspani

rL
Alibag (Road) 520 800 1780 1650

Pen (Rail) 500* 700* 2200 1800
Mormugao (Rail) 50
Fo
Paradip (Rail) 170 700

400 (post 95)
* After the completion of Konkan Railway (1996)
(Continued)
(Continued)
Outbound distances to market centres
Town Pen Alibag Town Pen Alibag

(Rail) (Road) (Rail ) (Road)
y
Delhi 1420 1490 Baroda 440 510
nl
Hissar 1520 1590 Raipur 1140 1200
Ambala 1610 1680 Gwalior 1225 1170
Ludhiana 1695 1765 Indore 880 680
O
Jaipur* 1150 1250 Jabalpur 1000 1130
Lucknow 1400 1450 Bombay 70 85
Muzaffarnagar 1550 1620 Nagpur 830 940
n
Tatanagar 1700 2040 Aurangabad* 340 350
io
Bhilai 1280 1450 Kolhapur 540 350
Rourkela 1540 1900 Waltair 1520 1320
at
Durgapur 2000 2300 Secunderabad 810 680
Burnpur 1850 2250 Kothagudam 1060 980
Bokaro 1820 2200 ulBangalore 1140 970
Ranchi 1650 2000 Hospet 870 670
Barajamda 1600 1800 Bhadravati* 1390 780
irc
Calcutta 1970 2160 Calicut 1840 1170
Rajkot 780 850 Madras 1290 1310
Ahmedabad 540 630 Tiruchirapalli 1710 1300
C
Towns marked * have metre gauge only

d
ite
1. What is the best mode for inbound logistic supply of raw material?
2. What final markets should Laxmi Transformers choose to serve based on logistical
im
considerations? What is the best mode for the finished goods despatch to key
markets?
rL
3. Contingent on a particular choice for inbound movement, can the same mode be
used for servicing some final markets?
4. Is a rail siding from Pen to Alibag worth the investment?
Fo
5. If ships are to be used for inbound logistics, how many barges would be required for
the lighterage operations from the mother vessel to the jetty at Alibag?
6. Would the scheduling of carriers in any mode affect the economics of the mode
choice?
In such a situation, attention must be paid to the value chain of the product, which could
have a significant impact on inventory related costs. In particular, the inventory costs have
to be calculated keeping in mind the stage at which the material is in, since raw material and
y
finished goods have quite different values of holding and handling. The main quantities to
nl
be considered in inventory calculations for each mode are pipeline (or transit) inventories,
batching inventories and buffer stock inventories.
O
Each mode choice scenario has to be worked out with the relevant detail of all the
associated costs. The sea mode, in particular has a number of cost factors, since the volume
of movement is high enough that inventory costs while building up the loading quantity
n
and, while consuming the unloaded quantities have to be estimated. Also, barge costs have
io
to be accounted for, with a suitable level of detail.
Since the spatial spread of final markets for the firm is still uncertain, the logistics costs
at
due to a mode choice decision may affect the choice of locations for marketing efforts, to
some extent. In other words, while considering each mode of movement, the “optimal”
ul
choice of markets to serve (with appropriate assumptions about market share) can be
assumed. For this purpose, a segmentation of the final market can be done either on the
irc
basis of location of the customer, or on the basis of size of the customer’s demand, in order to
decide which markets to focus on. Recall that one of the motivations of locating the plant at
Alibag was to serve new mini-steel plants and ferrous foundries in Western India.
C
d
ite
im
rL
Fo
CASE 2
y
nl
The market share for cement transportation by rail had declined from 56% to 43% over
O
the past decade. In an attempt to arrest the sliding market share, the Indian Railways (IR)
wished to focus more on customer service by expanding the scope of its offer to a third party
n
logistics service provider. The client system was in the private sector (unlike the other major
commodities carried by IR) and hence was more demanding and representative of the future
io
orientation of industrial activity in India. Cement had to reach the retail customer, due to
which the distribution side had reasonable scope for a third party logistics service provider.
at
This case examines what the IR can do in the context of a large scale modern cement
plant, called Rajashree Cement (RC), in terms of operations at a specific client site. This had
ul
implications for infrastructure, services, organization and information systems. A new
concept called Engine on Load (EOL) was being experimented by RC and the IR to improve
irc
the throughput of bulk cement from the Malkhaid plant of RC to the Bangalore market.
After three months of EOL, executives from RC and IR were reviewing-(i) whether to
continue with the EOL system or not and (ii) if, to continue, what should be the terms of
C
the contract between RC and IR. This would have implications for IR’s service levels to their
clients.
d
ite
RAJASHREE CEMENT Engine On Load

im
In January 2004, Mr Rawat, General Manager, Malkhaid Plant; Mr Mehta, Vice President
– Logistics, Grasim Industries Limited (GIL); and Mr Chaddha, Logistics Manager,
rL
Malkhaid Plant, of Rajashree Cement (RC) and, Mr Gupta, Chief Operations Manager;
and Mr Murthy, Divisional Railway Manager, Secunderabad (SC) Division of South
Central Railway (SCR) of the Indian Railways (IR) were discussing (i) whether to continue
Fo
with the Engine on Load (EOL) system or not and (ii) if to continue, what should be the
Case prepared by G Raghuram, R Jain, (Indian Isntitute of Management, Ahemdabed, Indian.

380015), D Kumar and RN Prasad, (Railways Staff College, Vadodara). We acknowledge the research
assistance by Ameesh Dave. We are thankful to Grasim Industries and South Central Railway for the
data discussion support, and to the Indian Railways for financial support.
terms of the contract between RC and SCR. They felt that the recommendations would
have wider implications and should be considered by the Railway Board of the IR.
Background
Indian Railways
y
IR was a critical transport provider to the Indian economy, carrying about 40% freight ton
nl
kilometers and 20% of the country’s passenger kilometers. Over two-thirds of the revenue
came from freight operations. While freight was crucial to IR, and has been growing in an
O
absolute sense, it had steadily been losing market share to the road sector and more recently
to coastal transportation. The service offer from the IR had traditionally been station to
station transportation, and even in that, with a more explicit focus on the originating station.
n
In an attempt to arrest the sliding market share, the IR wished to focus more on customer
io
service by expanding the scope of its offer to a third party logistics service provider. Cement
was considered as one of the focus areas for this.
at
Significance of Cement to IR
ul
In 2001-02, seven commodities accounted for over 90% of the freight revenue. Coal was the
dominant commodity, followed by food grains, iron and steel, iron ore, cement, petroleum,
irc
oil and lubricants, and fertilizers [Indian Railways, 2003]. Salient aspects of cement
transportation by rail were—(i) the market share for cement transportation by rail had
declined from 59% in 1991-92 to 40% in 2002-03, (ii) the client system was in the private
C
sector (unlike the other major commodities carried by IR) and hence, was more demanding
and representative of the future orientation of industrial activity in India, and (iii) cement
had to reach the retail customer, due to which the distribution side had reasonable scope for
d
a third party logistics service provider.

ite
In 2002-03, cement accounted for about 9% (46 million tons (mt)) of the originating
traffic of IR. Nearly 24% (11 mt) of IR’s cement was loaded in SCR, 75% (8 mt) of which
was in the SC division. Cement accounted for 18% of freight loading in SCR and 25% in the
im
SC division. Exhibit 1 gives the cement and total freight loading statistics for SC division,
SCR and IR.
rL
Some of the concerns raised by the cement industry were non-availability of appropriate
wagons, lack of proper loading and unloading systems, seamless intermodal connectivity for
further movement in the supply chain, non uniform methods across railway zones in indents
and claims processing, and the inflexibility of IR to deal with smaller parcel sizes and
Fo
insufficient information on wagon availabilities and movements.
Structure of the Cement Industry

The installed capacity in large plants (above 0.2 mt per annum) was 140 mt, spread across
125 plants and 54 companies. Out of the 54 companies, the top 11 accounted for nearly
64% of the production in 2002-03. There were four companies having a market share of over
10% each. These were ACC, GACL, GIL, and L&T. The industry structure had been
Case 2: Rajashree Cement 285
undergoing a change in terms of consolidation, and fewer players were expected to control a
larger share of the market. The most recent acquisition had been that of L&T by GIL
towards the end of 2003. This would make the merged entity the largest cement player in the
country, with a market share of more than 20%. ACC and GACL had started working as an
alliance and were expected to merge soon.
y
There had been changes in the production structure of cement in a variety of ways. One of
nl
them involved making cement in two stages; first as clinker as an intermediate product close
to the raw material source and then transporting it to grinding plants where it was ground,
blended and bagged into value added cement for further distribution to the market. Another
O
change was to transport cement in bulk form for bagging near market centers. The bulk
cement could also be processed to a value added product as ready mix concrete for the
n
customer. The product variety was also increasing with the manufacture of blended cements.
Bulk movement of cement required special loading, unloading, and handling relative to
io
bagged cement, while offering advantages of lower transportation costs. The cement industry
was keen to look at integrated solutions for distribution logistics wherein a single player
at
could offer all the services for distribution from the plant to the customer.
GRASIM INDUSTRIES AND RAJASHREE CEMENT

ul
irc
GIL was a flagship company of the Aditya Birla Group. It ranked among India’s largest
private sector companies, (with a gross sales of Rs. 54.12 billion in 2002-03). Starting as a
textiles manufacturer in 1948, GIL’s businesses comprised Viscose Staple Fiber, cement,
C
sponge iron, chemicals and textiles in 2004. The company held a dominant position in
several of its businesses. GIL was the world’s eighth largest cement producer, and the largest
d
in a single location. Exhibit 2 provides the locations and capacities of the GIL cement units.
RC, established in 1983, was the largest, single location grey cement manufacturer in the
ite
Grasim Cement Division under GIL. It had a capacity of 4.2 metric tons per annum
production with a total manpower of 844. The RC plant was located at Malkhaid,
im
Karnataka. Exhibit 3 provides the details on the expansion in capacity from 1984 to 2003 at
RC.
rL
CEMENT CLUSTERS
Cement was manufactured in India in regional clusters due to the availability of limestone,
Fo
which constituted over 80% of the raw material by weight. Limestone, however, constituted
about 5-7% of the cost of sales. The RC plant was located in the Gulbarga cluster of
limestone mines.
The Gulbarga cluster spanned the states of Andhra Pradesh and Karnataka. Exhibit 4
provides the cluster-wise map showing the large plants, the state-wise installed capacities and
plants in the Gulbarga cluster. As of June 2003, the total installed capacity was 13 mt,
forming over 8% of the national installed capacity. ACC and RC were the major plants in the
cluster. RC had a 1.4 mt grinding unit at Hotgi in Maharashtra, about 150 kms from its
plant at Malkhaid. ACC had a grinding unit near Mumbai. Exhibit 5 provides the details of
rail connectivity for the Gulbarga cluster.
LOGISTICS AND DISTRIBUTION
y
RC served the Maharashtra dealers from their grinding unit at Hotgi. It served Karnataka
nl
dealers from Malkhaid, both in bagged and bulk forms. About 0.6 mt (52,000 tons per
month) was sent in bulk by rail to service the Bangalore market to a bagging plant at
O
Dodballapur, near Bangalore. For Andhra Pradesh, it had four depots spread across the state
for stock transfer and secondary distribution to the dealers. Some bulk cement was also sent
by road for further value addition to a ready mix concrete plant near Hyderabad.
n
For the bulk movement, since specialized handling was a requirement, special purpose
io
wagons were necessary. IR encouraged customers having such requirements to invest in their
own wagons and offered a 22.5% freight subsidy under the Own Your Wagon (OYW)
at
scheme. Depending on the utilization, the returns on investment could be attractive to
customers like RC. RC had invested Rs 600 million on the wagons (amounting to three rakes
ul
a rake is a set of wagons constituting a train) and five wagons and the loading and unloading
facilities. The freight subsidy for the current level of cement movement was Rs 73 million,
irc
providing a return of about 12%. Exhibit 6 provides the details of economics of the OYW
scheme.
C
BULK CEMENT CLOSED CIRCUIT MOVEMENT

d
For supplying bulk cement to Dodballapur, there was a closed circuit movement of rakes
from Malkhaid to Dodballapur and back. The three rakes, averaging a turnaround of 100
ite
hours, led to about seven trips per month per rake. This resulted in a total of over 21 trips per
month. Since each rake could carry 2,400 tons, the supply at Dodballapur was as required. A
graphical view of the closed circuit movement, layout at Malkhaid and Dodballapur, and
im
operating details of the crew change, fuelling, engine change and wagon maintenance are
given in Exhibit 7.
rL
Since IR viewed the engine as a critical resource, after the engine pulled in the rake at RC’s
premises, it was allocated to other tasks, while the incoming rake was loaded at RC.
Consequently, there was a wait for engine after the loading was completed at RC, since the
SC division had to organize for it. Exhibit 8 gives the waiting time for engine at Malkhaid.
Fo
GROWING MARKET
The market for bulk cement in and around Bangalore was growing. RC felt that there was an
opportunity to meet this demand. RC wanted to improve the supply to Dodballapur to
70,000 tons per month. In consultation with SCR, the following alternatives were
considered:
✧ Increase throughput with the same number of trips by increasing load per wagon
and/or wagons per rake. Both these would require redesign of the wagon (due to the
topology over which the journey had to be made) and acquisition of fresh stock.
✧ Increase the number of trips with additional rakes. This would require acquisition of
fresh stock, for which one possibility was modifying and using oil tankers that had
y
been rendered surplus due to increased pipeline transportation)
nl
✧ Increase the number of trips with the same rakes by improving the turnaround time.
The first two alternatives required additional investments by RC. Recognising the rather
O
significant waiting time for engine at Malkhaid, RC proposed that the third alternative be
considered. SCR agreed to take this up.
n
ENGINE ON LOAD EXPERIMENT
io
IR had introduced an Engine on Load (EOL) scheme, wherein the engine would be kept
at
attached with a rake during loading and unloading operations, thus minimising the delays
caused due to loaded/unloaded rakes waiting for it. Such delays led to loss of throughput and
ul
consequent loss of revenues to IR. The relevant Railway Board Circular is given in Exhibit 9.
RC and SCR decided to experiment with improving the turnaround time by adopting the
irc
EOL scheme. The economics of the EOL for the closed circuit movement, as perceived by
RC, are given in Exhibit 10. The approval by SCR for this scheme is given in Exhibit 11. The
EOL was inaugurated on 15th September, 2003 (Exhibit 12).
C
For the EOL, it had been mutually agreed that three hours would be the total time in
which RC would complete the loading of the incoming rake. (The three hours was the lower
d
bound of the time taken for an engine to pick up a task subsequent to the one just completed,
during the normal assignment process). An hourly penalty of Rs 3,800 would accrue if this
ite
was exceeded. This can be viewed in the context of the two types of engines that were used.
The indigenous engines in a twin set cost Rs 80 million and a recent import cost Rs 140
million. At a 10 per cent cost of capital, the engine cost per hour was about Rs 900 and
im
Rs 1,600 respectively.
In December 2003, RC and SCR wanted to review the EOL experiment. When a team
rL
from the Indian Institute of Management, Ahmedabad and Railway Staff College, Vadodara
visited RC as a part of the study on the role of IR as a third party logistics service provider for
the cement industry, RC proposed that the team be involved in the review. The team
members agreed and gathered data on the closed circuit movement performance till the end
Fo
of December, 2003. 78 trips had been undertaken from Malkhaid after the introduction of
EOL, of which only 45 actually availed of EOL. Exhibits 13 and 14 give the rake-wise
detention details gathered by the team at Malkhaid and Dodballapur respectively. It was now
essential to calculate and analyse the components of the closed circuit movement, both for
the rakes which availed EOL and those that did not, to facilitate the review. The components
of the turnaround time before the EOL introduction is given in Exhibit 15.
Based on the experience of the EOL scheme, the team suggested that RC and SCR
consider the following issues:
✧ Should RC and SCR have the option of declaring a trip as not for EOL? Can this be
unilaterally decided or should it be a joint decision? What should be the required
advance notice?
y
✧ From the time the rake reaches the terminal, what should be the guaranteed time for
nl
engine availability? What should the penalty be on RC if load is not available? What
should the penalty be on SCR if engine is not available? (Currently, there was no
O
penalty on SCR).
✧ What steps should RC take to ensure timely loading at the terminal?
n
✧ What steps should SCR take to ensure EOL, and by implication powering the
incoming train with an appropriate engine?
io
✧ Is the issue EOL or timely availability of engine (incoming and outgoing engines
need not be the same, should SCR want the flexibility)?
at
✧ To improve customer service towards becoming a third party logistics service
provider, should not SCR (and thus, IR) focus on guaranteed total (a) delivery time
ul
from Malkhaid (origin) to Dodballapur (destination) and (b) turnaround time for
the whole circuit?
irc
C
d
ite
im
rL
Fo
y
nl
O
Exhibit 1: Cement and Total Originating Loading Statistics
Secunderabad Division South Central Railway Indian Railways
n
io
Year Cement Total % Variation Over Cement Total Cement Total
Previous Year
at
Mt Average mt Average mt Average mt Average mt Average mt mt
per Wagons per Wagons per Wagons per Wagons per Wagons per annum per annum
annum per day annum per day annum per day annum per day annum per day
ul
1990-91 5.76 703 17.32 2175 4.4 4.9 8.19 912 31.78 3503
irc
1991-92 5.38 661 18.88 2325 9.0 6.9 8.60 929 34.78 3723
1992-93 5.92 723 20.67 2593 9.5 11.5 8.63 931 35.83 3937
1993-94 6.78 801 24.23 3073 17.2 18.5 9.60 1027 40.30 4570
Case 2: Rajashree Cement

C
1994-95 5.55 786 20.51 2970 15.3 3.4 8.18 1042 35.73 4766
1995-96 6.28 757 21.51 2521 4.9 15.1 9.33 1059 36.68 4267
d
1996-97 7.00 767 24.09 2849 12.0 13.0 9.89 1069 40.30 4581
ite
1997-98 7.90 772 26.17 3056 8.6 7.3 10.42 1053 44.11 5021 37.36 429.38
1998-99 6.70 664 24.25 2824 7.3 7.6 9.65 989 42.45 4920 36.75 420.92
1999-00 7.22 705 28.73 3298 18.5 16.8 11.37 1184 51.25 5917 43.62 456.42
im
2000-01 7.05 827 30.41 3517 5.8 6.6 10.76 1093 56.84 6604 42.90 473.50
2001-02 6.56 770 31.31 3619 3.0 2.9 9.70 1137 60.41 7010 44.04 492.50
rL
2002-03 8.34 975 33.41 3843 6.7 6.2 11.18 1305 62.29 7222 46.25 518.74
Source: [South Central Railway, 2004 and Indian Railways, 2003a]
289
Fo
Exhibit 2: Grasim Industries (AV Birla Group): Cement Units
Plant Location State Capacity (mt)
Rajashree Cement Malkhaid Karnataka 4.20
y
Birla Super Grinding Unit Hotgi Maharashtra 1.40
nl
Aditya Cement Shambupura Rajasthan 1.75
Vikram Cement Khor M.P. 3.00
O
Grasim Cement Rawan Chhattisgarh 2.06
Grasim Cement South Reddypalayam Tamil Nadu 1.03
Bhatinda Grinding Unit Bhatinda Punjab 1.20
n
SDCC Sikka Jamnagar 1.08
io
Source: [RC, 2003]
at
Exhibit 3:
ul
Rajashree Cement – Malkhaid Plant Capacity
irc
Plant Year of Capacity (mt) Cumulative Capacity
Commissioning (mt)
C
Unit-I 1984 0.50 0.50

Unit-I Expansion 1989 0.30 0.80
Unit-II 1990 1.00 1.80
d
Unit-III 1995 1.20 3.00

ite
Unit-II Internal Optimization 2000 0.20 3.20

Capacity Enhancement By Debottlenecking June 2003 1.00 4.20
im
Source: [RC, 2003]

rL
Fo
Exhibit 4: Clusterwise Cement Map of India (Large Plants)

(as on 31.3.2003)
y
nl
O
n
io
at
ul
irc
Capacity Shown Million Tonnes
Sl. No. of Ind. Sl. No. of Ind.

C
No. State Plant Dept. No. State Plant Dept.
01 Andhra Pradesh 21 20.73 12 West-Bengal 4 43.13

02 Rajasthan 14 17.72 13 Panjab 3 42.04
d
03 Madhya Pradesh 10 16.19 14 Urissa 3 43.04

04 Gujarat 10 14.38 15 Bihar 1 41.10
ite
05 Tamil Nadu 13 14.56 16 Delhi 1 40.50

06 Chandigarh 5 10.36 17 Kerala 1 40.42
07 Maharashtra 5 10.35 18 Jammu & Kashmir 1 40.20
im
08 Karnataka 5 19.79 19 Assam 1 40.20

09 Uttar Pradesh 7 15.06 20 Meghalaya 1 40.20
10 Jharkhand 5 14.57 21 Hariyana 1 40.17
11 Himanchal Pradesh 3 14.06 Total 124 43.13
rL
Gulbarga Cluster
Fo
Sl. No. Region/State Location Installed Capacity (mt)
Andhra Pradesh
53. CCI Ltd. Tandur 1.00
58 India Cement-Visaka Tandur 1.12
2.12
(Continued)
(Continued)
Sl. No. Region/State Location Installed Capacity (mt)
Karnataka
82 ACC Ltd. Wadi 2.11
y
83 ACC Ltd.- New Wadi 2.60
84 Vasavadatta Cement Sedam 1.20
nl
85 Rajashree Cement Malkhaid 4.20
O
87 CCI Ltd. Kurkunta 0.20
88 HMP Cements Ltd. Shahabad 0.48
10.79
n
Source: [CMA, 2003]
io
at
Exhibit 5: Rail Connectivity for the Gulbarga Cluster
ul
irc
C
d
ite
im
rL
Fo
Source: [Indian Railways, 2003b]

Exhibit 6: Economics of Own Your Wagon Scheme for

Rajashree Cement
Investments
y
Wagons (40 *3 Rakes plus 5 for Rs 200 million
nl
contingency = 125 Boxes)
Malkhaid (Loading) Silo Rs 10 million
O
Dodballapur Unloading Rs 390 million
Total Investment Rs 600 million
n
Benefits
io
Freight Subsidy 22.5% for 10 years
(Rs 116 on a total of Rs 516)
at
7 trips per rake per month @2400 52,000 tons per month = 624,000 tons
mt per rake per annum
Savings 624,000*116 = Rs 73 million
ul
ROI 73/600 = 12 %
irc
Alternatives
Net savings: Rs 116 or Rs 150 per - Rail OYW Rs 460: Rs 400

ton? (Malkhaid to Dodballapur) + Rs 60
C
(Dodballapur to dealers)
- Rail regular wagon Rs 610: Rs 550
(Malkhaid to Bangalore in bags) +
d
Rs 60 (Bangalore handling at goods shed

ite
and then to dealers)

Source: [RC, 2003 and Authors Analysis]
im
rL
Fo
Exhibit 7: RC Bulk Cement Closed Circuit Movement
y
nl
O
n
io
at
ul
irc
Malkhaid Layout
C
d
ite
im
rL
Dodballapur Layout
Fo
y
nl
O
Crew Change : Chittapur, Guntakal, Dharmavaram and Baiyappanahalli
Engine Change : Guntakal and Gooty (for maintenance, if required) and Dharmavaram (for
n
engine compatibility, if required). Three types of engines were used from
io
Malkhaid: WDM2 Multi (about 5000 horse power, 16, 17 and 18 series),
WDG3A
at
Shaktiman (3100 horse power, 13 and 14 series) and WDG4 GM (4000 horse
power, 12 series). The WDG3A had difficulty in managing the gradient in a few
ul
sections in GTL division. It could stall if it was slowed down. A banker/double
head/multi would be preferred. The WDG3A required a banker on the ghats in
irc
the Dharmavaram – Dodballapur section for one block section. The WDG4 were
not safety certified for the ghat sections.
Fuelling : Vikarabad, Guntakal, Gooty and Baiyappanahalli, as required. A top up to 5000
C
litres (for WDM2) and 6000 litres (for WDGs) at Gooty would enable a trip to
either Malkhaid or Dodballapur and back
d
Wagon Maintenance : Gooty (The maintenance provided a break power certificate for 4000 kms,
implying that the next maintenance would be required before this distance was
ite
covered. This implied a maintenance once in three trips.)

SCR – SWR Issues : SCR suggested that SWR should conduct trials using WDG3A without a banker
im
on the ghat section. Alternately, SCR was ready to provide a banker to operate
only in the ghat section, if SWR agreed. This would avoid the need for engine
change at Dharmavaram.
rL
: SWR suggested that if the engine from Dharmavaram came with enough fuel, the
same can be turned around at Dodballapur. However, the crew would need rest
during the unloading operations, for which a running room was being
Fo
constructed by RC at Dodballapur
Source: [RC, 2003 and Authors Analysis]
Exhibit 8: Average Waiting Time for Engine to Pull Out Loaded Rakes
(Hours:Minutes)
Apr’02 - Mar’02
y
Month Average Time Taken for Arrival of Engine
nl
Cement Rake Clinker Rake
O
Bagged Loose
Apr02 12:45 8:00 5:50

May02 20:30 13:50 7:50
n
Jun02 31:40 13:00 8:50
io
Jul02 17:20 9:20 6:30
Aug02 7:00 29:30 5:40
at
Sep02 10:50 13:30 6:00
Oct02 ul 21:00 11:00 10:20
Nov02 13:45 14:10 9:50
irc
Dec02 24:10 10:50 9:30
Jan03 19:15 13:30 11:15
Feb03 18:45 12:23 9:40
C
Mar03 16:06 9:42 8:31

Total 213:06 158:46 99:46
d
Average 02-03 17:45 13:13 8:18

ite
Minimum time 7:00 8:00 5:40

Maximum time 31:40 29:30 11:15
im
Apr’03 - Nov’03
rL
Month Average Time Taken for Arrival of Engine
Cement Rake Clinker Rake
Bagged Loose
Fo
Apr03 30:38 6:48 11:27

May03 29:10 9:52 21:41
Jun03 15:06 12:20 19:03
Jul03 28:56 12:53 9:15
Aug03 23:49 12:38 11:43
(Continued)
(Continued)
Sep03 14:39 6:24 12:02
Oct03 17:07 3:06 12:06
Nov03 15:15 6:00 14:00
Total 174:43 70:04 111:19
y
Average 03-04 21:50 8:45 13:54
nl
Minimum time 14:39 3:06 9:15
Maximum time 30:38 12:53 21:41
O
Source: [RC, 2003]
n
io
Exhibit 9: Railway Board Circular
at
GOVERNMENT OF INDIA
MINISTRY OF RAILWAYS
ul
(RAILWAY BOARD)
irc
N0: TC-I/94/214/9 New Delhi, dt 26-6-1997
The General Manager,

C
All Indian Railways

Sub: Provision of Engine on Load Scheme at selected terminals and
sidings.
d
1. For quite some time the question of reducing the detention caused to the wagons at
ite
the terminals/siding because of their waiting for Power has been engaging the
attention of the Board. It has also been observed that such detentions to wagons
caused heavy loss of loading capacity as also loss of revenue to the Railways.
im
2. On the basis of extensive review, it has been observed that detention to rakes could be
minimised substantially if the locomotive is kept attached with a rake during loading
rL
and unloading operations. The moment the loading/unloading is completed, the rake
could be moved out without much loss of time; thereby improving the turn round of
wagons resulting in generation of additional carrying capacity. This system, also known
Fo
as Engine on Load is already operational on Eastern Railway (Khalari-Unchahar

Circuit), South-Eastern Railway (Bacheli Yard on Kirandul-Kottavalasa Line) and
Northern Railway (NTPC siding at Dadri). The studies conducted on operation of
existing Engine on Load system, which is based on the debit/credit hours for calculation
of free time have shown remarkable results in that it has led to a substantial reduction in
detentions and better turn round of wagons. This scheme, however, did not make much
headway at other places because the terminal/siding owner did not show much interest
in improving the handling facilities and infrastructure as they were not getting any
direct monetary advantage and perceived that this scheme benefited only the
Railways.
3. Keeping this in view, the basic features of the scheme are being outlined in a Draft
Memorandum of Understanding, which is under finalisation in Board’s office. The
proposed features are enumerated as under:
y
(i) This scheme is open to selected terminals/sidings, handling atleast one rake per day,
nl
who are desirous of making investments for improvement of siding infrastructure
and handling facilities for reduction in time taken for rake loading/unloading
O
operations.
(ii) The investments in sidings will precede the grant of monetary incentive i.e. the
installation and commissioning of improved terminal handling facilities and
n
resultant saving in detention of rakes is a must for payment of these incentives.
io
(iii) The terminal/siding user will submit technical plans with capital cost, method of
financing and period during which the facilities will be installed, to the Railways for
at
vetting and ascertaining the adequacy of the scheme etc.
(iv) The revised free time for loading/unloading would be four hours or less depending
ul
on the facilities and the special features of the siding.
(v) The monetary incentive will be 50% of the notional earnings accruing to the
irc
Railways on account of reduction in the loading/unloading time of the rake. These
notional earnings would be calculated after deducting the cost of detention of
locomotives as a result of the implementation of the scheme.
C
(vi) The incentive will be paid for a limited period of upto 3 years, commensurate with
the investments made.
d
(vii) After the period of recovery of investment, the rebate would be replaced by a
ite
nominal concession to the extent of the expenses on maintenance of the facilities

plus a token 5% thereof as long term incentive, subject to a ceiling of 2-3% of the
annual freight charges.
im
(viii) In case the party defaults and savings do not accrue, then normal demurrage charges
on rakes as also loco detention charges would be recoverable from the party.
rL
(ix) Monetary incentive will be paid on half-yearly basis.

(x) The monetary incentive is towards investments made only for infrastructure
improvements and provision of other facilities in the sidings, leading to reduction in
Fo
time taken for loading/unloading.

(xi) The scheme will take effect from a prospective date only.
4. The draft MoU is under finalisation.
5. In the meantime, Zonal Railways may address prospective rail-users who can opt for the
scheme and explain the provisions to them so that it is appreciated in the correct
perspective.
6. Any suggestions from Zonal Railways to further improve and strengthen the scheme,
may be sent alongwith comments of associate Finance.
7. This issues with the concurrence of the Finance Department in the Ministry of
Railways.
(Hindi version will follow)
y
(K.K. Sharma)
nl
Jt. Director, Traffic Comml. (Rates) I
O
Railway Board.
Copy to: (1) General Manager (Commercial) and General Manager (Optg.), All Indian Railways
including Konkan Railway.
n
(2) EDTT (M), EDCE (G), EDF(C), DDTC(R), Railway Board.
io
at
Exhibit 10: Economics of EOL
Preamble
ul
irc
The following table gives an activity wise breakup of the time taken in the turnaround cycle
for bulk rakes (Inclusive of both historical data and Proposed Time for each activity with
EOL system)
C
Time in Hours
Particulars 2002-03 2003-04 Proposed
d
with EOL
ite
Apr 02-Mar 03 Apr 03-Jul 03
Loading Time 3.36 3.27 3.3

Placement/Shunting/Formation(*) Time 7.36 3.11 1.3
im
Total Time for loading and placement 11.12 6.38 5

Time taken for engine arrival to pullout loaded rake 13.14 16.42 2 (***)
rL
Transit time Malkhaid to Dodballapur 34.42 53.4(**) 34

Unloading time at Dodballapur 9.46 25.54(**) 9
Transit time Dodballapur to Malkhaid 30.09 29.13(**) 30
Fo
Total Bulk Rake turnaround time 99.03 132.07 (**) 80

Average no. of trips per month 22.11 16.58 27.38
Actual average number of trips 22 20 -
(*) Also inclusive of detention of empty rake for want of stock in silos
(**) The high values are due to detention of loaded rakes due to Transport Strike in April and maintenance at Dodballapur in June
(***) Time taken as 2 hours for any contingency , ideally should be zero
(Continued)
(Continued)
Expected Benefits
Effective increase in number of trips per month 5

Incremental quantity moved per month (tons) 11900
Incremental quantity moved per annum (tons) 142800
y
Freight rebate @ 22.5 % (Rs/ton) 105.52
nl
Logistics Gain per month (Rs million) 1.26
O
Cost envisaged towards engine detention of two hours @ Rs 3900
per hour (Rs million) per month 0.21
Net Logistic Gain per month (Rs million) 1.05
n
Net Annual Logistic Gain (Rs million) 12.5
io
Gain to Railways
Additional revenue per month due to incremental
at
quantity (Rs million) 4.33
Additional revenue annually due to incremental quantity
(Rs million)
Detention Charges per month (Rs million)
ul 51.9
0.21
irc
Detention Charges annually (Rs million) 2.5
Total Gain to Railways (Rs in million) 54.4
C
Source: [RC, 2003]

d
ite
im
rL
Fo
Exhibit 11: Approval by South Central Railway

South Central Railway
Office of the Divisional Railway Manager,
Commercial Branch,
y
Sanchalan Bhavan – II floor,
Secunderabad.
nl
C/C/490/Shunting Charges/96 Date: 18-9-03
Vice President,
O
M/s Rajashree Cements,
May Fair Complex, S.P.Road,
Secunderabad.
n
Dear Sir,
io
Sub: Your request for detaining engine on load for loading BCCW rake.
Ref: Your Ir No. RC/MQR/CEM/EOL dated 06-8-03
at
Your request for loading BCCW wagons with engine on load, keeping with Railway
engine is agreed. Necessary instructions have been issued to station staff to permit to keep the
ul
engine on request for loading BCCW wagons.
It may be noted that detention charges have to be paid at prescribed rate for detention of
irc
engine beyond three hours, from the time of arrival of engine with empty rake into the
siding, till the departure of engine with load from siding.
C
Yours Sincerely,
For Divisional Railway Manager,
d
Secunderabad
ite
Copy to:
CCM for kind information
COM for kind information
im
Sr. DOM/SC for information and necessary action.

CCI/TDU for necessary action
rL
SM/MQR & Siding clerks for recording and levying engine detention charges beyond
three hours as stated above.
Fo
Exhibit 12: Inauguration of EOL (15.09.2003)
y
nl
O
n
io
at
ul
irc
Source: [RC, 2003]
C
d
ite
im
rL
Fo
y
nl
Exhibit 13: Rakewise Detention Details at Malkhaid
O
September 2003
n
S Ra- Date Stock Arriving Engine Placement Release Departure Total Free Time Rate Amo- Remarks
No ke at Siding No. Time Time from Siding Dete- Time Charge per unt
io
ntion Hour
Date Time Date Time Date Time Date Time
at
Hrs Hrs Hrs Rs Rs
1 A 15.09.03 40 Bccw 15.09.03 2:00 14615 15.09.03 2:30 15.09.03 7:30 15.09.03 8:00 6:00 3:00 3:00 3870 11610
2 B 15.09.03 40 Bccw 15.09.03 16:30 14630/14577 15.09.03 17:00 15.09.03 23:00 15.09.03 23:30 Engine withdrawn no
ul
EO L Cleared with 12019
3 C 15.09.03 40 Bccw 15.09.03 22:30 18513/18897 16.09.03 12:00 16.09.03 17:00 16.09.03 19:30 Party requested
irc
since no time gap
4 A 17.09.03 40 Bccw 17.09.03 17:00 14559 17.09.03 18:00 17.09.03 23:30 18.09.03 0:00 Enginewithdrawn anot-
her power 14818 cleared

C
5 B 18.09.03 40 Bccw 18.09.03 20:30 12006 19.09.03 5:00 19.09.03 8:00 19.09.03 12:00 Engine withdrawn at party
requestsince no material
6 C 19.09.03 40 Bccw 19.09.03 16:00 14678 19.09.03 20:00 19.09.03 23:00 20.09.03 7:00 No EOL
d
7 A 20.09.03 40 Bccw 20.09.03 8:00 14818 21.09.03 3:00 21.09.03 7:00 21.09.03 11:00 No EOL
8 B 23.09.03 40 Bccw 23.09.03 22:00 12020 23.09.03 22:30 24.09.03 4:30 24.09.03 7:00 Withdrawn engine
ite
cleared with 14693
9 C 24.09.03 40 Bccw 24.09.03 1:00 14639 25.09.03 4:00 25.09.03 8:00 25.09.03 8:30 Cleared with 14819
10 A 25.09.03 40 Bccw 25.09.03 8:15 14819 25.09.03 13:00 25.09.03 17:00 25.09.03 18:00 Cleared with 12030
im
11 B 26.09.03 40 Bccw 26.09.03 19:30 14639 26.09.03 20:00 27.09.03 1:00 27.09.03 2:00 6:30 3:00 3:30 3870 15480
12 C 27.09.03 40 Bccw 27.09.03 20:00 14819 28.09.03 3:00 28.09.03 7:00 28.09.03 11:00 Cleared with 14678
13 A 28.09.03 40 Bccw 28.09.03 20:00 12039 29.09.03 2:00 29.09.03 7:00 29.09.03 11:00 Cleared with 14917/
rL
14938
14 B 30.09.03 40 Bccw 30.09.03 3:00 17110 30.09.03 4:00 30.09.03 9:00 30.09.03 10:00 Cleared with 12029
Total 27090
303
Fo
y
304
nl
October 2003
O
ntion Hour

n
Hrs Hrs Hrs Rs Rs
io
1 C 30.09.03 40 Bccw 30.09.03 23:30 14678 01.10.03 1:00 01.10.03 4:00 01.10.03 4:3 5:00 3:00 2:00 3870 7740 EOL
2 A 02.10.03 40 Bccw 02.10.03 11:30 12006 02.10.03 12:00 02.10.03 16:00 02.10.03 16:30 5:00 3:00 2:00 3870 7740 EOL
at
3 B 03.10.03 40 Bccw 03.10.03 19:30 14813 03.10.03 20:00 04.10.03 4:00 04.10.03 4:30 9:00 3:00 6:00 3870 23220 E O L
4 C 04.10.03 40 Bccw 04.10.03 13:30 14678 04.10.03 14:30 04.10.03 18:30 04.10.03 19:30 6:00 3:00 3:00 3870 11610 E O L
ul
5 A 06.10.03 40 Bccw 06.10.03 1:30 12015 06.10.03 2:00 06.10.03 7:30 06.10.03 8:00 7:00 3:00 4:00 3870 15480 E O L Power Interception
with 18897/16499
6 B 06.10.03 40 Bccw 06.10.03 17:00 14813 06.10.03 17:30 07.10.03 1:30 07.10.03 2:00 9:00 3:00 6:00 3870 23220 E O L
irc
7 C 09.10.03 40 Bccw 09.10.03 8:30 14818 09.10.03 9:00 09.10.03 15:00 09.10.03 15:30 7:00 3:00 4:00 3870 15480 E O L
8 A 10.10.03 40 Bccw 10.10.03 7:30 12025 10.10.03 8:00 10.10.03 12:00 10.10.03 12:30 5:00 3:00 2:00 3870 7740 EOL
C
Subtotal 53:00 24:00 29:00 3870 112230
9 B 11.10.03 40 Bccw 11.10.03 4:30 12004 11.10.03 5:00 11.10.03 10:30 11.10.03 13:00 12004 with drawn cleared
with 13039
d
10 C 12.10.03 40 Bccw 12.10.03 19:00 16696/702 12.10.03 19:30 13.10.03 1:30 13.10.03 2:00 7:00 3:00 4:00 3870 15480 E O L
11 A 15.10.03 40 Bccw 15.10.03 2:30 16705/703 15.10.03 3:00 15.10.03 7:00 15.10.03 10:10 16705/703 with drawn
ite
cleared with 13039
12 B 16.10.03 40 Bccw 16.10.03 4:00 12026 16.10.03 4:30 16.10.03 9:30 16.10.03 10:00 6:00 3:00 3:00 3870 11610 E O L
13 C 16.10.03 40 Bccw 16.10.03 15:00 12026 16.10.03 21:00 17.10.03 0:00 17.10.03 6:45 Partys request since no
im
time gap cleared with

14691
14 A 18.10.03 40 Bccw 18.10.03 13:30 14092 18.10.03 14:00 18.10.03 19:00 18.10.03 19:30 6:00 3:00 3:00 3870 11610 E O L
rL
15 B 19.10.03 40 Bccw 19.10.03 7:00 14999 19.10.03 7:30 19.10.03 10:30 19.10.03 11:00 4:00 3:00 1:00 3870 3870 EOL
16 C 19.10.03 40 Bccw 19.10.03 20:30 14660 19.10.03 21:00 20.10.03 2:00 20.10.03 2:30 6:00 3:00 3:00 3870 11610 E O L
Subtotal 29:00 15:00 14:00 3870 54180

Fo
(Continued)
y
nl
O
(Continued)
n
ntion Hour
io
Hrs Hrs Hrs Rs Rs
17 A 21.10.03 40 Bccw 21.10.03 14:30 13037 21.10.03 15:00 21.10.03 21:00 21.10.03 21:30 7:00 3:00 4:00 3870 15480 E O L
at
18 B 22.10.03 40 Bccw 22.10.03 20:30 14813 22.10.03 21:00 23.10.03 2:00 23.10.03 2:30 6:00 3:00 3:00 3870 11610 E O L
19 C 25.10.03 40 Bccw 25.10.03 1:30 13044 25.10.03 3:00 25.10.03 8:00 25.10.03 13:30 No E O L cleared with
14667, Loco no. 13044
ul
not permitted due to loco
trouble
20 A 26.10.03 40 Bccw 26.10.03 5:30 17228/16474 26.10.03 6:00 26.10.03 11:00 26.10.03 11:30 6:00 3:00 3:00 3870 11610 E O L
irc
21 B 28.10.03 40 Bccw 28.10.03 5:30 14589/14689 28.10.03 6:00 28.10.03 13:00 28.10.03 13:30 8:00 3:00 5:00 3870 19350 E O L
22 C 29.10.03 40 Bccw 29.10.03 16:00 13037 29.10.03 16:30 29.10.03 23:00 30.10.03 0:50 As per T N No. 1/41 with

drawn went for fueling
C
23 A 30.10.03 40 Bccw 30.10.03 18:30 16017 30.10.03 19:00 30.10.03 0:00 31.10.03 1:15 No E O L cleared with
14613
24 B 31.10.03 40 Bccw 31.10.03 15:30 12023 31.10.03 16:00 31.10.03 20:00 31.10.03 20:30 5:00 3:00 2:00 3870 7740 EOL
d
Subtotal 32:00 15:00 17:00 3870 65790
ite
im
rL
305
Fo
y
306
nl
November 2003
O
ntion Hour

n
Hrs Hrs Hrs Rs Rs
io
1 C 01.11.03 40 Bccw 01.11.03 13:30 13039 01.1103 14:00 01.11.03 18:00 01.11.03 18:30 5:00 3:00 2:00 3870 7740 EOL
2 A 02.11.03 40 Bccw 02.11.03 7:00 14613 02.11.03 7:30 0211.03 11:30 02.11.03 12:00 5:00 3:00 2:00 3870 7740 EOL
3 B 04.11.03 40 Bccw 04.11.03 23:00 14613 04.11.03 23:30 05.11.03 5:30 05.11.03 6:00 7:00 3:00 4:00 3870 15480 EOL
at
4 C 06.11.03 40 Bccw 06.11.03 12:00 12029 06.11.03 12:30 06.11.03 16:30 06.11.03 17:00 5:00 3:00 2:00 3870 7740 EOL
5 A 07.11.03 40 Bccw 07.11.03 23:00 12030 07.11.03 23:3 08.11.03 3:30 08.11.03 4:00 5:00 3:00 2:00 3870 7740 EOL
ul
6 B 08.11.03 40 Bccw 08.11.03 23:30 12001 09.11.03 0:00 0911.03 4:00 09.11.03 4:30 5:00 3:00 2:00 3870 7740 EOL
7 C 09.11.03 40 Bccw 09.11.03 23:30 14943/572 10.11.03 0:00 10.11.03 8:00 10.11.03 8:30 9:00 3:00 6:00 3870 23220 EOL
irc
Subtotal 41:00 21:00 20:00 3870 77400
8 A 10.11.03 40 Bccw 10.11.03 23:30 14927/13007 11.11.03 0:00 11.11.03 5:00 11.11.03 5:30 6:00 3:00 3:00 3870 11610 EOL
9 B 12.11.03 40 Bccw 12.11.03 19:00 12024 20.11.03 16:00 20.11.03 20:00 21.11.03 16:05 On partys request No
C
EOL Since Mech break
down at doballapur
w/indent
d
Subtotal 6:00 3:00 3:00 3870 11610
ite
(Continued)
im
rL
Fo
y
nl
(Continued)
O
ntion Hour
n
Hrs Hrs Hrs Rs Rs
io
10 C 14.11.03 40 Bccw 14.11.03 7:00 14572/943 21.11.03 8:00 21.11.03 12:00 22.11.03 0:30 On partys request No
EOL Since Mech break
down at doballapur
at
w/Indent
11 A 21.11.03 40 Bccw 21.1103 23:30 22.11.03 9:00 22.11.03 12:00 22.11.03 21:30 No EOL cleared with
14902/17178
ul
12 B 24.1103 40 Bccw 24.11.03 18:45 17673 24.11.03 22:00 25.11.03 2:00 25.11.03 5:40 No EOL cleared with
14813
irc
13 C 25.11.03 40 Bccw 25.11.03 12:00 16602/17744 25.11.03 19:00 25.11.03 24:00 26.11.03 3:00 No EOL cleared with
14816
14 A 26.11.03 40 Bccw 26.11.03 1:00 14904 26.11.03 9:00 26.11.03 12:00 26.11.03 18:00 No EOL cleared with

C
14684/697
15 B 27.11.03 40 Bccw 27.11.03 14:00 14813 2711.03 14:30 27.11.03 19:30 27.11.03 20:00 6:00 3:00 3:00 3870 11610 EOL
16 C 28.11.03 40 Bccw 28.11.03 21 :30 14093 28.11.03 22:00 29.11.03 6:00 29.11.03 6:30 9:00 3:00 6:00 3870 23220 EOL
d
17 A 30.11.03 40 Bccw 30.11.03 0:00 14891 30.11.03 0:30 30.11.03 4:30 01.12.03 7:30 As per TN no 1/41
power 4891 Withdrawn
ite
cleared with 14588/888
Subtotal 15:00 6:00 9:00 3870 34830
im
rL
307
Fo
y
308
nl
December 2003
O
ntion Hour

n
Hrs Hrs Hrs Rs Rs
io
1 B 01.12.03 40 Bccw 01.12.03 8:00 17336 01.12.03 10:00 01.12.03 13:00 01.12.03 19:00 No E O L cleared with
12025
2 C 02.12.03 40 Bccw 02.12.03 16:00 14697 02.12.03 16:30 02.12.03 19:30 03.12.03 11:30 No E O L cleared with
at
12002
ul
13037
4 B 04.12.03 40 Bccw 05.12.03 23:30 14585 05.12.03 0:00 05.12.03 4:00 05.12.03 11:10 No E O L withdrawn for
fueling
irc
5 C 05.12.03 38 BCCW05.12.03 23:30 14660 06.12.03 0:00 06.12.03 3:00 08.11.03 3:30 4:00 3:00 1:00 3870 3870 E O L
6 A 07.12.03 40 Bccw 07.12.03 12:30 14999 07.12.03 13:00 07.12.03 16:00 07.12.03 16:30 4:00 3:00 1:00 3870 3870 E O L
C
14639
8 C 10.12.03 38 Bccw 10.12.03 2:30 13038 10.12.03 3:00 10.12.03 8:00 10.12.03 8:30 6:00 3:00 3:00 3870 11610 E O L
Subtotal
d 14:00 9:00 5:00 3870 19350
9 A 11.12.03 40 Bccw 11.12.03 10:30 14613 11.12.03 11:00 11.12.03 14:00 11.12.03 14:30 4:00 3:00 1:00 3870 3870 E O L
ite
10 B 12.12.03 40 Bccw 12.12.03 16:30 14045/35 12.12.03 17:00 12.12.03 20:00 12.12.03 20:30 4:00 3:00 1:00 3870 3870 E O L
11 C 14.12.03 40 Bccw 14.12.03 11:30 12004 14.12.03 12:00 14.12.03 16:00 14.12.03 16:30 5:00 3:00 2:00 3870 7740 E O L
12 A 15.12.03 40 Bccw 15.12.03 22:30 14613 15.12.03 23:00 16.12.03 4:00 16.12.03 4:30 6:00 3:00 3:00 3870 11610 E O L
im
13 B 17.12.03 40 Bccw 17.12.03 18:30 12010 17.12.03 19:00 17.12.03 23:00 17.12.03 23:30 5:00 3:00 2:00 3870 7740 E O L
14 C 18.12.03 40 Bccw 18.12.03 21:00 14613 18.12.03 21:30 19.12.03 0:30 19.12.03 1:00 4:00 3:00 1:00 3870 3870 E O L
15 A 19.12.03 40 Bccw 19.12.03 17:00 17744/ 19.12.03 17:30 19.12.03 20:30 19.12.03 21:00 4:00 3:00 1:00 3870 3870 E O L
rL
17462
Subtotal 32:00 21:00 11:00 3870 42570
(Continued)
Fo
y
nl
(Continued)
O
ntion Hour
n
Hrs Hrs Hrs Rs Rs
io
16 B 21.12.03 40 Bccw 21.12.03 10:30 12015 21.12.03 11:00 21.12.03 14:00 21.12.03 14:30 4:00 3:00 1:00 3870 3870 E O L
17 C 22.12.03 40 Bccw 22.12.03 8:30 14660 22.12.03 9:00 22.12.03 12:00 22.12.03 12:30 4:00 3:00 1:00 3870 3870 E O L
at
14779
19 B 25.12.03 40 Bccw 25.12.03 14:30 12045 25.12.03 15:00 25.12.03 19:00 25.12.03 19:30 5:00 3:00 2:00 3870 7740 E O L
ul
20 C 26.12.03 40 Bccw 26.12.03 23:00 14637 26.12.03 23:30 26.12.03 4:30 27.12.03 5:00 6:00 3:00 3:00 3870 11610 E O L
21 A 27.12.03 40 Bccw 27.12.03 20:00 12002 27.12.03 20:30 27.12.03 0:30 28.12.03 1:00 5:00 3:00 2:00 3870 7740 E O L
irc
13049
23 C 30.12.03 40 Bccw 30.12.03 19:30 14786/87 30.12.03 20:00 30.12.03 23:30 31.12.03 13:55 No E O L cleared with
13054

C
Subtotal d 24:00 15:00 9:00 3870 34830
ite
im
rL
309
Fo
Summary (as per L.NO.C/C/490/SHUNTING CHARGES/96/DT. 18.09.2003)
Period No of No of Total Free Time Rate per Amount

Rakes EOL Detention Time Charged Hour
Loaded Rakes Hrs Hrs Hrs Hrs Rs
y
th th
Ist Period (Sep 15 to 30 ) 14 2 12:30 6:00:00 6:30:00 3870 25155
nl
Total 14 2 12:30 6:00:00 6:30:00 3870 25155
O
Ist Period (Oct 1st to 10th ) 8 8 53:00 24:00 29:00 3870 112230
Iind Period (Oct 11th to 20th ) 8 5 29:00 15:00 14:00 3870 54180
IIIrd Period (Oct 21st to 31st) 8 5 32:00 15:00 17:00 3870 65790
n
Total 24 18 114:00 54:00 60:00 3870 232200
io
Ist Period (Nov 1st to 10th) 7 7 41:00 21:00 20:00 3870 77400
at
IInd Period (Nov 11th to 20th) 2 1 6:00 3:00 3:00 3870 11610
IIIrd Period (Nov 21st to 30th ) 8 2 15:00 6:00 9:00 3870 34830
Total 17 10
ul 62:00 30:00 32:00 3870 123840
irc
Ist Period (Dec 1st to 10th) 8 3 14:00 9:00 5:00 3870 19350
IInd Period (Dec 11th to 20th) 7 7 32:00 21:00 11:00 3870 42570
IIIrd Period (Dec 21st to 31st)
C
8 5 24:00 15:00 9:00 3870 34830

Total 23 15 70:00 45:00 25:00 3870 96750
d
ite
im
rL
Fo
Exhibit 14: Rakewise Detention Details at Dodballapur
y
September 2003
nl
Rake Arrival Unloading Rake Rake Removal Total Time
Releasing by Rlys. (Shunting
O
Time Rake Unloading
Time for (Shunting+ Waiting and
Star- Com- Unloa- unloading) Time for Rake
n
S ting pletion ding No. of at BSBT Engine Waiting)
No Rake Date Time Date Time Time (Hrs) Wagons (Hrs) Date Time (Hrs) (Hrs)
io
1 02.09.2003 3:25 02.09.2003 3:25 7:44 4:19 40 4:19 02.09.2003 8:40 0:56 5:15
2 03.09.2003 23:15 03.09.2003 23:33 3:40 4:07 40 4:25 04.09.2003 7:15 3:35 8:00
at
3 04.09.2003 17:45 04.09.2003 17:55 22:00 4:05 40 4:15 05.09.2003 7:40 9:40 13:55
4 05.09.2003 18:00 05.09.2003 18:30 22:37 4:07 40 4:37 05.09.2003 22:50 0:13 4:50
ul
5 07.09.2003 4:30 07.09.2003 4:45 8:49 4:04 40 4:19 07.09.2003 9:45 0:56 5:15
6 09.09.2003 0:05 09.09.2003 0:20 4:26 4:06 40 4:21 09.09.2003 10:15 5:49 10:10
irc
7 09.09.2003 18:05 09.09.2003 18:30 22:37 4:07 40 4:32 10.09.2003 1:05 2:28 7:00
8 11.09.2003 1:15 11.09.2003 1:30 5:45 4:15 40 4:30 11.09.2003 7:00 1:15 5:45
9 12.09.2003 23:00 12.09.2003 23:35 3:41 4:06 40 4:41 13.09.2003 8:20 4:39 9:20

C
10 14.09.2003 5:00 14.09.2003 5:25 9:38 4:13 40 4:38 14.09.2003 10:30 0:52 5:30
11 14.09.2003 14:30 14.09.2003 15:35 19:40 4:05 40 5:10 14.09.2003 20:30 0:50 6:00
12 A 16.09.2003 15:30 16.09.2003 16:05 20:15 4:10 40 4:45 16.09.2003 20:35 0:20 5:05
d
13 B 17.09.2003 3:00 17.09.2003 3:25 7:32 4:07 40 4:32 17.09.2003 9:45 2:13 6:45
14 C 18.09.2003 4:35 18.09.2003 5:00 9:25 4:25 40 4:50 18.09.2003 10:00 0:35 5:25
ite
15 A 19.09.2003 2:35 19.09.2003 2:55 7:12 4:17 40 4:37 19.09.2003 8:20 1:08 5:45
16 B 20.09.2003 23:00 20.09.2003 23:25 3:32 4:07 40 4:32 21.09.2003 11:45 8:13 12:45
17 C 21.09.2003 19:00 21.09.2003 19:18 23:24 4:06 40 4:24 22.09.2003 2:15 2:51 7:15
im
18 A 22.09.2003 16:15 22.09.2003 16:29 20:59 4:30 40 4:44 22.09.2003 21:15 0:16 5:00
19 B 25.09.2003 17:30 25.09.2003 17:50 22:05 4:15 40 4:35 25.09.2003 23:10 1:05 5:40
20 C 26.09.2003 7:40 26.09.2003 8:08 12:14 4:06 40 4:34 26.09.2003 14:15 2:01 6:35
rL
21 A 27.09.2003 9:55 27.09.2003 10:08 14:15 4:07 40 4:20 27.09.2003 15:45 1:30 5:50
22 B 28.09.2003 15:55 28.09.2003 16:10 20:26 4:16 40 4:31 29.09.2003 2:25 5:59 10:30
23 C 29.09.2003 16:20 29.09.2003 16:30 20:35 4:05 40 4:15 29.09.2003 21:05 0:30 4:45
311
Fo
24 A 30.09.2003 19:50 30.09.2003 20:02 0:10 4:08 40 4:20 01.10.2003 2:50 2:40 7:00
Average 4:10 4:31 2:31 7:00
y
312
nl
October 2003
O
Time Rake Unloading

n
Star- Com- Unloa- No. unloading) Time for Rake
S ting pletion ding of at BSBT Engine Waiting)
io
No Rake Date Time Date Time Time (Hrs) Wagons (Hrs) Date Time (Hrs) (Hrs) Remarks
1 B 01.10.2003 15:30 01.10.2003 16:04 20:07 4:03 40 4:37 02.10.2003 16:45 20:38 25:15
at
2 C 02.10.2003 22:50 02.10.2003 23:07 3:12 4:05 40 4:22 03.10.2003 10:20 7:08 11:03
3 A 04.10.2003 16:30 04.10.2003 16:50 20:55 4:05 40 4:25 04.10.2003 21:00 0:05 4:30
4 B 05.10.2003 4:45 05.10.2003 5:13 9:26 4:13 40 4:41 05.10.2003 10:55 1:29 6:10
ul
5 C 06.10.2003 4:20 06.10.2003 15:35 19:40 4:05 40 15:20 08.10.2003 1:00 29:20 44:40 37.15 Hrs-Insufficient silo space
6 A 08.10.2003 2:15 08.10.2003 21:30 1:35 4:05 40 23:20 09.10.2003 2:20 0:45 24:05 19.15 Hrs-Insufficient silo space
irc
7 B 09.10.2003 3:25 09.10.2003 23:40 4:06 4:26 40 24:41 10.10.2003 7:45 3:39 28:20 20.15 Hrs-Insufficient silo space
C
11 C 14.10.2003 23:35 15.10.2003 0:10 4:16 4:06 40 4:41 15.10.2003 10:15 5:59 10:40
12 A 17.10.2003 1:00 17.10.2003 1:27 5:39 4:12 40 4:39 17.10.2003 11:35 5:56 10:35
13 B
d
17.10.2003 22:35 17.10.2003 23:08 3:22 4:14 40 4:47 18.10.2003 6:40 3:18 8:05
14 C 18.10.2003 10:20 18.10.2003 11:15 16:26 5:11 40 6:06 18.10.2003 19:00 2:34 8:40
ite
15 A 20.10.2003 3:15 20.10.2003 3:33 7:42 4:09 40 4:27 20.10.2003 10:10 2:28 6:55
16 B 21.10.2003 4:20 21.10.2003 4:55 9:01 4:06 40 4:41 21.10.2003 11:45 2:44 7:25
im
rL
22 B 30.10.2003 1:20 30.10.2003 1:55 6:05 4:10 40 4:45 30.10.2003 9:55 3:50 8:35
23 C 31.10.2003 6:35 31.10.2003 7:00 11:20 4:20 40 4:45 31.10.2003 14:15 2:55 7:40
Average 4:15 14:29 5:34 20:03
Fo
y
nl
November 2003
O
Time Rake Unloading
n
io
1 A 01.11.2003 1:55 01.11.2003 2:21 6:25 4:04 40 4:30 01.11.2003 7:20 0:55 5:25
at
2 B 02.11.2003 4:15 03.11.2003 15:43 19:57 4:14 40 39:42 03.11.2003 20:48 0:51 40:33 35:28 Hrs Insufficient silo space
3 C 03.11.2003 22:20 04.11.2003 20:45 0:55 4:10 40 26:35 05.11.2003 5:45 4:50 31:25 22:25 Hrs Insufficient silo space
4 A 05.11.2003 6:35 06.11.2003 2:00 6:07 4:07 40 23:32 06.11.2003 10:00 3:53 27:25 19:25 Hrs Insufficient silo space
ul
5 B 07.11.2003 9:50 07.11.2003 10:10 14:40 4:30 40 4:50 07.11.2003 18:50 4:10 9:00
irc

C
11 B 23.11.2003 1:30 23.11.2003 1:55 6:35 4:40 40 5:05 23.11.2003 11:40 5:05 10:10
12 C 24.11.2003 0:10 24.11.2003 0:30 4:35 4:05 40 4:25 24.11.2003 5:40 1:05 5:30
13 A
d
24.11.2003 10:35 24.11.2003 10:55 15:00 4:05 40 4:25 24.11.2003 19:05 4:05 8:30
14 B 26.11.2003 5:10 26.11.2003 5:30 10:26 4:56 40 5:16 26.11.2003 10:50 0:24 5:40 52 min PLC control voltage prob.
ite
15 C 27.11.2003 5:35 27.11.2003 6:05 10:10 4:05 40 4:35 27.11.2003 13:00 2:50 7:25
16 A 28.11.2003 4:30 28.11.2003 5:05 9:10 4:05 40 4:40 28.11.2003 11:35 2:25 7:05
17 B 29.11.2003 15:30 29.11.2003 15:35 19:38 4:03 40 4:08 29.11.2003 20:10 0:32 4:40
im
18 C 30.11.2003 13:30 30.11.2003 13:40 17:45 4:05 40 4:15 30.11.2003 19:05 1:20 5:35
Average 4:12 20:18 2:51 23:10
rL
313
Fo
y
314
nl
December 2003
O
Time Rake Unloading

n
io
1 A 02.12.2003 18:35 02.12.2003 19:08 23:14 4:06 40 4:39 02.12.2003 23:40 0:26 5:05
at
2 B 03.12.2003 10:50 03.12.2003 11:04 15:08 4:04 40 4:18 03.12.2003 17:20 2:12 6:30
3 C 04.12.2003 20:15 04.12.2003 20:25 0:25 4:00 38 4:10 05.12.2003 4:15 3:50 8:00
4 A 06.12.2003 4:35 06.12.2003 4:51 8:45 3:54 40 4:10 06.12.2003 11:35 2:50 7:00
ul
5 B 06.12.2003 17:05 06.12.2003 17:25 21:19 3:54 40 4:14 06.12.2003 23:59 2:40 6:54
irc
C
13 A 18.12.2003 1:55 18.12.2003 2:08 5:55 3:47 40 4:00 18.12.2003 10:20 4:25 8:25
d
14 B 19.12.2003 20:05 19.12.2003 20:15 23:59 3:44 40 3:54 20.12.2003 3:00 3:01 6:55
ite
15 C 20.12.2003 17:40 20.12.2003 17:58 23:35 5:37 40 5:55 21.12.2003 3:45 4:10 10:05 01:15 Hrs Boot positioner problem
17 B 23.12.2003 15:20 23.12.2003 15:58 19:55 3:57 40 4:35 24.12.2003 7:20 11:25 16:00
18 C 25.12.2003 5:05 25.12.2003 5:38 9:32 3:54 40 4:27 25.12.2003 12:05 2:33 7:00
im
21 C 29.12.2003 12:50 29.12.2003 13:12 17:12 4:00 40 4:22 29.12.2003 18:00 0:48 5:10
rL
Average 4:19 11:08 4:30 15:38
Fo
Exhibit 15: Statement of Turn Around Time
Time Taken Transit Time

for Engine Time
Idle Time Arrival Malkhaid Taken for Total
y
Turn No of Time taken After to Dodba- Unloading Turn
nl
around Rakes Prior to For Loading Dodba- llapur to at Dodba- Around
status Loaded Loading Loading completed llapur Malkhaid llapur Time
Hrs Hrs Hrs Hrs Hrs Hrs Hrs
O
Turn around status 109 5:14 4:37 12:52 34:23 29:48 10:13 97:10
prior to 15.09.03
(01.04.03 to
n
12.09.03)
io
Turn around status
from 15.09.03 to
31.12.03
at
Of rakes on which 45
EOL availed
Of rakes on which
EOL not availed
Weighted average
33
21

5:09

4:34
ul
6:18

34:05 29:40

8:13

88:02
irc
for the year 03-04 (monthly
(01.04.03 to average)
31.12.03)
C
Weighted average 22 7:36 3:35 13:13 34:42 30:09 9:46 99:03

for the year 02-03 (monthly
average)
d
Idle time at Dodballapur/Malkhaid in lieu of any extreme disruption is discounted

ite
REFERENCES
im
1. CMA, 2003. Cement Manufacturers’ Association, New Delhi (Downloadable from

http://www.cmaindia.org).
rL
2. Indian Railways, 2003a, Year Book 2001-02, Ministry of Railways, New Delhi.
3. Indian Railways, 2003b, Trains at a Glance, Ministry of Railways, New Delhi.
4. RC, 2003, Various Internal Documents, Rajashree Cement, Malkhaid.
Fo
5. South Central Railway, 2004. Internal Document, Secunderabad.
1. Should RC and SCR have the option of declaring a trip as ‘not-for-EOL’? Can this
be unilaterally decided or should it be a joint decision? What should be the required
advance notice?
2. From the time the rake reaches the terminal, what should be the guaranteed time
for engine availability? What should the penalty be on RC if load is not available?
What should the penalty be on SCR if engine is not available? (Currently there is no
penalty on SCR).
3. What steps should RC take to ensure timely loading at the terminal?
y
4. What steps should SCR take to ensure EOL, and by implication, powering the
nl
incoming train with an appropriate engine?
5. Is the issue EOL or timely availability of engine? Incoming and outgoing engines
O
need not be the same, should SCR want the flexibility.
6. What performance measures should be used for IR to improve customer service and
n
position itself as a third party logistics solution provider?
io
at
ul
Possible performance measures of interest to RC and SCR are-(a) the turnaround time at the
terminals (b) the delivery time between Malkhaid and Dodballapur, and (c) the total
turnaround time for the whole circuit. These should be analysed from the point of view of
irc
RC and SCR.
The questions are largely related to establishing a sustainable contract between RC and
C
IR. While RC’s objective is to increase the throughput of bulk cement, using the same asset
base, IR’s objective is to ensure that the EOL concept takes off. The EOL concept would
essentially provide turnaround time improvements at the terminal, while for RC, the
d
turnaround of the entire fleet between the origin and the destination would be of
ite
significance.
Based on the three-month data, an analysis can be attempted to see how the EOL and the
non-EOL movements have performed. It would be important to break the data set from
im
Exhibits 13 and 14, into three separable sets, since there are three rakes doing the round
trip movement between the origin and the destination. Summary statistics can then be
generated to fill up the format in Exhibit 15.
rL
This analysis would bring out an approach to answering the questions, if not providing
specific recommendations. The terminal turnaround versus total turnaround time would
also get highlighted, since, while satisfying the EOL requirements, IR often passed the
Fo
problems on to the transit segment. The consequence was that while the terminal
performance looked good, the transit and total turnaround time performance really did not
improve.
Simulation of the performance of this supply chain is also possible with the data that is
provided. This may be useful for looking at the impact of policy variables such as EOL at the
destination end, type of engine being assigned and different rake maintenance strategies of
the railways, and impact of randomness in transit times and occurences of accidents.
CASE 3
y
nl
An important decision in petroleum supply chains that operate through port-based
O
refineries is to determine optimal jetty capacity for handling crude and refined products.
Western Oil Limited, a refinery in Western India was set for an expansion that would more
n
than quadruple its capacity over the coming three years.
This case examines how to expand the captive jetty capacity to match the projected
io
refinery output in terms of volume and variety. An appropriate trade off between the cost of
jetty expansion and demurrage due to ship delays had to be considered.
at
ul
irc
WESTERN OIL LIMITED (A)
Western Oil Limited (A)The following letter was received by Professor Prashanth in October
C
2007. Please help him in the proposed study.

4th October 2007
d
Mr. K Sitaram
Head – Shipping Operations, and Member, Investment Committee
ite
Western Oil Limited

Makers house,
im
Mumbai - 400020
Prof. R. Prashanth
rL
Well-known Institute of Management in Western India (WIMWI)

Karnavati
Dear Prof. Prashanth,
Fo
I hope you would recall me as a participant in the Strategic Port Management Programme
held at WIMWI in May 2006.
Prepared by Chetan Soman and G. Raghuram. We acknowledge the research support by Shivani Shukla.
Teaching material of the Indian Institute of Management, Ahemdabad, is prepared as a basis for class
discussion. Cases are not designed to present illustrations of either correct or incorrect handling of
administrative problems. Copyright © 2008 by the Indian Institute of Management, Ahmedabad.
As you are probably aware, our company has sailed through troubled waters right from its
inception, but now that should be a thing of the past. When speaking of the Western Oil
Limited (WOL) in its ‘new-avatar’, we are thinking of a truly world class refinery and equally
good infrastructure facilities to support it in its current and expanded capacity. But we have
to take into account the diverse and changing product-process-market characteristics.
y
In particular, WOL’s management is reviewing the jetty capacity requirements, and
nl
alternate methods of providing the additional capacity to match the three-fold expansion
(due in the next couple of years) in its Sikkanar refinery output.
O
Knowing your interest and expertise in this area, we would like you to develop a model
based on operations management/research tools or such other tools as you may deem fit to
use and advise us on jetty capacity requirements at various refinery throughput levels. You
n
can assume that there is sufficient number of product tanks available. You may also want to
suggest some suitable software to enable us to examine such decisions in the future.
io
I am enclosing a few pages introducing our company, projected marine movements, the
at
jetty operations and the relevant cost estimates.
Thanking you
(sd) Sitaram
ul
irc
INTRODUCTION
The Western Oil Limited (WOL) is a part of the large corporate house - Western - which has
C
presence in various sectors: Telecommunication, Textile, Steel, BPO, Oil and Gas etc. The
Oil and Gas Group is further organized into three separate divisions, each focusing on one
d
link in the petroleum value chain:

ite
1. Exploration and Production

Western Oil and Gas was one of the first private companies to bid for exploration blocks in
im
early 1990s, under the New Exploration Licensing Policy of Government of India, along
with the other major private players (Reliance and Essar). It has acquired rights for oil and
gas explorations blocks in Assam and Gujarat in India.
rL
2. Refinery (WOL)
Fo
WOL’s oil refinery is located on India’s west coast in close proximity to the crude rich Gulf
States. Sikkanar is an all weather deep draft natural port. More than 60% of India’s crude
imports land in and around this region. Besides, the refinery’s location enables access to the
fast growing markets in the north and western region of India through product pipelines.
The eastern and southern parts of India are generally serviced through the coastal route
circling the country.
Case 3: Western Oil Limited (A) 319
WOL’s refinery has a capacity of 7.5 million tons per annum (mtpa) with an investment of
close to Rs 10 billion. The refinery is built with the state of the art technology with technical
and project assistance from the world’s leading consultants and equipment suppliers in the
field. It is designed to handle a diverse range of crude mixes. The refinery is configured to
produce Euro II and Euro III grades of petrol (MS - motor spirit) and diesel (HSD—high
y
speed diesel). With mid-stream upgradation of processes and technologies, the refinery has
the capability to process the most sour, acidic and heavy crude.
nl
The refinery is fully integrated with its own dedicated 120 mega watt co-generation power
plant, port and terminal facilities. It includes a single point mooring (SPM) capable of
O
handling vessels up to 350,000 deadweight tons (DWT) capacity, marine product dispatch
capacity of 14 mtpa, and railcar and truck loading facilities.
n
The refinery has built-in environment-friendly technologies for pollution management
and has also planted over one million trees to ensure a green corridor around the entire
io
refinery complex.
at
The refinery is set for an expansion in the next couple of years. Exhibit 1 shows the
planned refinery throughput milestones. The key idea behind the expansion is to produce
higher value-added products and to have economies of scale. Exhibit 2 gives the national
ul
production, import and refining capacity of crude, and production, import and export of
products.
irc
3. Marketing
C
WOL took rapid marketing strides by setting up its retail network all across India. The
company has set up more than 800 retail outlets. (Effective 1st April 2002, marketing of
transportation fuel, including MS and HSD was de-regulated and all companies meeting the
d
eligibility criteria laid down by the Ministry of Petroleum & Natural Gas, Government of
ite
India, were permitted to market these products, and also to set up a retail network after
receiving formal authorization from the Government). In addition to this, there are another
900 retail outlets under various stages of construction, and 1100 franchisees and sites
im
identified. Currently, the company has 17 depots to ensure timely supplies to all its retail
outlets. In addition to the above, more such depots are identified.
rL
Following the industry trend, a host of other products and services are planned to
compliment MS and HSD at the retail outlets. Some of these are restaurants, dhabas,
truckers stop, car wash, service stations, ATMs, convenience stores, cafés, etc. Some of the
companies with which WOL has an association for supplies of other products and services
Fo
are Castrol (for lubricants), Pepsi, Coca Cola, Frito-Lay, and State Bank of India.
The company had planned to set up over 5000 retail outlets across the country by 2008.
However, given the higher crude prices in the recent years and the Government not hiking
the fuel prices fearing a political backlash, it doesn’t make economic sense for private players
to sell the products in India. Hence, WOL has decided to focus more on the exports.
Projected Marine Movements

Given the new focus on the export markets, it is but natural to have more marine
movements. Exhibit 3 shows the projected marine movements and various expected parcel
sizes to be sold for the various products. This is provided for various throughput levels. It is
evident that WOL expects larger parcel sizes for larger marine movements. This brings in
y
economies of scale in jetty operations and transportation.
nl
Jetty Operations
O
The jetty located at Sikkanar is owned and operated by WOL itself. Currently, WOL has to
get pilots on board from the Kandla port, but soon this dependence is likely to vanish. There
n
is only one berth at the moment. The depth of water alongside the berth is about 20 meters.
io
The length of the berth is 310 meters. It can accommodate vessels ranging from 10,000 to
100,000 DWT.
at
The petroleum product tankers arrive and anchor at the anchorage and await their turn
for berthing alongside the jetty for high waters. Although the WOL jetty is located in a non-
tidal port, for ease of navigation, the current practice is to berth and de-berth activities only
ul
when tide is moving from low to high. Exhibit 4 gives the tide calendar as a sample for the
month of September 2007. Exhibit 5 brings out the implication of the current practice due
irc
to tide variations on the waiting time of the ships.
At the jetty, WOL has two dedicated pipelines of 16” each for loading white oil and one
C
dedicated pipeline of 12” for loading black oil. Typically, these ships are designed to load the
petroleum products as per flow rate depending on their DWT and certain ship board
conditions like ballast discharge rate. From the shore side, the company can pump the
d
petroleum products into the ships at the maximum rate of 2000 m3/hr per pipeline for white
oil. The pumping rate for black oil is a maximum 1400 m3/hr. However, historical data
ite
suggests that, generally, only 85% of the above flow rates are achievable. Exhibit 6 shows the
schematic representation of the pipelines present.
im
Currently, only one of the pipelines is used for pipeline loading. This is because, in the
case of multiple products loading, waste called ‘slop’ is generated as we transition from one
product to another. This slop is pumped back to the refinery using the second pipeline. Very
rL
soon, it is planned to set up a slop tank near the jetty, from where the slop would be
transported by road. This would allow double line loading of white oil with a rate of upto
4000 m3/hr. The slop is generated even though a required operation called ‘pigging’ is done
Fo
between the two different products. This takes upto three hrs. In future, WOL plans to have
dedicated product pipelines. With dedicated pipelines, the pigging time loss won’t happen.
This will also reduce the need for the slop tanks. Instead, warping (alignment of the
pipelines’ loading arm with the ship) will have to be done which takes upto two hrs. Around
10% of the ships will have loading of multiple products.
In addition, there could be delays because of pump and/or tankage problems at the jetty.
Last year, 2% of the vessels were delayed by two to three hours on account of this. Upon
completion of loading operations, the product tankers are deberthed during high waters after
inspection, documentation, etc.
Presently, there are restrictions in terms of non-availability of night navigation. It means
that the ships can neither be berthed nor can they be deberthed after daylight hours.
However, with work underway, it is expected that this restriction of non-availability of night
y
navigation will be overcome latest by the end of 2007.
nl
Exhibit 7 shows the ship arrival times and the quantity loaded during the period
December 2006 to July 2007. It is suspected that although most arrivals were scheduled or
O
based on the product stock levels, the actual arrival times were random. (As is usual in such
arrival patterns, the inter arrival times are expected to follow a negative exponential
distribution).
n
Relevant Costs
io
Due to the various operating conditions governing the berthing and loading, delays could
at
occur to the ships. A demurrage is payable if the actual operation time exceeds agreed laytime
which is typically 36 hours for loading and documentation. The clock for the actual
ul
operation time starts six hours after the first available favourable daylight tide or actual
berthing time, whichever is earlier. The clock stops when all documentation is on board.
irc
There are agreed deductions available on the actual operation time for documentation and
shifting of the vessel. Delays on account of vessel and/or weather would not be part of actual
operation time.
C
Once night navigation is in place, the first available favourable tide will no longer be
restricted to the daylight hours for short vessels i.e. upto 40,000 DWT. For very large vessels,
the 36 hours are typically extended to 48 or 60 hours. If the ship does not arrive during the
d
agreed laycan (arrival period), then the actual operation time clock starts only after berthing
ite
which can now happen at the jetty’s convenience.

The demurrage rates are typically in the range of USD 20,000 to 30,000 per day. One
would expect larger rates for the larger vessels but this need not be the case. The rates are
im
often supply and demand driven. Exhibits 8a, 8b and 8c show typical ‘Statement of Facts’ for
the vessels and the demurrage calculations. These statements capture, in detail, the vessel
rL
movement, the related activities and their duration. Exhibit 8a relates to a vessel which
arrived in the night with the first available favourable tide being at 6.00 am. Exhibit 8b
relates to a vessel whose first available daylight tide was available on arrival, with berthing
done rightaway. Exhibit 8c relates to a vessel which just missed the tide. This vessel also had
Fo
two product loading involving pigging operation.

Acknowledging the high demurrage rates, the general industry philosophy is that “a berth
should wait for the vessels and not the other way around”. The WOL management feels that
they should not have more than 65% berth occupancy. The cost of constructing a new jetty,
that can accommodate practically all the vessel sizes, is about Rs 100 crores (approximately
USD 25 million). The jetty operating expenses are about USD 0.25 million per year.
Exhibit 1: Refinery Throughput Milestones
Time Capacity Remarks
As of now 7.5 mtpa
y
December, 2007 10.5 mtpa Integration of secondary process including cracker, hydrotreatment,
sulphur production and value added products.
nl
September, 2009 16 mtpa Addition of coker, fuel oil handling and heavy crude oil diesel
production.
O
December, 2009 32 mtpa Duplication of 16 mtpa facilities.
n
io
Exhibit 2: National Petroleum Imports, Production and Exports
at
(000' t)
Crude Products
Production Import Refining
Capacity
ul
Production Import Export
irc
*
2002-03 33044 81989 116968 104140 7228 10289
2003-04 33373 90434 127368 113463 8001 14620
C
2004-05 33981 95861 127368 118579 8828 18211

2005-06 32190 99409 132468 119750 11677 21507
2006-07 33988 110858 148968 135260 16966 32394
d
** Base Upper Base Upper

ite
case case case case
2007-08 41230 107270 158700 4070 4110 30950 29870

2008-09 42330 120370 194700 3320 3400 40640 37860
im
2009-10 42490 150510 210210 4280 4400 67880 62200

2010-11 41190 170410 225880 5490 6600 79560 71060
2011-12 39510 195490 240960 6130 7460 92680 83980
rL
Base case: Low Demand projections

Upper case: High Demand projections
Fo
Source:* http://petroleum.nic.in/petstat.pdf
* * http://planningcommission.nic.in/aboutus/committee/wrkgrp11/wg11_petro.pdf
Exhibit 3: Projected Marine Movements and Parcel Sizes
Products Density Grade 10.5 mtpa 16 mtpa 32 mtpa
Total Parcel Total Parcel Total Parcel
y
marine size marine size marine size
movement movement movement
nl
ton/m 3 (kt/year) (kt) (kt/year) (kt) (kt/year) (kt)
LPG 0.53 Domestic 858 13 1661 13
O
Gasoline1 0.74 Grade 1 2386 60 990 30
0.74 Grade 2 2336 40 627 40 4924 60
0.74 Grade 3 4115 90
n
ATF 1 0.79 Jet export 950 29 2419 60 3441 60
io
HSD1 0.84 Grade 1 3600 90 1485 40
0.84 Grade 2 3689 60 1485 60 2696 80
0.84 Grade 3 5629 100
at
PCN 1 0.66 Domestic 122 20 1198 50 713 80
SKO1 0.79 Domestic 396 30
VGO2
FO 2
0.86
0.87
Domestic/ Export
Domestic

1914
ul
40
584

60

irc
Total Marine Movement (kt/year) 9011 13157 26050
1
white oil
2
black oil.
C
For LPG, the maximum loading rate is 800 m3/hr.

kt = 000 tons
d
ite
Exhibit 4: Tide Calendar for Sikkanar for September 2007

im
Units are meters, timezine is IST

September 2007 low is 0.4 m, high is 7.0 m, range is 6 6 m
rL
Sunday Monday Tuesday Wednesday Thursday Friday Saturday
0826 0827 Full 0828 0829 0830 0831 0901

H0012 5.4 H0105 5.8 H0150 6.1 H0233 6.4 H0316 6.6 H0358 6.7 H0443 6.7
Fo
L0657 1.0 L0743 0.8 L0824 0.6 L0905 0.5 L0944 0.6 L1025 0.8 L1106 1.0
H1321 6.3 H1354 6.6 H1427 6.8 H1500 6.9 H1533 7.0 H1608 6.9 H1644 6.7
L1959 1.7 L2032 1.4 L2105 1.0 L2139 0.7 L2214 0.5 L2251 0.4 L2331 0.4
0902 0903 LQtr 0904 0905 0906 0907 0908
H0529 6.6 L0016 0.5 L0108 0.8 L0216 1.0 L0345 1.2 L0518 1.3 L0631 1.2
L1151 1.4 H0621 6.3 H0723 6.0 H0847 5.8 H1035 6.8 H1157 6.0 H1249 6.2
H1722 6.4 L1241 1.8 L1348 2.1 L1533 2.4 L1730 2.3 L1850 2.0 11940 1.7
H1804 6.1 H1853 5.7 H2004 5.4 H2157 5.2 112339 5.4
(Continued)
(Continued)
0909 0910 New 0911 0912 0913 0914 0915
H0043 5.6 H0129 5.9 H0208 6.1 H0242 6.3 H0314 6.4 110345 6.4 H0416 6.4
L0725 1.0 L0807 1.0 L0843 0.9 L0913 1.0 L0941 1.0 L1009 1.2 L1035 1.3
H1327 6.4 H1400 6.5 H1428 6.6 H1455 6.6 H1521 6.5 H1546 6.4 H1611 6.2
L2018 1.4 L2048 1.2 L2114 1.0 12138 0.9 12201 0.8 12224 0.8 L2247 0.8
y
0916 0917 0918 FQtr 0919 0920 0921 0922
H0447 6.3 H0520 6.1 H0557 5.9 L0015 1.2 L0102 1.4 L0223 1.6 L0412 1.6
nl
L1102 1.5 L1130 1.8 L1202 2.0 H0643 5.7 H0749 5.5 H0932 5.5 H1107 5.7
H1635 6.1 H1659 5.9 H1725 5.6 L1245 2.3 L1413 2.5 L1644 2.5 L1802 2.2
O
L2312 0.9 L2341 1.0 H1757 5.4 H1846 5.2 H2032 5.0 H2244 5.1
0923 0924 0925 0926 Full 0927 0928 0929
L0534 1.4 L0633 1.2 H0050 6.0 H0135 6.4 H0217 6.8 H0259 7.0 H0342 7.1
n
H1200 6.0 H1239 6.3 L0721 0.9 L0805 0.8 L0846 0.7 L0928 0.8 L1009 0.9
L1846 1.8 L1922 1.3 H1315 6.5 H1350 6.7 H1426 6.8 H1502 6.8 H1539 6.7
io
H2358 6.5 H965 0.9 L2030 0.5 L2106 0.2 L2142 0.0 L2220 0.0
0930 1001 1002 LQtr 1003 1004 1005 1006
at
H0426 7.0 H0513 6.8 H0604 6.4 L0040 0.8 L0152 1.2 L0333 1.5 L0509 1.5
L1052 1.2 L1138 1.5 L1233 1.9 H0705 6.1 H0827 5.8 H1011 5.8 H1129 5.9
H1617 6.5 H1657 6.2 H1743 5.9 L1352 2.1
ul L1547 2.2 L1728 2.0 L1831 1.7
L2302 0.2 L2346 0.4 H1839 5.5 H2006 5.2 H2216 5.1 H2346 54
irc
Exhibit 5: Tides and Implications for Ship Movements
C
d
ite
im
rL
Fo
At Sikkanar, currently all the ship movements are only during the low-to-high tide periods.
In the above figure, the ship has arrived at 11:39 PM marked by + (at the center of the
figure). It cannot berth straight away, since the tide is not favorable. Further ship movement
is possible at around 4 AM if night navigation facilities are present and only after 6 AM in
their absence. Before the movement, the ship also has to get customs clearance, medical
y
clearance, shore readiness signal, and the pilot on board. Also, a maximum of three shipping
movements are possible through the channel during any favourable tide.
nl
O
Exhibit 6: Pipeline Layout
n
io
at
ul
irc
C
d
ite
im
rL
Refinery to Product Tank Farm: 2 kms

Product Tank Farm to Jetty: 22 kms
Fo
Exhibit 7: Ship Arrival Times and Quantity Loaded for Different Products
y
nl
O
n
io
at
ul
irc
C
d
ite
im
rL
Fo
Exhibit 8a
I. Statement of Facts
Vessel: Ship 20 Cargo: HSD
Laytime allowed:
y
Minimum of (First available daylight tide + 6 hrs, Berthed and all-fast time) + 36 Hrs. Delays because of weather,
nl
problems on vessel side not to be counted. Shifting time from Anchorage to berth should not be counted.
Demurrage rate: $25000/day
Last port: Fujairah Next Port: West Africa
O
Activity From To
Date Time Date Time
n
End of sea passage 02.03.07 2130
io
NOR Tendered 02.03.07 2200
Anchored 02.03.07 2225
at
Custom on Board 02.03.07 2300
Free Pratique 02.03.07 2325
Custom clearance granted 03.03.07
ul 0125
Anchor aweigh 06.03.07 1054
Pilot on board 06.03.07 1127 06.03.07 1350
irc
Berthing tug used 06.03.07 1130 06.03.07
Mooring 06.03.07 1236 06.03.07 1335
NOR Received 06.03.07 1350
C
Gangway down 06.03.07 1350

Pre Operation meeting 06.03.07 1350 06.03.07 1410
Tank inspection 06.03.07 1410 06.03.07 1510
d
L/A connection 06.03.07 1445 06.03.07 1500

ite
Safety check list 06.03.07 1440 06.03.07 1515

Ship Line up 06.03.07 1515 06.03.07 1525
Shore Line up 06.03.07 1525 06.03.07 1954
im
Loading HSD 06.03.07 1954 08.03.07 0900

Stoppage by shore 08.03.07 0400 08.03.07 0530
Re-ullaging/Cargo calculation 08.03.07 1100 08.03.07 1200
rL
Anchored awaiting cargo documents 08.03.07 1348

Tank inspection/Ullaging 08.03.07 0900 08.03.07 1000
Cargo Calculations 08.03.07 1000 08.03.07 1030
Draining/Disconnection of L/A 08.03.07 0930 08.03.07 1048
Fo
Documents on board 08.03.07 1630

Documentation completed 08.03.07 1645
POB for unmooring 08.03.07 1120 08.03.07 1310
Unberthing tug used 08.03.07 1230 08.03.07 1300
Unmooring 08.03.07 1230 08.03.07 1248
Vessel sailed 08.03.07 1730
Quantity loaded (tons) 59,560

II. Demurrage Calculation for Ship 20
Activities Date Time
Vessel anchored 02.03.07 2225

NOR tendered 02.03.07 2200
y
Time to count as of 03.03.07 1200
nl
Berthed 06.03.07 1335
Hose connected 06.03.07 1500
O
Commenced loading 06.03.07 1954
Completed loading 08.03.07 0900
Hoses off 08.03.07 1048
n
io
Time to count till 08.03.07 1630
Gross time taken 124:30:00
at
Time allowed for documentation 4:00:00
Shifting ul 2:30:00
Net time at loadport 118:00:00

irc
Time allowed as per charterparty 36:00:00
Time on demurrage 82:00:00
Demurrage rate (per day) $25,000
C
Amount payable $85,417

Amount paid $85,347
d
ite
im
rL
Fo
Exhibit 8b
Vessel: Ship 20 Cargo: HSD
y
Vessel: Ship 42 Cargo: Naphtha
nl
Laytime allowed:
O
Demurrage rate: $27000/day Last port: Fujairah Next Port: Fujairah
Activity From To
n
Date Time Date Time
io
Anchored 12.05.07 1605
at
Custom clearance granted 12.05.07
ul 1730
Pilot on board 12.05.07 1605 12.05.07 1800
irc
Berthing tug used 12.05.07 1620 12.05.07 1810
Mooring 12.05.07 1730 12.05.07 1815
NOR Received 12.05.07 1845
C

Tank inspection 12.05.07 1900 12.05.07 2015
d
L/A connection 12.05.07 1900 12.05.07 1915

ite

Ship/Shore Line up 12.05.07 2015 12.05.07 2036
Loading Naphtha 12.05.07 2036 14.05.07 0135
im

Stoppage by ship 13.05.07 1435 13.05.07 1615
rL
Tank inspection/Ullaging 14.05.07 0135 14.05.07 0235

Cargo Calculations 14.05.07 0235 14.05.07 0305
Draining/Disconnection of L/A 14.05.07 0215 14.05.07 0236
Fo

POB for unmooring 14.05.07 0538
Unberthing tug used 14.05.07 0540
Unmooring 14.05.07 0600
Quantity loaded (tons) 29,853

y
Berthed 12.05.07 1830
nl
O
Hoses off 14.05.07 0200
n
io
at
Shifting 1:00:00
Net time at loadport

Time allowed as per charterparty
ul 30:30:00
36:00:00
irc
Amount payable $0
C
Amount paid $0
d
ite
im
rL
Fo
Exhibit 8c
Vessel: Ship 58 Cargo: HSD and ATF
Laytime allowed:
y
nl
Demurrage rate: $26000/day. Last port: Singapore Next Port: Djibouti
O
Activity From To
Date Time Date Time
n
io
Anchored 18.06.07 1415
Custom clearance granted 18.06.07 1630
at
Pilot on board 20.06.07 1130
Berthing tug used
Mooring
ul
20.06.07

1255

20.06.07

1400
irc
NOR received 20.06.07 1442
L/A connection for ATF loading 20.06.07 1442 20.06.07 1450
C
Tank inspection 20.06.07 1500 20.06.07 1536

Ship/Shore line up 20.06.07 1600 20.06.07 1612
d
Loading ATF 20.06.07 1612 21.06.07 1154

ite
Rate reached to max 2100 cu.m 20.06.07 1625

Stoppage by shore (for pig launching) 21.06.07 0730 21.06.07 0842
Resume Loading ATF 21.06.07 0842
im
Ref. ullaging for ATF 21.06.07 1154 21.06.07 1254

L/A disconnection ATF 21.06.07 1214 21.06.07 1218
L/A connection for HSD loading 21.06.07 1218 21.06.07 1224
rL
Ship/Shore line up 21.06.07 1254 21.06.07 1754

Loading HSD 21.06.07 1754 21.06.07 2218
Rate reached to max 1600 cu.m 21.06.07 1810
L/A disconnection HSD 21.06.07 2324 21.06.07 2330
Fo
Ullaging & calculation for ATF & HSD 21.06.07 2218 21.06.07 2324
POB for unmooring 22.06.07 0130 22.06.07
Unberthing 22.06.07 0130 22.06.07 0205
ATF: Quantity loaded (tons) 28,496
HSD: Quantity loaded (tons) 5048

y
nl
Berthed 20.06.07 1400
O
Hoses off 21.06.07 2330
n
io
at
Shifting 2:30:00
Net time at loadport
Time allowed as per charterparty
ul 50:00:00
36:00:00
irc
Amount payable $15,167
C
Amount paid $14,722

d
ite
GLOSSARY OF TERMS
(Compiled from various sources)
im
Term Definition
Anchor A device which is attached to anchor chain at one end and lowered
rL
into the sea bed to hold a ship in position; it is designed to grip the
bottom when it is dragged by the ship trying to float away under the
influence of wind and current; usually made of heavy casting or
Fo
casting
Anchorage A place suitable for ships to anchor
Arrived Ship A ship is considered arrived and the laytime can commence when
certain conditions specified in the charterparty are fulfilled, e.g. reach
the designated position for loading or discharging, vessel is ready in
all respects for cargo operation and notice of readiness properly given
Aweigh Describes an anchor which has been lifted off the sea bottom and
has its weight fully taken by the anchor chains
Berth 1. Cabin or place to sleep in a ship;
2. Place for mooring a ship in port or anchoring
y
Cargo Goods carried in a ship
Charterer A person or firm who enters into a contract with a ship-owner for
nl
the transportation of cargo or passengers for a stipulated period of
time, i.e. a ship owner’s customer
O
Charterparty A written contract between ship-owner and charterer whereby a ship
is hired; all terms, conditions and exceptions are stated in the contract
n
Deadweight Tons Total weight of cargo, stores, fuel and water needed to submerge a
io
(DWT) ship from her light draught to her maximum permitted draught; it is
given by the difference between the load displacement and light
at
displacement (also known as lightweight)
Demurrage Fee paid by the charterer to the ship-owner when the latter’s ship is
Draught
ul
detained beyond the specified date agreed in the charterparty
The vertical distance measured from the lowest point of a ship’s hull
irc
to the waterline or the water surface
Free Pratique Official permission from the port health authorities that the ship is
without infectious disease or plague and the crew is allowed to make
C
physical contact with shore; otherwise the ship may be required to

wait at quarantine anchorage for clearance
d
Laytime Time allowed by the ship-owner to the voyage charterer to carry out
the cargo loading and/or discharging operations; laytime may be
ite
expressed as a certain number of days or number of tons of cargo

loaded/unloaded per day
im
Mooring Securing a vessel to a buoy or strong point ashore e.g. bitt by ropes; at
anchorage, by dropping anchor
Notice Of Readiness Notice presented to shipper or his agent by masters or ships’ agent
rL
(NOR) stating the readiness of the arrived ship to load; it determines when
the time starts to count
Pigging Passing a solid plug through a pipeline. The plug is small enough to
Fo
pass through the pipeline but large enough to touch the inside wall.
They have many uses - separating product, cleaning the pipeline,
measuring the pipeline’s wall thickness, etc.
Pilot A qualified person having local knowledge of navigation hazards, is
authorised to guide ships in and out of a port or channel
Ullage 1. Quantity represented by the unoccupied space in a tank or
compartment;
2. Depth of space from the tank top to the free surface of the
liquid;
3. Natural loss in weight or quantity e.g. due to evaporation
y
ABBREVIATIONS
nl
ATF Aviation Turbine Fuel
O
ATM Automatic Teller Machine
BPO Business Process Outsourcing
n
DWT Deadweight Tons
EOSP End Of Sea Passage
io
FO Fuel Oil
at
HSD High Speed Diesel
KT Kiloton ul
LPG Liquified Petroleum Gas
irc
L/A Loading Arm
MTPA Million Ton Per Annum
MS Motor Spirit
C
NOR Notice Of Readiness

PCN Polychlorinated Naptha
d
POB Pilot On Board

ite
SKO Superior Kerosene Oil

SPM Single Point Mooring
im
VGO Vacuum Gas Oil

rL
1. What are the criteria that should be considered during the determination of jetty
capacity?
Fo
2. What is the number of berths required to take care of marine movements at various
throughput levels? Do the sensitivity analysis for your recommendation.
3. Do you agree with the WOL’s idea of not to have more than 65% berth occupancy?
4. What kind of software is required to analyse the situation described in the case?
What would be the functionality of such a software?
5. Suggest an approach, if the number of product tanks is also a decision variable.
Ports and jetty operators use a thumb-rule of a desirable jetty utilization (e.g. 65%) before
further expansion is considered desirable. This, often is the case when the costs of additional
ship waiting time are not borne directly by the jetty operator. In the case of WOL, the
y
investments in the jetty capacity and the demurrage consequences are both borne by the
nl
same entity. Hence, it is useful to develop a trade-off between the investment levels and the
demurrage costs for various throughput levels.
O
At a first level, a process flow chart needs to be constructed, based on the data in exhibits
8A, B & C. An aggregate analysis can now be attempted using deterministic figures to get a
broad relationship between the throughput and the number of jetties. This approach would
n
be largely based on average times. Some average times are not so obvious, for example, the
io
average waiting times for a favourable tide.
For the next level of analysis, the actual distribution of the different time parameters
at
would be essential. A simulation model that considers these inputs along with different
investment levels (e.g. navigation during a wider tidal window, more pilots on call, piping
ul
and pumping capacities and allocation) and evaluates the relevant performance measures
can be used. From this, the various costs incurred such as demurrage, can be estimated.
irc
C
d
ite
im
rL
Fo
CASE 4
y
nl
FarmAid Tractors Limited (FTL) is a tractor company with a 20% market share and
O
aiming to be the market leader within five years. The tractor industry in India has become
very competitive, with growth in capacity outstripping the growth in demand. Customer
n
preferences and demands have changed in the context of the competitive environment. The
role of the dealer and servicing the dealer by the company are recognised by FTL as key
io
success variables in this business. The key issues of concern is to develop a partnership with
the dealers by bringing in a service focus, through improvements in supply chain
at
management and outbound logistics. Two critical areas for analysis are to be taken up,
namely (i) order processing and inventory planning and (ii) distribution structure including
ul
stockyard location. The issue that is taken up specifically is of determining optimal
stockyard location for states in India. The state of Gujarat has been given as an example for
irc
analysis.
C
FARMAID TRACTORS LIMITED

d
1. Introduction
ite
The tractor industry in India had become very competitive, with growth in capacity
im
outstripping the growth in demand. The industry production capacity was in response to a
healthy growth in demand during the nineties. (A brief description of the Indian tractor
industry is given in Exhibit 1). There were seven major players targeting both the domestic
rL
and international markets (Exhibit 2). FarmAid Tractors Limited (FTL), situated near Thane
in Mumbai was the third largest player in FY99 with a 20% market share. FTL was aiming to
be the market leader within 5 years.
Fo
FTL had one factory, 15 models (4 accounting for 90% sales), 18 regional offices (each
with a stockyard - one in a given state), 300 dealers and sold 60,000 tractors per year for
Prepared by G Raghuram. Assistance provided by Ms Sushma Chaudhary is acknowledged. Names

and some of the data have been modified. Teaching material of the Indian Institute of Management,
either correct or incorrect handling of administrative problems. Copyright © 2000 by the Indian
Institute of Management, Ahmedabad. Revised December 2004.
Case 4: Farmaid Tractors Limited 337
the past two years (Exhibit 3). FTL was relatively a new entrant in the industry, having
started in the early nineties. The promoters were familiar with the automobile industry,
and had competencies in the auto components business.
Customer preferences and demands had changed in the context of the competitive
environment. The role of the dealer and servicing the dealer by the company were
y
recognised by FTL as key success variables in this business. The key issues of concern were
to develop a partnership with the dealers by bringing in a service focus, through
nl
improvements in supply chain management and outbound logistics.
O
In summer 1999, as a part of a major effort in organisational restructuring and business
process reengineering, FTL decided to engage the services of Prof Prashanth from a
reputed institute of management in Western India for a consultancy assignment. The
n
assignment was on supply chain management, focused on outbound logistics. After an in-
depth study of FTL (the manufacturing, production planning and despatch processes), and
io
some of the regional offices, stockyards and dealers, various issues and decision areas were
identified by Prof. Prashanth. Exhibit 4 gives a list of the issues, and decision and action
at
areas which were planned to be analysed and for which data was sought from FTL. An in-
company logistics team of four young executives, coordinated by Mr Rajesh Bhatt,
ul
supported Prof. Prashanth in this exercise.
Two critical areas for analysis were first taken up, namely:
irc
(i) Order processing and inventory planning
(ii) Distribution structure including stockyard location
C
2. Order Processing and Inventory Planning

d
Orders were placed by regional offices for delivery to stockyards, model-wise on a monthly
basis. Orders had to be consolidated, based on dealers’ requirements by the 20 th of a
ite
month for receipts during the following month. Over the next five days, there could be
discussion between the plant (production planning) and the regional offices to modify the
order, keeping in view any possible production constraints. Even though the production
im
(and the dispatches) were planned and scheduled for a whole month, there were always end
of month pressures for modifications and additions. FTL marketing executives and the top
management were sensitive to the monthly market share points that industry analysts
rL
watched and reported.

Inventory planning at the stockyards to enable high service levels to the dealers, and at
the plant to respond to seasonality were key concerns. The forecasts that would drive this
Fo
planning also needed to be examined. Exhibit 5 gives the framework proposed by Prof.
Prashanth to address these.
3. Distribution Structure Including Stockyard Location

There were two concerns here, namely, the need for a central despatch yard and location
(and number) of stockyards. Exhibit 6 gives relevant issues as part of the consultant’s
framework of addressing these concerns.
Currently, all despatches were made from the factory to the stockyards through a daily
allocation process which took into account the unmet demand at stockyards, ready for
despatch inventory and availability of trucks based on the transporters’ inputs. Once
assigned to a transporter, the tractors were moved by the transporters to their godowns,
since there was no space in the factory for holding finished stocks. It was often noticed that
the transporters actually moved the tractors out of their godowns after an average of two
y
days, primarily due to the non-availability of the intended trucks for despatch. FTL was
nl
concerned about this since the period of “lack of control” over the tractors was enhanced
due to this. A central despatch yard at a suitable highway location, 20 kms away from the
O
plant was being considered to address this. (There was no space nearer or adjacent to the
existing plant for such expansion).
For better servicing of dealers, location of stockyards was another issue. The primary
n
transportation mode was decided as road, using long platform trucks that could carry up to
io
five tractors. The secondary transportation from the stockyards to the dealers would now
begin to use trucks that could carry two tractors rather than move on own power. This was
at
expected to reduce transit damages and also offer the tractor in a mint condition to the
dealer. One of the debates was whether a stockyard should be close to the entry point in a
state or close to the marketing office, which was usually in a large commercial centre near
ul
the centre of the state. For an in-depth analysis, Exhibit 7 gives the potential stockyard
locations for Gujarat, with the monthly operating costs and distances. The current
irc
stockyard location was Ahmedabad. Exhibit 8 gives the location of the 19 FTL dealers in
Gujarat, along with the expected monthly demand and distances. The total Gujarat demand was
expected to be 500 tractors per month. Exhibit 9 gives a map of Gujarat showing all locations.
C
The primary transportation cost was expected to vary from Rs 2.5 to Rs 3.0 /tractor/
km, depending on the truck technology. The secondary transportation cost was expected to
d
vary from Rs 3.0 to Rs 3.5 /tractor/km depending on the service level offered. It was also
a matter of concern that a dealer should not have to be more than 500 kms from a
ite
stockyard. Some of the more aggressive marketing executives felt that this should not
exceed even 350 kms, which would be a one day transportation lead time. Another issue
was a possible minimum on what a stockyard should handle per month, especially if
im
stockyard management were to be outsourced. The company executives felt that a number
of 200 tractors per month was a reasonable figure, to enable it to be attractive to the
rL
outsourcee.
Exhibit 10 gives the results (optimal stockyard locations with total costs) of the analysis
based on a math programming model for various scenarios of parameters affecting the
Fo
stockyard location. The top management was interested in the actual allocation of dealers
to stockyards for these scenarios and their implications in terms of the criteria considered.
The final recommendations for stockyard locations of the major states are given in Exhibit 11.
4. Supply Chain Organisation

To ensure better supply chain coordination for higher service levels to the dealers and
customers, it was also decided that the distribution unit right upto stockyards would be under
a new supply chain organisation which would include the production units (Exhibit 12).
The overall spirit of the service orientation sought to be achieved by better supply chain
management was communicated in a letter (Exhibit 13) to the logistics team members in
response to certain queries.
y
Exhibit 1: The Indian Tractor Industry
nl
In 1999, the Indian economy was still highly dependent on agricultural growth and not
surprisingly, it was the largest tractor market in the world. However, in terms of total tractors
O
in use in the country, it was eighth in the world. The country had a tractor density of 10.5
tractors per thousand hectares of Gross Cropped Area (GCA) compared to the international
average of about 28 tractors per thousand GCA.
n
io
Demand and Supply
at
The tractor industry was segmented based on the power of the tractor engine expressed in
terms of horse power (HP). Among the segments, 31-40 HP segment led with 54.7% of total
ul
tractors sales in FY2000. Among the regions, North India constituting UP, Punjab and
Haryana led with a contribution of 44% of tractor sales for FY99. The demand for tractors
has increased from 121,106 tractors in FY90 to 260,762 tractors in FY2000 at a CAGR of
irc
8%, but the annual growth in demand varied substantially. For example, in FY99 and
FY2000, the rise in sales was reducing with growth rates of 3% and 2.9% respectively.
C
During the last 20 years from FY78 to FY98, the tractor industry had performed
substantially better in comparison to the growth in Indian agriculture production and GDP.
As given in Table 1 below, the CAGR of tractor sales was almost four times the growth of
d
agriculture production.
ite
Table 1: Tractor Sales v/s GDP Growth

im
Particulars CAGR (FY78 to FY98)
Agriculture production 2.2%

rL
GDP 5.15%
Tractor sales 8.31%
The major factors influencing demand of tractors were monsoon, land-holding pattern,
Fo
availability of credit, growth in income of farmers and implementation of scientific farming

practices. The industry was not cyclical, as many would presume about the automotive
industry in general. Demand contraction occurred in 1982 and again in 1993, the two years
of severe credit squeeze.
The optimal land-holding required for different HP tractors was approximately 8 to 10
hectares for 1800 cc or 25 HP tractors, 25 to 30 hectares for 35 HP and beyond that, it
was higher HP tractors. But, the actual relationship between different land-holdings and
the power of tractors was dependent on earning stability, type of soil, type of operation and
affordability by farmers.
As per FY91 data, nearly 78.2% of India’s agriculture land was held by small and
marginal farmers having less than the prescribed land holding. This had gone up from
76.2% in FY86. The all-India average land-holding figure stood at 1.55 hectares in FY91,
y
down from 1.69 hectares in FY86. For these small and marginal farmers, there was very
little potential for economic use of tractors by outright purchases.
nl
Industry Structure
O
The HP-wise composition of tractor industry sales as shown in Table 2, reveals that 31-40
HP tractors constituted the largest segment with 54.7% of the total tractor sales in
n
FY2000. It can also be seen that demand for the less than 30 HP segment, which used to be
io
the largest in FY94, had been pushed back by the 31-40 HP segment.
at
Table 2: Segment-Wise Market Shares
Segment FY94
ul FY99 FY2000
< 30 HP 36% 22.5% 26.1%

irc
31-40 HP 46% 57.8% 54.7%
41-50 HP 15% 15.9% 14.7%
> 51 HP 3% 3.8% 4.4%
C
This graduation to a higher HP segment can be explained if one considers the fact that
region-wise share of tractor sales had also shifted from the northern states to other parts of
d
the country. Initially, the population of tractors was concentrated in states like Punjab and
ite
Haryana which received the benefits of the so-called ‘Green Revolution’. The soil in these
states were alluvial in nature and thus required a low powered tractor for tilling it.
However, over the years, with an increase in irrigated cropped area, the population of
im
tractors began to increase in other states as well. These included states located in the
western and southern parts of the country where the soil - laterite soil, black soil etc - was
harder and needed higher powered tractors.
rL
From the table above, we see that in FY2000 the share of the less than 30 HP segment
improved by around 3.5% while that of the next higher segment fell by 3%. This could be
attributed to the fact that in the 1999-2000 budget, the excise duty for the plus 1800cc
Fo
tractors (above 30 HP) was increased from 13% to 16% while that of the less than 1800cc
segment remained at 8%.
Region-Wise Share of Tractor Sales

North India constituting UP, Punjab and Haryana led among the regions, by contributing
44% of tractor sales in FY99, the data for which is presently available. But, its contribution
had come down in the last five years from a peak level of 53.5% in FY94. Among the
states, UP stood first with a contribution of 23% of the country’s tractor sales. MP and
Punjab stood next with around 12-13% contribution each.
The state-wise sales figures over the years show some important changes (Table 3). Sales
from agriculturally developed states like Punjab, Haryana and Uttar Pradesh as a part of
total tractor sales had come down from 53.5% in FY94 to 44% in FY99. This contrasts
y
with the increase in share for the central and western regions of the country (Rajasthan,
nl
Madhya Pradesh, Maharashtra and Gujarat). This region’s share rose from 20.4% to
27.8% in the same period. Sales in the northern regions had been below the all India
CAGR of 8.6% in the last ten years. On the other hand, the central and western regions of
O
the country had recorded double-digit growth rates.
n
Table 3: State-Wise Market Shares
io
States FY90 FY94 FY98 FY99 CAGR
(FY90 to FY99)
at
Punjab 22,026 26,636 31,644 31,047 3.9%
Haryana
UP
15,307
31,233
16,579
30,656
ul 22,711
50,433
21,877
59,211
4.0%
7.4%
Bihar 4,891 2,900 10,282 12,875 11.4%
irc
Rajasthan 9,315 11,129 25,152 24,054 11.1%
MP 10,018 11,418 32,250 32,711 14.1%
Andhra Pradesh 5,395 5,881 11,171 9,862 6.9%
C
Gujarat 6,508 10,033 21,501 22,208 14.6%

Maharashtra 5,737 6,730 15,059 16,011 12.1%
d
Karnataka 3,665 5,179 10,176 6,501 6.6%

Tamil Nadu 3,864 7,030 10,392 7,369 7.4%
ite
Others 3,147 3,887 9,608 11,145 15.1%

All-India 121,106 138,058 250,379 254,871 8.6%
im
As seen in Table 4, the increased growth from central and south Indian states had led to
growth in the medium and high powered tractor sales, as they had harder soil compared
with the alluvial soil in the northern states. In the future, new tractor sales were expected to
rL
take place mainly in the western and southern regions of the country while the mature
markets of the north would see higher replacement sales.
Fo
Table 4: Major Markets for Tractors
Tractor power (HP) Major markets
Less than 30 HP UP, MP, Rajasthan, Haryana and Bihar

31-40 HP MP, Rajasthan, Punjab, UP, Haryana, Gujarat and Bihar.
41-50 HP Punjab, MP and Rajasthan.
More than 51 HP Punjab and also some volumes in Kerala and Maharashtra
Source: SIAM/TMA
Exhibit 2: The Key Market Players
As of 1999, the industry was controlled by seven major players, namely, Mahindra &
Mahindra (M&M), Escorts, FarmAid Tractors Limited (FTL), Punjab Tractors Limited
(PTL), TAFE, Eicher and HMT. All except FTL had been in the tractor market for more
y
than 20 years. M&M continued its leadership position for the sixteenth successive year with
nl
a market share of 25% in FY2000. Escorts had taken the second position from PTL with a
market share of around 20%. Escorts was followed by FTL (20%), PTL (15%) and TAFE
(12%). While M&M had a balanced presence in all the HP segments, except that of less than
O
20 HP, others like PTL were stronger in the low and medium powered segments.
In recent years, the tractor industry had seen some new entrants introducing their
n
products in the higher HP range. These included Ford New Holland and L&T John Deere.
This had added around 65,000 units to existing capacities. In FY99, capacity stood at
io
approximately 350,000 units. This meant a capacity utilisation of around 72%, down from
around 87% in FY98. This pointed to a growing problem of overcapacity in the industry.
at
Many of the players had strong linkages with the automobile industry. Recently, one of
the largest and the oldest players in the tractor industry, Escorts Ltd merged itself with
ul
Escorts Tractors Ltd after buying out the stake of its foreign collaborator Ford New Holland.
irc
Mahindra and Mahindra Limited (M&M)
C
M&M was the world’s largest tractor manufacturer. It was the largest jeep and tractor
manufacturer in India with a tractor market share of 27%. Achieved economies of scale due
to these associated operations in LCV and Utility Vehicles. Strong distribution network and
d
brand equity. Recently embarked on a Business Process Reengineering Programme to achieve

ite
reduction in the production time, costs and improvements in product quality. One of the
core competencies of M&M lay in the engineering skills and its R&D which had helped the
company develop a diversified product range. They had acquired a majority stake in Gujarat
im
Tractors Limited.
Escorts
rL
With the recent merger, Escorts has emerged as the second largest player in the Indian tractor
market with a market share of 20%. (Escorts may not have substantial economies of scale due
Fo
to differences in the products of the two companies). It was also poised to takeover the
tractors division of public sector HMT Ltd. Engaged in a restructuring exercise, resulting in
the integration of the entire tractor business in one company and the other businesses like
two-wheeler, auto components, telecom etc in separate companies. The distribution
network, though widespread in the North, was weak in the emerging markets of South and
West. The highly successful brand name FORD had been withdrawn due to the divestment
of the stake of Ford New Holland in favour of Escorts. Now, the tractors were being
marketed in the name FARMTRAC and the change had been well advertised among the
customers. Escorts had seen declining sales, had lower margins than the industry and had a
very high debt/equity ratio.
FarmAid Tractors Limited (FTL)

Had a presence in all market segments. Had a 20% market share and was aiming to be the
y
number one player in another five years, achieving a market share of 30%, either by
nl
expansion or by acquisition.
O
Punjab Tractors Limited (PTL)
Strong distribution network and brand equity. Had adapted its tractors to suit specific state/
n
crop conditions. Had been growing at a faster rate than the industry. Had one of the highest
operating margins in the industry. Sales were against advances. It had a working capital
io
surplus. Well positioned in the > 40 HP segment to exploit future growth opportunities.
at
TAFE
ul
Closely held company of the Amalgamation Group. It had the strongest presence in the 30-
40 HP market segment, but had no presence in either < 30 HP or > 40 HP segments. The
irc
company had substantial presence in all the major markets, though, its earlier focus was
Southern India where the competition was weak. Setting up new facilities to manufacture
13,000 tractors per year. Most of the vendors were well known automobile ancillary group
C
companies like Bimetal Bearings, India Pistons etc.
Eicher
d
ite
All the sales were limited to < 30 HP category which had seen declining market share, though
the absolute sales may be increasing. Distribution was limited to Haryana, Punjab and
Western UP. There was a proposed merger with Royal Enfield, which could provide access to
im
Enfield’s network in South. Had one of the lowest net profit margins in the industry.
HMT
rL
Had a presence in all the market segments. Despite very good product quality, had not been
a very strong marketing company. Had been experiencing a declining market share, although
Fo
volumes were increasing. Had not been strong in the Green Revolution states of Punjab,
Haryana and Western UP. It had not been competitive on prices. Proposed takeover by
Escorts Ltd though the final approval was pending with CCEA.
As seen in the table below, production capacities in the tractor industry had increased
with the coming in of John Deere and Ford New Holland last year. However, this increase
had not kept pace with a corresponding increase in production. Capacity utilisation in
FY99 stood at around 72%, clearly indicating a growing problem of overcapacity in the
industry. In FY2000, the utilisation was expected to drop to 64%. In FY01, an additional
17,000 units is likely to be added to New Holland’s capacity while another 20,000 units
was expected to be added to John Deere’s capacity in the next two years. FTL was also
planning to expand either by adding capacity or through acquisition.
Production Capacities
y
Company FY99 FY2000 (Expected)
nl
Bajaj Tempo 6,000 6,000
Eicher 45,500 45,500
O
Escorts 56,250 56,250
FTL 60,000 65,000
HMT 18,000 18,000
n
M & M 77,000 89,500
io
Punjab Tractors 50,000 72,000
TAFE 37,800 37,800
at
L&T John Deere 0 10,000
Ford New Holland 0 18,000
Total ul 350,550 418,050
Total production 253,904 267,356
Source: TMA/Annual reports
irc
C
Exhibit 3: FTL Tractor Production
FY Numbers
d
1996 45000
ite
1997 53000
1998 60000
1999 60000
im
2000 65000
(Expected)
Source: Company Records
rL
Fo
Exhibit 4: Logistics Issues and Decision Areas
1. Objectives of FTL’s logistics: To maximise market share and profitability in a

competitive environment, by enabling availability of the right model in best quality at
the right place at the right time.
2. Symptoms/issues
a) Stockouts
i) Inability to supply the specific models indented by the dealers (end-user

impact: lost sale/forced conversion to another model)
ii) Delayed supply of the specific model indented by the dealers (end-user impact:
loss of goodwill)
iii) Stockout cost considered anywhere from 50% of contribution to 200% of
y
contribution.
nl
b) Excess stocks
i) Inventory carrying cost: working capital cost of Rs. 100 per tractor per day (@
O
0.05% per day on a tractor worth Rs. 200,000)
ii) Scope for damages
n
iii) An average of 20 days stock before sale to dealer costs Rs. 2000 per tractor (Rs
12 crores per annum)
io
iv) An average of 20 days stock and a further 15 days credit by the dealer costs the
dealer Rs. 3500 per tractor (Rs 21 crores per annum) (a dealer typically earns
at
Rs. 8000 per tractor as commission, and possibly Rs. 1500 as 15 days credit,
apart from margins on accessories)
ul
c) Inter-stockyard transfers
i) Extent during 1997/98
irc
ii) Reasons, costs
d) Losses: transit/storage resulting in repair/replacement/replenishment (70% of
C
tractors received a ‘yellow’ card on receipt at dealers - implying not ready for sale.
75% of these were set right in the first week. The remaining sometimes got
complicated in ‘investigations’, resulting in non-settlement of claims/dues even
d
upto four years)

ite
i) Losses chargeable to transporter

ii) Losses chargeable to insurance
im
iii) Other losses, including opportunity cost of sale

iv) Modewise losses
e) End-of-month pressures
rL
i) Inventory carrying cost associated with skewed sales to dealers

ii) Cost of communication
Fo
f) Need to revise production schedules

i) Extent during 1997/98
ii) Scope for improvement through better forecasting and planning
g) Transportation costs
i) Average primary (Rs 2500 per tractor, Rs 15 crores per annum)
ii) Average secondary (Rs 1500 per tractor, Rs 9 crores per annum)
3. Decision and action areas

a) Central despatch yard
i) Allocation based on truck availability: savings in inventory
ii) Increased flexibility due to later allocation across destinations: savings in
inventory
y
iii) Loading and storage under supervision: reduced losses
nl
iv) Infrastructure cost
O
v) Availability of trucks/movement to railsiding
b) Stockyard locations
i) Stockyard in every state due to tax savings
n
ii) Since secondary transportation cost is not significantly higher than primary
io
transportation cost, entry point in the state is often the best
iii) Multiple stockyards will be justified based on total transportation cost (primary
at
and secondary), limitations on secondary movements (if on own power) and
stockyard infrastructure and management cost (each stockyard costs about
Rs 25,000 per month) ul
iv) Stockyard to have minimum throughput volume (say 200 tractors per month)
irc
v) Stockyards to be managed by C&FA
c) Mode choice
C
i) Transportation Cost
ii) Inventory cost at unloading end
iii) Inventory cost at loading end
d
iv) Inventory cost due to pipeline

ite
v) Cost due to losses

vi) Buffer inventory cost at unloading end due to reliability in transit
im
vii) Availability of transport/lead time for availability

d) Inventory norms
rL
e) Forecasting and planning systems

i) Current system
ii) Revised system with rolling horizon
Fo
f ) MIS for stockyard management

i) Selection of C&FA
a) Experience/familiarity with tractor business
b) Professional outlook of management and employees
c) Understanding of transport business
d) Skills for inspection and documentation on receipt and before delivery
e) Security infrastructure, preparedness for contingency

f ) Infrastructural support: land for ordered storage and retrieval, fax,
telephone, loading/unloading ramp
g) Professional charges
h) Time-frame for contract
y
ii) Monitoring of C&FA
nl
a) Proper record keeping
b) Proper execution of instructions
O
c) Orderly storage and retrieval (in full communication with area office)
d) Damages during unloading/storage/despatch
n
g) MIS for transporter assessment
io
i) Selection of transporter
a) Quality in delivery (accidents, thefts)
at
b) Reliability in transit time (average, variance)
c) Price
d) Number of trucks owned
e) Nature of financing
ul
irc
f ) Type of drivers employed
g) Permits invested in
C
ii) Monitoring of transporter

a) Date of despatch, date of delivery, variance analysis,
b) Transshipments onroute
d
c) Report daily movement

ite
d) Spot contingency due to accident (role of central despatch/nearest C&FA/

destination C&FA)
e) Stock allocation during contingency by central despatch
im
f ) Despatchwise
g) Transporterwise
rL
h) Modewise
h) Need for transport development
i) Loading and unloading practices
Fo
ii) Truck design

iii) Financing of transporters
iv) Coordination with other tractor manufacturers for return loads
i) Inputs for product development
i) Avoid protruding parts for better packing efficiency and reduction of damages
ii) Kitting of loose parts
Exhibit 5: Order Processing and Inventory Planning
Regional Office to Factory (Stockyard-wise, Model-wise)

✧ Monthly inventory plan with safety stock for stockyards. Safety stock to provide 98%
y
service level, model-wise
nl
✧ Order quantity = expected demand during protection interval + safety stock - current
stock (including stock in transit)
O
✧ Protection interval = Order interval (monthly) + lead time (0.8-1.3 months)
✧ Attempt to reduce end of month skew and move to weekly ordering, with two weeks
n
lead time
io
Seasonality
at
✧ Annual inventory plan for seasonality (uniform production versus demand driven, if
required by using overtime and/or subcontracting).
ul
irc
Month Demand Production Inventory due to
uniform production
January 5,000 5,000 1,100
C
February 4,000 5,000 2,100

March 4,500 5,000 2,600
April 6,000 5,000 1,600
d
May 5,900 5,000 700

June 5,700 5,000 0
ite
July 4,500 5,000 500

August 4,000 5,000 1,500
September 4,500 5,000 2,000
im
October 5,500 5,000 1,500

November 5,400 5,000 1,100
December 5,000 5,000 1,100
rL
Total 60,000 60,000
Forecasting
Fo
✧ Company level (for aggregate and seasonality planning)

✧ Regional office level (consolidation of dealer forecasts and cross checking the same,
for placing orders from the factory)
✧ Dealer level (customer tracking)
Exhibit 6: Distribution Structure
Central Despatch Yard

✧ Investment cost: Rs 15 million
y
✧ Operating cost: Rs 2 million per year
nl
✧ Flexibility in allocation after physical verification of truck availability
O
✧ Inventory saving over current practice: 2 days (1,20,000 tractor days per year) i.e. Rs
12 million per year
n
Some Drivers for Secondary Logistics
io
✧ Change from own power to truck based movement
✧ Stockyard near dealer
at
✧ B/C category inventory to be held by stockyard only
Stockyard Location Analysis

ul
irc
✧ Due to 4% Central Sales Tax considerations, by and large each state had to have
atleast one stockyard
✧ Current policy for stockyard locations was proximity to regional marketing office,
C
which was usually in a major city in the centre of the state

✧ Both the number and location of stockyards were questioned and relaxed where
d
logistical servicing of dealer gained importance

ite
✧ A math programming model could be used for analysis

im
Exhibit 7: Stockyard Location, Cost and Distance

rL
Sr Stockyard Operating Cost Distance from

No Location per Month (Rs) Thane (Kms)
1 Valsad 25,000 136

Fo
2 Surat 20,000 263

3 Vadodara 30,000 448
4 Ahmedabad 30,000 545
5 Rajkot 25,000 761
y
350
nl
Exhibit 8: Dealer Location, Demand and Distance
O
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

n
Surendranagar
Dealer
Himmatnagar
Location
Dharampur
Bhavnagar
io
Porbandar
Jamnagar
Junagadh
Mehsana
Palanpur
Godhara
Bharuch
Rajpipla
Nadiad
Bardoli
Dholka
Anand
Amreli
Patan
Morbi
at
No of Tractors/
ul
Month 35 30 25 40 25 20 20 20 35 20 30 45 20 20 20 30 20 25 20
1 Valsad 633 272 62 163 514 32 385 315 424 616 630 293 419 647 491 456 715 204 431
irc
2 Surat 566 205 31 96 447 109 318 248 357 549 563 226 352 570 424 389 648 141 364
3 Vadodara 399 38 125 71 280 266 151 81 190 382 396 59 185 403 257 238 481 82 200
C
4 Ahmedabad 258 73 225 182 200 377 40 136 79 313 327 52 74 292 146 125 412 195 116
5 Rajkot 105 255 492 365 175 560 162 321 304 88 102 234 299 67 371 255 187 255 111
* All distances are in Kilometers.
d
ite
im
rL
Fo
Exhibit 9: Stockyard and Dealer Locations
y
nl
O
n
io
at
ul
irc
Case 4: Farmaid Tractors Limited

C
d
ite
im
rL
351
Fo
Exhibit 10: Scenario Analysis for Gujarat
y
352
(Rs)
nl
Cost/tractor/km Current Secondary Secondary Secondary Secondary distance limit Secondary distance limit:
distance limit: distance limit: distance limit: none Minimum no of 500 kms Minimum no of
O
None 350 kms 500 kms tractors to be serviced tractors to be serviced
by a stockyard: 200/month by a stockyard: 200/month
Primary: 2.5 10,28,999 8,73,533 8,78,209 8,75,454 8,75,484 875,484

n
Secondary: 3.5
io
Ahmedabad Valsad Valsad Valsad Valsad Valsad
Rajkot Ahmedabad Ahmedabad Ahmedabad Ahmedabad
Rajkot
at
Primary: 3.0 11,19,855 8,22,880 9,43,085 8,87,380 8,22,800 8,99,080
Secondary: 3.0
ul
Ahmedabad Valsad Valsad Valsad Valsad Valsad
Ahmedabad Vadodara Vadodara
irc
Rajkot
C
Exhibit 11: Recommendations for Stockyard Locations
d
State Existing Yard(s) Optimal Locations State Existing Yard(s) Optimal Locations
ite
Andhra Pradesh Hyderabad Hyderabad Madhya Pradesh Bhopal Indore
Vijaywada Raipur
im
Tamil Nadu Chennai Hosur Rajasthan Jaipur Kota

Trichy SriGanganagar Jodhpur
SriGanganagar
rL
Karnataka Bangalore Belgaum Punjab Jalandhar Patiala

Davangere
Gujarat Ahmedabad Valsad Haryana Karnal Gurgaon
Fo
Ahmedabad
Fo
Exhibit 12: Organisation Structure
rL
im
ite
d
C
irc
ul
at
Case 4: Farmaid Tractors Limited
io
n
O
353
nl
y
Exhibit 13
November 2, 1999
Mr Rajesh Bhatt
Logistics Team
y
FarmAid Tractors Limited
nl
Thane, Mumbai
Dear Rajesh,
O
Regarding the restructuring of the supply chain, I have the following points.
1. The entire supply chain should service the customer requirements in a coordinated
n
way. We aim to be like a “service” organisation.
io
2. We build on the premise that marketing’s role is to be aware of the customer
requirements, and interface with the customer, both directly as well as through the
at
dealer. Hence, ideally, supply chain’s responsibility should be up to servicing the dealer
requirements. Thus, dealer and dealer customer interface becomes the front office
function, servicing the dealer right from the raw material vendor through a series of
ul
transport, storage, and conversion activities is the back office function.
irc
3. Given the above ideal supply chain structure, we can start examining activities prior to
reaching the dealer, one by one, to see if there would be better effectiveness and
efficiency if the activity was performed by the marketing (front office) or by supply
C
chain (back office). The activities in reverse order would be

(a) transportation from area stock yard to dealer
d
(b) allocation from area yard to dealer

(c) receipt, storage, inspection at area yard
ite
(d) transportation from central despatch to area yard

(e) allocation to area yard
im
(f ) transportation from finished goods to area yard

(g) ready for despatch inspection at the plant and so on.
rL
4. If (a) above is better performed by marketing, then it could be part of marketing role.
Apart from this activity, it appears that all the other activities should be a back office
function under supply chain, so that complete responsibility is taken to ensure
Fo
availability of the required tractors. Focus on activities like transportation, storage,

inspection even at geographically dispersed places should reach the same level as a
particular manufacturing process in a factory environment.
Thanking you,
Yours sincerely,
Prashanth
1. What are the factors leading to the perceived ‘lack of control’ and poor delivery
quality?
2. What forecasting technique should be used for inventory planning at the plant and the
y
stockyards?
nl
3. Is there a need for a central dispatch yard? What are the pros and cons?
4. What is a good model for determining optimal location of stockyards and the
O
associated allocation of dealers to the stockyards in the state of Gujarat?
5. Interpret the implications of the recommended locations versus the existing locations
n
as given in Exhibit 11.
io
at
Quality, as determined by dealers, is an important driving element in the new supply chain.
It throws up a number of steps and stages of supply which lead to poor quality. These need
ul
to be mapped out.
The demand forecast plays a major role as the next step in the whole analysis. Forecasts
irc
are made at three levels; at the national level, across all models, for overall capacity planning
at the factory; regional forecasts across dealers in a region so that regional stockyards can
C
order; and finally, dealer level forecasts. Seasonality needs to be addressed properly for the
required service levels. Dealer level forecasts may have their limitations, and alternatives to
this have to be considered.
d
The need for a central dispatch yard can be assessed from the perspective of inventory
ite
costs, quality costs and control issues.

The case has a number of scenarios in which to analyse the optimal stockyard location
with transportation costs (primary and secondary) and stockyard location costs being the
im
main factors. Constraints like minimum number of stockyards in a state and minimum
number of tractors to be handled by stockyards have to be taken into account too.
A mathematical program can be formulated for solving this problem using the demand
rL
forecasts and cost estimates that are available. This model is usually of the location-
allocation type, and would decide the locations of stockyards together with the allocations of
dealers to these stockyards. Exhibit 10 gives the outcome through a model of this type.
Fo
The optimal stockyard locations in the various states, driven primarily by costs, could be
different from the existing locations, which are based on proximity to the marketing offices.
The optimal locations may, therefore, need a buy-in, for which facilitating mechanisms for
improved communication and co-ordination would be required.
In the context of the above issues, the supply chain organization may also need re-
structuring, to achieve better co-ordination between production, order processing and
dispatching.
CASE 5
y
nl
The Senior Vice President, Marketing, of Titan Industries Limited (TIL) was examining
O
the domestic growth scenario of Titan watches. The sales of TIL had dropped 11 per cent
over the previous year. The traditional strengths of Titan had been in marketing, especially
n
retailing, brand building and product design. To continue to remain profitable, the key
attributes were recognised as continuous product innovation, and ensuring availability of
io
the desired product or at least range. The top management led by the Senior Vice-President,
Marketing felt that better supply chain management would be one key strategy and would
at
lead to better professional practice within the organisation.
ul
irc
SUPPLY CHAIN MANAGEMENT at Titan Industries Limited
C
Bhaskar Bhat, Senior Vice President-Marketing of the high growth, high profile Titan
Industries Limited, was examining the domestic growth scenario of Titan watches. Year-to-
date sales as of August 1996 showed a drop of 11 per cent over the previous year. While some
d
of it could be attributed to the general slowing down of the economy, there was also the
lurking suspicion that growth in this segment of the high-end quartz analog watch market
ite
may be slowing down, owing to saturation and increasing competition from foreign entrants,
thanks to liberalisation.
im
The traditional strengths of Titan have been in marketing (especially retailing – making
watch buying a pleasure in itself ), brand building, and product design. To continue to
remain profitable and keep the market happy, the key attributes in test in the current context
rL
were recognised as continuous product innovation (offering dynamic variety over the already
large variety offered), and ensuring availability of the desired product or at least range.
Further, any expansion or related diversification required cash for which one possible source
Fo
was the large inventory (estimated at one year’s requirement from raw material to retail
outlet). The attention of top management led by Bhat naturally moved on to better supply
Written by G. Raghuram. Cases of Indian Institute of Management, Ahmedabad, are prepared as a

basis for class discussion. Cases are not designed to present illustrations of either correct or incorrect
handling of administrative problems. Copyright © 2004 by the Indian Institute of Management,
Ahmedabad Sales (excerpted from the Annual Report 1995-96).
Case 5: SCM at Titan Industries Limited 357
chain management as one key strategy. Apart from addressing many of the concerns raised
above, Bhat felt that better supply chain management would lead to better professional
practice within the organisation.
Background
y
Titan was set up in mid 1985 by the Tata group, bringing in a few select top executives from
nl
other Tata companies (including Bhat) along with a few executives from HMT (the public
sector watch manufacturing company). After obtaining approvals, Titan set up the
O
manufacturing unit in Hosur, Tamil Nadu, close to Bangalore, on the Bangalore-Madras
national highway, just across the border from Karnataka. (The Tamil Nadu government
offered incentives to industry to locate in Hosur, leveraging on the proximity to Bangalore.)
n
Titan entered the Indian watch market during 1986-87 by just assembling watches and
io
increased its manufacturing content through a phased manufacturing programme.
The appendix provides some basic information on world watch production, the Indian
at
watch market, a fact sheet on the company, and excerpts from the annual reports on
Government policy. Following the appendix, the organisation chart of the company as well
ul
as the manufacturing and marketing setups are enclosed. This is followed by a chart showing
the manufacturing structure of a watch.
irc
In spite of the Indian watch market growing by only 5 per cent, Titan continued its
dominance, selling a total of 3.40 million watches in the domestic market as against sales of
3 million watches in the previous year. In addition, 4,30,000 watches were exported.
C
At the end of March 1996, Titan had a market share of close to 50 per cent of the quartz
analog watches manufactured and sold in India. Together with Timex watches, which were
d
entirely sold and marketed by Titan, the combined market share stood at 70 per cent of the
Indian made quartz analog watches and close to 25 per cent of the estimated 21 million
ite
watches sold in India during the year, including mechanical, quartz, digital, and electronic
watches from indigenous and imported sources, both legal and otherwise.
Titan’s chain of exclusive retail showrooms increased from 80 to 86 and the number of
im
Titan shops was 90 as of March 31, 1996. During 1996-97, Titan would further strengthen
its retail network. The number of outlets would be significantly increased, showrooms
rL
upgraded in terms of general appearance, merchandise and service, and the Titan shop retail
chain raised to a new level altogether. Titan shops would be rechristened “Timezone” and
would sell multiple brands negotiated by the company. Every shop would be upgraded, as
with the showrooms, and the Timezone chain would be heavily promoted as a chain of
Fo
trusted watch shops retailing a wide variety of reliable, reputed, and high quality watches and
clocks and offering excellent service.
Titan was also in the process of setting up a joint venture with Hour Glass of Singapore,
one of the world’s largest wholesalers and retailers of luxury watches. A memorandum of
understanding to this effect had already been signed and the modalities, including the
proposed investment and the brands which Hour Glass in India would carry, were in the
process of being firmed up. The new company would set up watch boutiques in leading
Indian cities for the sale of luxury watches, mainly Swiss, and would also be responsible for
wholesale and after-sales operations.
With cumulative sales of close to 20 million watches since 1987, the company was laying
renewed emphasis on after-sales-service. Titan was constantly increasing the number of
service outlets and exploring ways and means of upgrading these outlets to enable it to
y
continue to provide high quality service and thereby ensure total customer satisfaction.
nl
The success achieved by Titan in marketing excellence continued to receive recognition.
During September 1995, Titan was once again voted “the most admired consumer durable
O
company in India” in a poll conducted by the A & M magazine. Besides winning numerous
gold medals under various categories at the Bombay, Madras, and Bangalore Ad Clubs award
functions, Titan won the very prestigious “campaign of the year” awards presented by the
n
Bombay and Bangalore Ad Clubs for its Tanishq advertising campaign, developed by Lintas.
Titan also figured in the Far Eastern Economic Review’s survey of Asia’s 200 leading
io
companies, ten of which were Indian.
at
Despite its many success, in the words of the Managing Director, Xerxes Desai, “Titan
recognises the need to further invigorate its sales and marketing organisation and strategies in
view of the new competitive threats which are emerging. A new programme has been
ul
initiated to combat such threats and improve the ‘sell through’ rates at retail. At the same
time, the Directors would like to extend their very special thanks to its prime business
irc
associates consisting of franchisees, stockists, dealers, and after-sales service providers for the
significant contribution which they have made over the years to the company’s success. The
shareholders, directors, and management of Titan Industries count on their unqualified
C
support as the company completes the first decade of its operations and commences another
decade which will be marked by the emergence of new competitive forces in the Indian
d
market. We believe that close cooperation between Titan’s central selling organisation and its
key business partners will enable the company to maintain its pre-eminent position in the
ite
Indian market.”
Manufacturing (excerpted from the Annual Report

im
1995-96)
rL
Titan’s two new plants, the eurowatch plant and the jewellery plant, had their first full year of
commercial production. Production was progressively stepped up at both plants – the
eurowatch plant producing about 80,000 watches, and the jewellery plant producing almost
40,000 jewellery pieces and around 2,000 jewellery watches.
Fo
The eurowatches, launched this year in many European markets, had been extremely well
received by the trade as well as the end customer. The company expected the sale of these
watches in international markets to progressively increase in the coming months and had
taken steps to further increase manufacturing capacity to meet this demand.
Titan totally produced 3.77 million watches, 3.70 million movements, and 1.98 million
cases. Production would have been higher but for industrial action during wage negotiations
which slowed output in Hosur.
Titan’s manufacturing group had initiated a programme to place a whole new emphasis on
world class manufacturing. PQCD (Productivity, Quality, Cost Control and Delivery on
Time) had in fact become the new buzzword throughout the company. There was also a
whole new focus on “customer satisfaction” as understood in its broadest sense. Titan would,
in the current financial year, make an application for the JRD Quality Value Award. This
y
award had been instituted by the Tata group in memory of the late Mr. J.R.D. Tata and his
never-ending quest for perfection and quality, and recognised companies within the Tata
nl
group which excelled in quality management.
The CIF value of imported materials and components consumed during the year was
O
Rs 40.29 crore as against Rs 21.55 crore in the previous year, largely because of the increase
in volumes and imports of raw materials and components needed for the eurowatch
n
collection created for the European market.
io
Supply Chain Management: Material Flow
at
The supply chain management in Titan was complex. At any time, 800 varieties of watches
were made available to the retail outlets. These could be classified into seven regular ranges
ul
and four special ranges (either the export lines sold in India or designer specials). The
number of varieties that had been offered and for which Titan would be technically ready to
accept an order would be 1700.
irc
These watches were sold through a variety of channels. The most important were
redistribution stockists, institutional sales, franchisee operated shops, traditional outlets, and
C
Titan shops. The watches were distributed to these outlets through CFAs who were spread
regionally: 12 for the south, 10 for the west, 10 for the north, and 6 for the east. The watches
moved to CFAs from three assembly plants, located in Hosur, Dehradun, and Ooty.
d
Transportation from CFAs to retail outlets was essentially through road, often personally
ite
carried by a representative of the retail outlet. Transportation from the assembly plants to
CFAs was by air, wherever possible, or else by road.
A watch assembly consisted of six components, five of which formed the appearance parts
im
and the sixth was the movement (not visible). The appearance parts were cases, dials, hands,
straps, and the crown. The overall variety was a function of variety in each of these parts.
Currently, Titan operated with a variety range of 300 for cases, 1200 for dials, 800 for hands,
rL
650 for straps, very few for crowns, and 32 for movements.
Titan manufactured and assembled the movements (at Hosur) and manufactured about
50 per cent of the cases (also at Hosur). All the remaining parts were sourced from outside
Fo
and made into matched sets at Hosur, before being assigned to Hosur, Dehradun, or Ooty
assembly plants. Transportation to the Dehradun plant from Hosur was by rail through a
third party who organised a courier to carry the parts as a parcel. Transportation to Ooty was
by road.
The matched sets were put together at the finished products store using assembled
movements and the boughtout items which were intended from the raw material store. The
assembled watch heads (without the strap) in Hosur were sent to a strapping section for
fixing the straps, then packed and dispatched to CFAs through the despatch section.
In Hosur, a large assembly plant assembled both watches assigned to Hosur as well as all
the movements required for assembly across the three assembly units. A separate movement
manufacturing area made the over 100 components required for each movement. There was
y
a lot of standardisation of components across the 32 movement varieties. For example, there
nl
were only five varieties of the backplate, one of the most important components in a
movement.
O
The cases were made at the case plant in Hosur. For movements, three items were brought
in from outside. The batteries and the quartz were imported from Japan while the electronic
circuit board was manufactured by Titan in Goa and sent to Hosur.
n
All the watch related manufacturing units (except the new jewellery and bracelet units) in
io
Hosur were located in one campus, though the layout across various buildings caused a lot of
cross and duplicate moves of various components.
at
Technically sophisticated quality control procedures were instituted at every stage of
procurement and manufacture, including the concept of pre-despatch inspection by Titan at
its vendor locations. ul
Supply Chain Management: Information Flow
irc
The information flow was driven by an annual forecast for case manufacture made in
C
October/November for the following financial year. This was required for capacity planning
as well as planning for sourcing with an estimated six month lead time. The large variety in
cases also added to the complexity. A variant level forecast for cases was made quarterly, but
d
three months ahead of the quarter, since procurement/manufacturing lead times were three
months. It appeared that the case planning was the most important thing since all other
ite
appearance parts were sourced from outside with relatively shorter lead time. While
movements were also manufactured in-house, like cases, the variety was not as large.
im
Once the quarterly forecast was accepted, a material procurement request was sent by the
purchase department for the all the required appearance parts.
Based on actual availability of matched sets, a launch plan was prepared for assembly
rL
indicating the planned monthly watch production.
Company Performance
Fo
Exhibits 1 and 2 give the balance sheet and profit and loss account of the company for the
past four years. Financial information on inventories which play a key role in supply chain
management is given in Exhibit 3. Details of the operating and other expenses, which
highlight the inputs consumed, are given in Exhibit 4.
Sales information is provided in Exhibits 5, 6, 7 and 8. While sales data for the range,
region, and channel are secondary sales, CFA-wise sales are primary. As one measure of
supply chain efficiency, the inter-CFA stock transfer data for the south zone (which is
representative of the national scenario) is given in Exhibit 9.
To understand the requirements of the supply chain, ABC categorisation as well as sales
and forecast data are provided, the latter two for one case. Exhibit 10 gives the cluster-wise
(range-wise), case-wise, and dial-wise ABC categorisation for a sample of watches for 1995-
y
96 and 1996-97. Exhibits 11a and 11b give dial-wise month-wise primary sales for 19
nl
months until October 1996 for Case 141 (a representative sample). Exhibit 16 provides
information on the sales forecast and subsequent revisions for various dials in Case 141.
Month-wise production plan (launch plan) for these same dials is given in Exhibit 17.
O
An analysis of inventory vs. consumption is provided in Exhibit 12, while Exhibit 13
examines the total vs. obsolete inventory break up. The sourcing plan for cases, dials, and
n
hands is given in Exhibit 14. The purchase department’s objectives and procedures are given
in Exhibit 15. Exhibit 18 provides a sample production scheduling for a few select
io
components of a movement.
at
Supply Chain Management: Perceived Problem Areas
ul
One of the important concern areas was high levels of inventory, including obsolete
inventory. Various reasons were attributed to this situations: repeated forecast modification,
irc
long lead times, long time buckets, excess buffer planning (A category items - 2 months, B
category - 1-1/2 months, and C category - 1 month), production driven by capacity
utilisation, and lack of coordination resulting in mismatched sets. Even at the strapping
C
section, watch heads waited for straps.

Another concern was loss of sales owing to stockout. There was a debate as to whether the
focus should be on the top 150 variants or on a wider range of 800 variants essential for sales.
d
With liberalisation, one of the options was cutting back on inhouse manufacture and
ite
focusing purely on design, sourcing, assembly, distribution, retailing, and of course

marketing.
im
While vendor development and management currently had a focus on quality, delivery
flexibility was an issue yet to be addressed in a significant manner.
Production batch sizes were also an area of concern, especially since setup times, were
rL
perceived to be significantly high compared to the high rates of production. This caused
problems of large inventories and matching components.
Bhat was wondering what may be the appropriate steps he should take in terms of
Fo
ensuring a better marketing manufacturing coordination, both at the planning level and at
the damage control level, to minimise inventories and have a leaner and more responsive
supply chain support. Any step would of course require a proper justification to take the
entire organisation along.
APPENDIX
World Watch Production

World watch production 1050 million
Toy watches, clocks, pen and stick watches 200 million
y
Wrist watch production 850 million
nl
One watch for every 7 persons
O
Wrist Watch Production (By Type)
(million)
n
1989 1994
Quartz Analog 341 500
io
Quartz Digital 227 250
Mechanical 114 100
at
Total 682 850
*Share of analog and digital watches is growing
ul
*Share of mechanical watches is declining
irc
Indian Watch Market
(in lakh)
1991—92 1992—93 1993—94 1994—95

C
All Watches 180.0 184.0 193.0 203.0

Mechanical 75.0 50.0 35.0 22.0
d
QAW 84.0 107.0 128.0 148.0

Digital 21.0 27.0 30.0 33.0
ite
Brands of QAW
Titan 23.1 23.6 22.2 25.4
Timex 1.8 8.0 12.0
im
HMT Quartz 14.5 13.0 7.3 6.0

Allwyn 6.0 2.5 1.0 1.0
IMFQ & Others 40.4 66.1 89.5 103.6
rL
Total 84.0 107.0 128.0 148.0
India is only one of the dozen countries in the world manufacturing all the components
Fo
for a watch (mechanical and quartz). The others are Japan, Hongkong, Switzerland, France,
USA, Germany, S. Korea, and China.
Till 1985
Till the 1950s, watches used to be imported. The first company to produce watches in India
was HMT, a giant public sector corporation, which established a watch assembly unit in
collaboration with Citizen of Japan in 1961.
HMT produced 40 lakh watches per annum. The next biggest player was Allwyn watches,
an Andhra Pradesh State Government undertaking, which had an annual production of 4 lakh.
There were also other marginal players in India like Indo-Swiss, Indo French, and Jayna
which mainly produced mechanical watches. HMT chose to concentrate on mechanical
watches even though the world moved to quartz analog watches in 1966. HMT entered the
y
quartz segment only in 1981.
nl
Allwyn watches, though innovative in design and fairly active in the automatic watch
segment, did not establish long-term identity.
O
1986-87
Titan entered the market in 1986-87. During the period between 1986-87 to 1992-93, Titan
n
acted as catalyst transforming the market from mechanical to quartz. Titan produced only
io
quartz watches. Titan thus penetrated the market by bringing about brand/type switch from
mechanical to quartz and imported to Indian.
at
1988-89
ul
Allwyn introduced low price polyamide range branded “Trendy”, inspired by the “Swatch”
success.
irc
1989-90
Titan’s response was to create a new brand “Aqura” projecting a casual and refreshing side of
C
an individual’s personality.
1990-91
d
In the watch market:

ite
There were a number of competitors spanning several price brands and product types.
Titan wanted to create and cater to high priced segment.
im
HMT started an extensive promotion of quartz watches.
1991-92
rL
There was a steep increase in import costs on account of the devaluation of the rupee. Titan
revised its price sharply by raising by 20 per cent and vacated price brands below Rs 600.
Fo
The IMFQ segment of the market gained a considerable cost advantage over the organised
sector.
1992-93
Timex brand name became available to Titan as a result of its entering into a joint venture
with the well known Timex Corp. of USA.
Allwyn incurred huge losses in 1989 and 1991 and was offered for sale.
1993-94
The shift of the Indian market from mechanical to quartz.
The hold on the watch market shifted from public sector companies like HMT and
Allwyn to private sector - Titan and Timex.
Development and growth of the grey watch market including IMFQs and smuggled
y
watches.
nl
Development of new retailing network initiated by Titan using exclusive showrooms to
sell its products.
O
Factsheet on Titan
n
Titan is acronym derived from - TATA Industries and Tamil Nadu.
io
Titan means descendent of the Greek god Uranus one who has extraordinary strength,
intellect, and power.
at
Titan Industries Limited is a joint sector company, promoted by Tata Industries Limited
and TIDCO to manufacture quartz analog watches and jewellery.
ul
Tata’s interests in watches dates back with the acquisition of minor stake in Hegde and
Golay and an aborted collaboration with Fairchild. TIDCO was also interested in setting up
irc
a watch manufacturing unit at the same time. Thus, together with France Ebauches, a
European movement giant, a tripartite agreement was signed and Titan Industries Limited
was born in 1984, and started selling watches by 1987.
C
Present Locations
Offices
d
Corporate Office Bangalore

ite
Regional Office Bangalore, New Delhi, Calcutta, Bombay

Branch Office Madras
International Marketing Office London, Paris, Dubai
im
Factories and Units

Main Plant Hosur
rL
Assembly Unit Dehradun

ECB Goa
ATP Hosur
Fo
Stepper Motors Hosur

Jewellery & Jewellery Watches Hosur
Euro Watches Hosur
Titan Vision
To be a leader in the watch market and to be admired and respected as innovative company.
To be well known in the International watch industry through its products.
To be a responsible corporate citizen.
Titan is perceived to be a professionally managed company with one of the finest watch
plants in the world and with a high degree of innovativeness in marketing. In fact Titan is
perceived to be in a fashion business rather than the watch business as a result of its beautiful
products, promotions, and showrooms.
Today Titan is a Rs 275 crore company, the largest player in the Indian watch industry.
y
Titan foresees the company play a major role in the international market. It is thus adding
nl
new brand and new dimensions to the personal accessories business to fulfil its aim and reach
its goal.
O
1. Major Events
Government of India approval for watch production August 1985
Government of India approval for Component manufacturing May 1986
n
Commencement of watch assembly March 1987
io
Government of India approval for Case manufacturing July 1987
Commencement of component manufacturing October 1987
at
Commencement of Case manufacturing November 1989
Government of India approval for Dehradun Unit
ul March 1990
Commencement of Dehradun Watch assembly March 1990
Commencement of Step motor manufacturing August 1992
irc
Commencement of Euro Case manufacturing May 1994
Commencement of Euro Watch assembly March 1995
C
Commencement of Ooty Watch assembly (Tansi Unit) October 1995

2. Capacity Details
d
Annual Capacity Watches Hosur 2.1

ite
Dehradun 1.8
Ooty 0.5
4.4
im
Cases Metallic 2.2

Euro Cases 0.15
rL
Euro Bracelets 0.5

Movement Components (in sets) 4.0
3. Investment (approx) as on 31-03-1996 Rs 2000 million US$ 55.5 million
Fo
4. Turnover (approx) Value in millions

Actuals (1995-96) Cumulative till Target (1996-97)
Rs 3071 Rs 15111 Rs 4154
US$ 88 US$ 432 US$ 119
5. Production Statistics
Description Actuals Cum. Till Target
(1995-96) 31.7.1996 (1996–97)
Regular Watches 3.6 20.05 4.175

Euro Watches 0.18 0.28 0.425
y
Movements 3.7 21.5 4.2
nl
Cases (Metallic) 1.9 10.9 2.2
Cases (Euro) 0.08 0.13 0.145
O
Bracelet 0.019 0.033 0.45
6. Employee Statistics
n
Men 1864
Women 397
io
Supervisory and Managerial Staff 471
Total Workforce 2261
at
Average age of Employees 24.7 years
Tamil Nadu Nativity ul 2066
Handicapped Employees
irc
Physically Handicapped 31
Hearing Impaired 100
Sight Impaired 6
C
Total 137
7. Plot Area Details sq.mts
d
Total Area 56,672

ite
Built up Area 32,385

Air-conditioned Area 16,966
im
8. Number of Variants
Watch Models more than 2500
Basic Movements 8
rL
Movement Versions 32
9. Sales
Fo
Description Actuals Cum. Till Target

(1995-96) 31.7.1996 (1996–97)
9.1 Watches Domestic 3.40 18.44 3.62

9.2 Watches Export 0.43 1.089 0.91
9.3 Movements Export 1.106
10. Domestic Sales Network (approx) No’s

Clearing and Forwarding Agents 38
Dealers 4738
Towns Covered 1100
Authorised Service Centres 226
y
Redistribution Stockists 55
nl
Company’s Service Centres 36
O
11. Export Sales Network (approx) No’s
Countries Covered 19
Distributors 07
n
Service Centre’s 19
io
Dealers 1904
Towns Covered 89 +
at
Government Policy
ul
(excerpted from the Annual Reports of 1994-95 and 1995-96)
irc
1995-96
Smuggling still remains a menace in our country. Government policy makers continue to
C
give insufficient attention to the repeated representations made by Indian manufacturers to

make adequate reductions in excise and customs duties and to destroy seized smuggled goods
as part of a strategy to combat smuggling.
d
Titan, and for that matter, the entire Indian watch industry, would like to see a more
ite
enlightened government policy where, in addition to customs duties on components being

reduced to 20 per cent and on raw materials to 10 per cent, excise duties are reduced to a
nominal 1 per cent and sales taxes are pegged uniformly at 3 per cent.
im
1994-95
In spite of repeated representations by Indian manufacturers, including Titan, sufficient
rL
attention has not been given by Government policy makers to the menace of smuggling.
While customs duties have come down in the last two years, the current level of duties still
remains high and, combined with the damaging effect of a 10 per cent excise duty as also
Fo
high rates of sales taxes, octroi and turnover taxes, makes it extremely difficult for the
legitimate Indian manufacturers to compete with smugglers, control over whose activities is
largely ineffective. Most affected are watches priced below Rs 750.
Titan is of the considered view that, in addition to further reductions in customs duties on
components to 20 per cent and on raw materials to 10 per cent or less, it is also essential for
the Central Government to lower the excise burden on watches to a nominal 1 per cent and
for state governments to peg sales taxes at 3 per cent. Equally importantly, vigilance,
detection and punitive action against those who indulge in smuggling and defrauding the
Government of India must be stepped up. Suggestions made to various ministries of the
Government of India have met with a muted response much to the relief of the global
smuggling fraternity. At the same time, government needs to review its present restrictions on
the import of complete watches. We recommend that, in the first stage, watches with a CIF
y
value of Swiss Francs 300 and above be allowed to be imported with special import licenses
and floor prices gradually lowered thereafter. We also recommend that the import duty in
nl
such watches be no more than 40 per cent.
O
Exhibit 1: Balance Sheet as on March 31
n
(Rs lakh)
io
SOURCES OF FUNDS 1993 1994 1995 1996
Shareholders Funds
at
Share capital 4227.63 4227.63 4977.63 5227.63
Reserves and surplus 7781.71 8633.68 9874.49 11145.94
Loan Funds
Secured loans
ul
10181.65 11878.53 15595.24 19945.13
irc
Unsecured loans 1262.33 3236.91 3567.95 10388.70
APPLICATIONS OF FUNDS
C
Fixed Assets
Gross block, at cost 14793.07 15861.04 24357.04 28513.32
Less: Depreciation 2751.17 3708.39 5010.99 6555.96
d
Net block 12041.90 12152.65 19346.05 21957.36

Advances on capital account and capital work
ite
in progress, at cost 1094.61 4511.52 1245.03 1009.57

Investments 6.00 1492.29 2181.85 2831.50
Current assets, loans and advances
im
Inventories 8639.40 10184.60 11207.05 15253.69

Sundry debtors 1406.46 1432.09 3358.28 7784.63
Cash and bank balances 850.35 1265.22 962.74 1385.10
rL
Loans and advances 3018.94 1861.29 1893.52 4389.34

Less:
Current liabilities and provisions
Fo
Current liabilities 2885.30 3819.51 4837.75 6387.62

Provisions 719.04 1103.40 1341.46 1516.17
Net current assets 10310.81 9820.29 11242.38 20908.97
Total 23453.32 27976.75 34015.31 46707.40
Source: Company Annual Reports, 1993-94, 1994-95, and 1995-96.

Exhibit 2: Profit and Loss Account

(Rs lakh)
1993 1994 1995 1996
INCOME
y
Sales 19121.45 22622.69 28034.24 35071.95
nl
Other Income 159.85 258.09 359.31 294.50
TOTA L 19281.30 22880.78 28393.55 35366.45
O
EXPENDITURE
Operating and other expenses 13599.00 15984.14 20028.48 24825.05
Excise duty 2026.14 2393.50 2364.62 2793.64
n
Depreciation 722.76 978.33 1311.35 1568.40
Interest 1846.19 1615.93 2179.77 3422.16
io
T O TA L 18194.09 20971.90 25884.22 32609.25
PROFIT FOR THE YEAR 1087.21 1908.88 2509.33 2757.20
at
TAXES
PROFIT AFTER TAXES 1087.21 1908.88 2509.33 2757.20
Povisions for taxes in respect to earlier year
Profit brought forward
Amount available for appropriations
ul
4.82
123.92
1206.31

78.40
1987.28

675.37
3184.70

1627.18
4584.38
irc
Appropriations 1127.91 1311.91 1557.52 1813.75
Transfer to debenture redemption reserve 57.00 55.00 29.00 28.00
Transfer to investment allowance reserve account 304.05
C
Proposed dividend (subject to deduction of tax) 688.86 1056.91 1268.52 1395.12

Transfer to general reserve 78.00 200.00 260.00 300.00
Balance carried to balance sheet 78.40 675.37 1627.18 2770.63
d

ite
im
Exhibit 3: Inventories as on March 31

(Rs lakh)
rL
1993 1994 1995 1996
Inventories
Consumable stores 438.33 482.01 435.39 405.45
Fo
Loose tools 64.88 96.71 108.39 165.71

Stock-in-trade 8136.19 9605.88 10663.27 14682.53
Raw materials and bought components 2167.80 2401.59 3051.37 4836.98
Work in progress 3710.12 3907.37 3857.94 3937.41
Finished goods 3296.92 2258.27 3753.96 5908.14
TOTAL 8639.40 10184.60 11207.05 15253.69
Exhibit 4: Operating and Other Expenses

(Rs Lakhs)
Operating and Other Expenses 1992–93 1993–94 1994–95 1995–96
Raw materials and components consumed 7574.62 8960.65 11580.74 14900.25
y
Stores and spare parts consumed 1445.95 1563.73 703.29 1010.34
nl
Purchase of finished goods 47.69 145.84 198.48 361.93
Payments to and provisions to employees 1283.71 1788.64 2262.06 3219.83
O
Other expenses 4003.76 4940.29 5907.29 7709.40
Auditors remuneration 4.42 6.88 8.40 10.24
Directors fees 0.26 0.17 0.30 0.31
n
Decrease/(Increase) in work in progress and finished goods 138387.66 16170.30 20252.95 24978.65
Less: Expenses capitalized 238.66 186.16 224.47 153.6
io
Source: Company Annual Reports 1993-94, 1994-95, and 1995-96.
at
ul
Exhibit 5: Rangewise Secondary Sales
irc
(000)
Cluster 1993–94 1994–95 1995–96
C
Type: Gents
Classique 197.18 209.73 6.00
Exacta 215.11 279.22 23.00
d
Royale 181.29 179.48 1.00

Regalia 27.40 16.34 68.00
ite
Others 27.09 26.79 1.00

Total: Gents 648.07 711.56 9.00
Type: Ladies
im
Classique 120.88 167.32 28.00

Exacta 80.26 101.45 21.00
Royale 92.14 85.37 8.00
rL
Regalia 4.92 2.21 123.00

Others 7.16 9.25 23.00
Total: Ladies 305.36 365.6 16.00
Fo
Type: Others
Pair 1.55 0.48 223.00
Grand Total 954.98 1077.64 11.00
Exhibit 6: Regionwise Secondary Sales

(000)
Region Channel Ytd Aug 96 Ytd Aug 95 % Growth Target 96-97 Annualized
Sales/000
y
SOUTH Showrooms 113.40 115.87 -2.13 313.17
nl
Titan Shops 42.00 43.11 -2.57 112.03
Multibrand 202.65 225.38 -10.09 554.06
TOTAL 358.04 384.36 -6.85 979.26 14.48
O
WEST Showrooms 64.89 64.38 0.79 210.91
Titan Shops 29.52 34.97 -15.58 100.27
Multibrand 122.80 130.62 -5.99 364.00
n
TOTAL 217.21 229.98 -5.55 675.18 8.70
NORTH Showrooms 28.47 32.37 -12.05 101.64
io
Titan Shops 30.52 35.14 -13.15 108.62
Multibrand 147.23 183.46 -19.75 516.99
at
TOTAL 206.22 250.97 -17.83 727.55 8.65
EAST Showrooms 26.44 28.64 -7.68 75.41
Titan Shops 11.71 ul 12.90 -9.22 34.13
Multibrand 135.61 170.76 -20.58 394.13
TOTAL 173.76 212.3 -18.15 503.67 10.80
irc
Total Showrooms 233.20 241.26 -3.34 701.13
Titan Shops 113.75 126.12 -9.81 355.05
Multibrand 608.29 710.22 -14.35 1829.18
Grand Total 955.23 1077.61 -11.36 2885.66
C
d
Exhibit 7: Channel’s Contribution to Total Sales Volume (Ytd-Aug 96)

ite
Channel Sales Volume % to Total Avg UCP Relative to Value (%) to

Total (%) Total
im
RDS 344463 36.3 1099 92.1 33.4

Institutional 159914 16.8 1055 88.4 14.9
rL
Franchisee 156497 16.5 1469 123.1 20.3

Traditional 95184 10.0 1181 99.0 9.9
Titan Shop 72498 7.6 1328 111.3 8.5
CSD 56623 6.0 976 81.8 4.9
Fo
Non-Traditional 32499 3.4 1307 109.6 3.7

Showrooms 17222 1.8 1605 134.5 2.4
Others 14732 1.6 1490 124.9 2.0
Grand Total 949632 100.0 1193 100.0 100.0
UCP Uniform Consumer Price in Rs

Exhibit 8: CFA Wise Primary Sales - 1996-97

(numbers)
CFA Apr May Jun Jul Aug Sep Total
y
Bangalore 6851 25255 15662 17974 18042 28049 111833
Madras 4216 11563 12828 12603 14159 20297 75666
nl
Hyderabad 5568 8028 7156 6245 10144 19723 56864
Coimbatore 1325 6572 6033 6328 7424 14910 42592
O
Madurai 476 4019 5344 5002 9068 12439 36348
Vijaywada 3117 7760 2559 3950 6685 9439 33510
Cochin 1186 4264 4095 4100 10545 7326 31516
n
Vizag 2059 4204 1857 2002 3550 6027 19699
io
Hubli 1649 3027 2063 1905 4113 5905 18662
Trichy 51 1601 2614 3015 4628 6646 18555
Calicut 446 1511 2320 1526 3765 3490 13058
at
Goa 513 991 1261 940 1852 3628 9185
Staff 300 472 511 250 498 0 2031
Region South
Bombay
27757
4947
79267
10712
ul
64303
14011
65840
17613
94473
19205
137879 469519
66488
irc
Ahmedabad 1001 2812 2862 5300 6620 18595
Pune 591 1616 3624 4865 5564 16260
Thane 118 1524 2244 3522 5779 13187
C
Aurangabad 241 849 1861 2501 2548 8000

Rajkot 251 1048 1300 2323 3007 7929
Bhopal 472 1675 550 2003 2800 7500
d
Nagpur 604 1578 1023 1553 1872 6630

Indore 286 950 1263 1355 1416 5270
ite
Raipur 288 806 704 1216 1436 4450

Region West 8799 23570 29442 42251 50247 154309
Delhi 9412 13245 16908 15256 19421 74242
im
Varanasi 1589 6906 1000 3001 3014 15510

Lucknow 600 3843 2164 3234 4690 14531
Ludhiana 147 1487 704 5125 6848 14311
rL
Meerut 472 1051 1969 2024 5204 10720

Jaipur 778 2042 1554 2114 3089 9577
Ambala 78 1790 1720 745 3124 7457
Fo
Jodhpur 464 1000 1772 2004 2200 7440

Chandigarh 122 265 805 1496 1374 4062
Jammu 182 561 701 940 762 3146
Region North 13844 32190 29297 35939 49726 160996
Calcutta 4313 9593 9887 16125 14830 54748
Patna 9300 17758 6367 6243 5084 44752
Jamshedpur 1697 5655 3601 3213 3048 17214
(Continued)
(Continued)
Bhubaneshwar 1296 4066 3943 3941 3291 16537

Guwahati 1319 1681 2411 3438 3521 12370
Siliguri 241 2504 2154 2566 2341 9806
y
Region East 18166 41257 28363 35526 32115 155427
nl
Region Total 68566 176284 151405 179556 226561 940251
CSD 9938 8100 5994 14168 17427 55627
Godrej 0 87960 0 0 0 87960
O
Domestic 78504 272344 157399 193724 243988 945959
Export 10576 35827 34622 59797 43130 183952
Grand Total 89080 308171 192021 253521 287118 1129911
n
io
at
Exhibit 9: Stock Transfer Issued Out Monthwise: Region South - 1996-97
ul (numbers)

irc
Bangalore 1273 4027 3165 2961 4099 3883 19408
Madras 929 2383 4025 3704 3922 2339 17302
Coimbatore 12 3144 2231 598 2022 639 8646
C
Hyderabad 0 959 349 1825 1310 327 4770

Cochin 0 270 74 40 439 1582 2405
Trichy 7 33 180 496 694 457 1867
d
Vijaywada 11 106 11 987 350 158 1623

ite
Hubli 6 33 239 660 117 95 1150

Madurai 0 334 30 39 549 130 1082
Calicut 0 166 60 111 120 373 830
im
Goa 0 0 330 235 195 23 783

Vizag 0 114 15 148 29 0 306
Total 2238 11569 10709 11804 13846 10006 60172
rL
Fo
Exhibit 10: Dial Categories for the Year 1995-96 and 1996-97
Category
Type Cluster Case Dial 1995-96 1996-97
y
Ladies EXACTA 103 D39 B B
EXACTA 103 D40 A C
nl
EXACTA 103 F25 A A
EXACTA 103 F26 B A
O
EXACTA 103 Z147 C
EXACTA 107 A27 C A
EXACTA 107 A28 C B
n
EXACTA 107 A29 C B
io
CLASSIQUE 109 A33 B B
CLASSIQUE 109 C62 B C
at
CLASSIQUE 109 F28 A C
CLASSIQUE 109 F29 B B
CLASSIQUE 109 ul N96 B B
CLASSIQUE 109 N97 B B
CLASSIQUE 111 A37 B C
irc
CLASSIQUE 111 D44 B B
CLASSIQUE 111 H58 A B
CLASSIQUE 111 L44 C C
C
CLASSIQUE 111 Z135 C

d
CLASSIQUE 113 C60 A B

ite
CLASSIQUE 113 K80 B A

CLASSIQUE 113 M90 B A
CLASSIQUE 113 N99 B C
im
CLASSIQUE 113 T218 C

rL

CLASSIQUE 117 A52 C B
Fo
CLASSIQUE 122 A71 A A

CLASSIQUE 122 A71R C
CLASSIQUE 122 A72 B A
CLASSIQUE 122 D48 B A
CLASSIQUE 122 D48R C
Gents EXACTA 122 L41 B B
EXACTA 138 B24 B B
(Continued)
(Continued)
Category
EXACTA 138 B25 B B

EXACTA 138 D11 B B
y
EXACTA 139 B26 A A
EXACTA 139 B27 A A
nl
EXACTA 139 B88 A A
EXACTA 139 D12 B A
O
EXACTA 139 L81 C C
EXACTA 140 B29 A A
EXACTA 140 C65 A B
n
ROYALE 140 E26 A A
io
ROYALE 140 T510 C
ROYALE 140 T511 C
at
CLASSIQUE 141 B30 A C
CLASSIQUE 141 ul C43 A A
CLASSIQUE 141 K56 B A
CLASSIQUE 141 K57 B B
irc
CLASSIQUE 141 K58 B C
CLASSIQUE 141 R50 C
CLASSIQUE 141 R51 C
C
CLASSIQUE 142 B33 A A

CLASSIQUE 142 F81 A A
d
CLASSIQUE 142 K81 A C

ite
CLASSIQUE 142 K82 B B

CLASSIQUE 142 R52 C
im

rL

CLASSIQUE 142 Z175 C C
CLASSIQUE 143 B36 A B
Fo
CLASSIQUE 143 B37 B A

CLASSIQUE 143 D19 A B
CLASSIQUE 143 G26 A A
CLASSIQUE 149 B49 A A
CLASSIQUE 149 C61A C
(Continued)
(Continued)
Category
CLASSIQUE 149 D21 A A

CLASSIQUE 149 G29A C
y
CLASSIQUE 149 K63 A A
nl
CLASSIQUE 149 Z160 C C
O
CLASSIQUE 149 Z180 B
CLASSIQUE 149YL Z216 A
n
CLASSIQUE 153 B92 B B
CLASSIQUE 153 K90 C C
io
CLASSIQUE 153 P19 C C
CLASSIQUE 153 P20 C B
at
CLASSIQUE 155 C47 B B
CLASSIQUE
CLASSIQUE
155
155
ul E80
L11
A
A
A
B
ROYALE 161 B73 A A
irc
ROYALE 161 C17 A A
ROYALE 161 D35 A A
ROYALE 161 P42 B A
C
ROYALE 161 P43 C A

ROYALE 162 B74 A A
d
ROYALE 162 C49 A A

ROYALE 162 J20 A A
ite
ROYALE 162 J22 A A

ROYALE 162 R57 C
CLASSIQUE 166 C12 B B
im

CLASSIQUE 166 K67 A B
Ladies ROYALE 172 J85 C B
rL
ROYALE 172 J86 C B

ROYALE 172 J87 C B
ROYALE 176 C52 B B
Fo
ROYALE 176 C53 A A

ROYALE 176 C54 B B
ROYALE 176 J53 A A
CLASSIQUE 177 E35 A A
CLASSIQUE 177 F33 A B
CLASSIQUE 177 P21 C B
CLASSIQUE 177 P22 B A
(Continued)
(Continued)
Category
Type Cluster Case Dial 1995–96 1996–97

Gents CLASSIQUE 179 D23 A A
y
CLASSIQUE 179 E86 A B
nl
CLASSIQUE 179 P23 B A
CLASSIQUE 179 P24 B B
O
EXACTA 180 D54 A A
EXACTA 180 D57 A A
EXACTA 180 G22 A A
n
EXACTA 180 G23 A A
EXACTA 180 H83 A A
io
EXACTA 180 H84 A A
EXACTA 180 H85 A A
at
EXACTA 180 L77 A A
EXACTA 180 L78 A A
EXACTA 180 ul L79 A B
EXACTA 180 Z124 C
EXACTA 180 Z141 C
irc
EXACTA 180 Z164 C
C
d
ite
im
rL
Fo
Exhibit 11a: Month Wise Primary Sales - Case - 141 (Gents Classique)
(numbers)
1995–96 Apr May Jun Jul Aug Sep Total
y
B30 53 1231 467 297 490 1181 3719
B31 121 385 37 174 590 802 2109
nl
C43 197 758 610 428 580 1574 4147
D17 49 97 42 54 51 114 407
O
D67A 0 0 0 0 0 0 0
D68A 0 0 0 0 0 0 0
D69A 0 0 0 0 0 0 0
n
D70A 0 0 0 0 0 0 0
E68 107 637 225 176 426 675 2246
io
E69 0 0 1 0 0 0 1
E70 0 3 3 0 0 0 6
at
F94 0 3 1 0 0 0 4
F95 76 120 66 93 60 114 529
G24
G25
11
10
48
63
ul12
44
11
16
27
20
40
20
149
173
irc
K56 24 415 764 241 356 1006 2806
K57 179 620 103 95 520 800 2317
K58 167 896 280 234 903 940 3420
C
K61 12 80 28 16 -2 27 161
K75 14 35 16 16 24 21 126
P12 4 13 24 -2 -19 11 31
d
P13 9 23 11 4 -12 38 73
P14 1 9 1 1 6 1 19
ite
Q33 0 0 0 0 0 0 0
R50 0 0 0 0 0 0 0
R51 0 0 0 0 0 0 0
im
Total 1034 5436 2735 1854 4020 7364 22443

Oct Nov Dec Jan Feb Mar Total
B30 395 685 356 530 752 1243 3961
rL
B31 167 92 637 354 316 687 2253

C43 683 740 465 661 743 1797 5089
D17 41 49 24 5 23 42 184
Fo
D67A 0 765 351 114 90 93 1413

D68A 0 148 169 29 209 296 851
D69A 0 824 298 93 64 95 1374
D70A 0 938 109 30 14 24 1115
E68 297 224 131 243 127 237 1259
E69 0 3 2 1 0 3 9
E70 0 0 0 0 0 0 0
(Continued)
(Continued)
1995–96 Apr May Jun Jul Aug Sep Total
F94 0 0 1 1 1 0 1
F95 92 54 19 75 51 14 305
G24 29 7 3 14 9 7 69
y
G25 9 13 4 19 6 11 54
nl
K56 238 269 138 288 362 388 1683
K57 530 641 313 440 407 1283 3614
K58 580 588 535 822 670 444 3639
O
K61 45 24 6 9 9 14 107
K75 8 14 10 6 9 6 53
P12 40 8 10 23 5 4 90
n
P13 31 11 9 28 6 25 110
io
P14 4 4 1 2 3 1 13
Q33 0 0 0 0 0 100 100
at
R50 0 0 0 0 0 465 465
R51 0 0 0 0 0 139 139
Total 3189 6101 3579
ul 3787 3874 7418 27948
irc
C
d
ite
im
rL
Fo
Exhibit 11b: Monthwise Primary Sales - Case - 141 (Gents Classique)
1996–97 Apr May Jun Jul Aug Sep Total Oct
B30 98 278 215 275 157 473 1496 205
y
B31 46 193 84 52 35 21 431 12
C43 108 323 209 233 665 2406 3944 784
nl
D17 3 2 0 5 7 78 95 37
D67A 15 25 32 35 41 264 412 58
O
D68A 24 25 11 22 37 171 290 41
D69A 16 16 20 20 10 111 193 22
D70A 9 3 5 3 4 13 37 3
n
E68 32 131 56 75 55 231 580 44
io
E69 0
E70 0 0 0 1 1 0 2 0
F94 0
at
F95 2 3 12 6 8 23 54 33
G24 2 3 0 1 1 4 11 0
G25
K56
5
126
18
306
8
209
ul 5
221
9
515
5
1597
50
2974
3
677
irc
K57 174 348 361 225 378 1592 3078 271
K58 67 437 281 189 71 57 1102 487
K61 4 6 2 1 2 3 16 0
C
K75 1 3 1 1 3 3 12 0
P12 3 14 3 3 6 0 23 2
P13 13 11 4 4 15 89 136 57
d
P14 1 1 2 1 0 2 3 0
Q33 10 0 0 0 0 0 10 0
ite
R50 229 385 208 248 325 133 1528 357

R51 187 282 138 56 81 131 875 42
im
rL
Fo
Exhibit 12: Inventory Vs Consumption

(Quantity: 000; Value: Rs. lakh)
Head Cases Dials M Straps L Straps M Glass
Stock as on 1-9-96 Qty 219 1125 441 401 230
y
Value 360 540 246 196 30
nl
Non Moving & Obsolete Qty 9 281 70 70 35
Value 15 135 39 34 7
O
Net Usable Qty 210 844 371 331 195
Value 345 405 207 162 23
Consumption in Sep96 Qty 77 190 123 77 155
n
% 37 23 33 23 79
Consumption in Oct96 Qty 55 120 102 43 15
io
% 26 14 27 13 8
Consumption in Nov96 Qty 13 45 51 28 5
at
% 6 5 14 8 3
Not usable in next 3 mth Qty 65 489 95 183 20
% ul31 58 26 55 10
irc
C
d
ite
im
rL
Fo
Exhibit 13: Total and Obsolete Inventory Break-up

(Quantity: 000; Value: Rs. lakh)
Item As on 31.03.96 As on 30.06.96 As on 31.07.96
y
Quantity Value Quantity Value Quantity Value
nl
Total Inventory Break-up - Appearance Parts
Cases 150 253 178 267 233 357
Dials 1249 522 1259 554 1322 631
O
Hands 988 141 1647 197 1562 197
Straps - M 561 455 469 325 451 322
Straps - L 296 130 508 246 431 222
n
Total 1501 1589 1729
io
Total Inventory Break-up Others
Case Component 112 95 87
at
Euro Comp 96 110 130
Preplated Parts 138 138 138
Movts+Comp ul388 308 288
Raw Mat 372 274 411
Cons+Tools 223 212 209
irc
Gold 58 43 73
Maint Spares 191 271 281
Packaging & Others 66 77 89
C
Others Total 1644 1529 1706

Grand Total 3145 3117 3435
Obsolete Inventory Break-up Appearance Parts
d
Cases 20 39 39 39
ite
Dials 347 157 157 157

Hands 347 57 57 57
Straps - M 147 82 82 58
im
Straps - L 64 21 21 21
Total 356 356 332
Obsolete Inventory Break-upOthers
rL
Case Component 15 15 15
Euro Comp 11 11 11
Preplated Parts 138 138 138
Movts+Comp 19 19 19
Fo
Raw Mat 12 12 12
Cons+Tools 48 48 45
Gold
Maint Spares
Packaging & Others 9 9 9
Others Total 252 252 249
Grand Total 608 608 581
Exhibit 14: Sourcing Plan for 1996-97
Cases Qty (000) Rs lakh
Centura Tumkur 508.5 302.37
y
Cosmos - Bidar 657.0 1155.83
Growel - Pune 299.5 561.22
nl
Hitech - Bangalore 559.0 335.94
Sree Laxmi - Tumkur 128.0 90.59
O
Saha Tumkur 26.0 15.74
TOTAL INDG 2178.0 2461.68
Dailywin - Hongkong 149.3 493.18
n
Kamyuen - Hongkong 191.5 712.16
io
K&S - Hongkong 17.0 83.45
TOTAL IMP 357.8 1288.8
at
TOTAL CASES 2535.8 3750.48
Dials
KDDL - Bangalore
KDDL - Parwanoo
ul 1818
1779
502
545
irc
KAWAGUCHIKO - Japan & Taiwan 276 137
YICHANG - Taiwan 370 266
TAIWAN SHOKO - Taiwan 203 277
C
TOTAL 4446 1727

Hands
d
MU Ltd - Hongkong (Mfg at Japan) 2383 254.89

ite
Universo - Switzerland 2088 337.00

TOTAL 4471 591.89
im
Exhibit 15: Purchase Department

rL
Objectives
✧ To provide uninterrupted supplies of materials for production requirement
Fo
✧ To keep the costs of inputs under control

✧ To help vendors in ensuring quality of material supplied by them
Activities
✧ Identification, Selection and Approval of Vendors to meet TIL’s product requirement
✧ Releasing of orders to meet MPR requirements
✧ Follow up with suppliers for supplies
✧ Monitoring supplier’s performance
y
✧ Assisting suppliers in Technical and Financial requirements
nl
✧ Wherever required coordinating with planning, stores, QC, Assembly and Finance
for receipt, inspection and payment to the suppliers
O
✧ Involving in make or buy decisions of cases
✧ Developing new variants of existing items procured
n
✧ Liaisoning with government agencies to obtain statutory licenses and approvals
io
✧ Indigenisation
Ordering Procedures
at
✧ MPR from concerned department with drawing
✧
✧
Checked for all relevant detailsul
QTN obtained/comparative QTNs/repeat order price
irc
✧ Proposal to section head/department head
✧ Approval obtained to place PO
✧ PO number booked in register
C
✧ Order released
✧ Order sent for approval/signature
d
✧ Order copies made to relevant departments

ite
✧ PO to supplier
✧ Suppliers acknowledgement & acceptance obtained
✧ Order copy and acceptance letter filled
im
Other Procedures
rL
1. Supplier Selection:
✧ Based on product range available
✧ Based on infrastructure, capacity available
Fo
✧ Based on financial capability

✧ Based Titan’s visit to suppliers’ work and impressions formed from this
✧ Based on past records of supply to other parties, if any
✧ Based on commitments received for delivery
✧ Based on quality evaluation of samples
2. Price Fixation:
✧ Products specs. identified
✧ Quotation taken
✧ Comparative quotations
y
✧ Negotiation
nl
✧ Price fixation based on lowest quote subject to quality requirements being met
3. Vendor Rating:
O
✧ Done based on department procedures
✧ Analysis of vendor rating
n
✧ Sent to supplier
io
✧ Discussed in detail with supplier
✧ Improved rating obtained based on discussions
at
ul
Exhibit 16: Sales Forecast: Case - 141 Gents Classique
irc
(numbers)
Apr-Jun 96 Apr-Jun 96 Jul-Sept 96 Jul-Sept 96 Oct-Dec 96

C
I - QTR Revised II - QTR Revised III - QTR
Dial
d
B30 0 0 800 800 1550

B31 0 0 300 300 0
ite
C43 0 0 800 800 1700

E68 0 0 0 0 0
K56 750 750 1000 1000 1500
im
K57 1000 1000 1000 1000 1500

K58 750 750 0 0 1500
rL
R50 1500 1500 1200 1200 0

R51 1500 1500 1200 1200 0
T856 0 0 0 0 2500
Fo
Exhibit 17: Production Plan: Case - 141 Gents Classique

(numbers)
Dial Oct-96 Nov-96 Dec-96

B30 100 0 700
y
B31 200 0 0
nl
C43 600 0 100
K56 0 0 200
O
K57 0 200 300
K58 1100 0 300
R50 0 0 500
n
R51 1500 0 0
T856 0 0 500
io
at
Exhibit 18: Production Scheduling of Selected Movement Components
ul (numbers)
irc
Component Description Target for Target for Oct Production Target for Nov Production
The Year Schedule on Schedule on
1996–97 a given Day a given Day
– Oct – Nov
C
Bridle Screw (FE42) 5326500 484000 15710 465000

Elec Mod Screw (FE40) 1861500 290000 21000 278000
d
Yoke Spring Screw (FE43) 1339500 140000 146000

Hand Setting Train Cover Screw 2050500
ite
Stop Lever Screw (FE44) 1866000

Minute Wheel Gaurd Screw-T90 277500
Dial Screw (FE5) 1624500 345000 331000
im
ECB Screw-T9092 304500

Yoke Screw-T9092 226500 60000 11065 50000
Dial Screw-T9092 333000
rL
Train Wheel Bridge Screw (FE4) 2094000 289000 278000

Electronic Module Screw (FE34) 5517000 864000 41920 830000 21535
Setting Lever Jumper Screw (FE3) 6300000 1033000 18930 993000 18500
Fo
Hand Setting Train Cover Foot 11430000 1706000 71880 1640000 60965
Yoke Spring Stud 6000 2269500 470000
Train Wheel Bridge Support Stud 150000
Train Wheel Bridge Foot 3394500 148000
Date Corrector Stud 12118500 1690000 38300 1140000 16120
Intermediate Date Wheel Stud 3357000 700000 39230 815000
Exhibit 19: Watch Out For More Growth
Xerxes Desai, the 58-year-old Oxford educated Philosophy student of Karl Popper, is a
man busy shuttling between nations. In an exclusive interview with Business Barons, Shobha
Ramaswamy, the Vice-Chairman and Managing Director of Titan industries reveals his
y
ambitious plans to make Titan a world brand.
nl
✧ What opportunities attracted you to enter the personal accessories industry with Titan
watches?
O
Desai: The existence of a large and growing market which was imperfectly supplied
and needed technical, managerial, marketing and financial inputs which the Tatas
were well equipped to provide.
n
✧ What strategies did Titan formulate to ward off competition? Do you perceive threat
io
from other companies, especially in the premium market?
Desai: Titan will use every weapon available in the armoury of a successful marketer
at
to confront competition, including product publicity, PR, sponsorships, sub-
branding, expanding and upgrading the company’s owned and franchised retail
ul
chains and working closely with the trade. We will have no qualms about using our
market clout and will always maintain an aggressive posture. The globalisation of
irc
Titan and making it an international brand name will also benefit Titan in India.
We believe that the least threat comes from European brands and the greatest threat
from Asian brands.
C
✧ What are Titan’s strengths and weaknesses?

Desai: Titan’s greatest strengths are its manufacturing, marketing and distribution
d
resources, the powerful brand name that has been built, the vast amount of goodwill
it enjoys with the trade, with media and with the general public, the fact that it is
ite
increasingly a global brand and, of course, the fact that it is a member of the Titan
group.
im
Titan’s major weakness derives from its success and the huge expectations which
people have of it: consequently, it is an organisation in a hurry, seeking, very rapidly,
to expand, move upmarket, enter new markets and diversify its portfolio. This
rL
places significant stress on our people, our finances and, sometimes, our bottomline.
✧ What is Titan’s marketing credo?
Desai: To provide value for money, at many price segments, to a diverse public and
Fo
be held in high regard for the products and services which we provide to our
customers – both in India and globally.
✧ What are Titan’s global plans?
Desai: Titan plans to be a significant, trusted and respected global brand in the $ 70
to $700 price segment. We will be a generalised brand, selling a fairly wide spectrum
of products with a heavy emphasis on innovation, contemporaries, styling, quality
and value for money.
✧ Titan Industries net profit margins are under pressure. Have you over invested in
Europe?
Desai: In fact, the profit before tax was marginally higher than in the corresponding
period the previous year. The profit after tax was marginally down because of the
introduction of MAT. But for MAT, Titan would not have any tax liability this year.
y
At the same time, it is current to say that profits are under pressure this year not only
because of the economic downturn resulting in a cut back in consumer spending but
nl
also because of the combined impact of the start-up phase of several projects, the
simultaneous entry by Titans into new markets, high interest rates and significant
O
dependence on short-term borrowed funds. Expansion, diversification and entry
into new markets were planned strategies and they were expected to have an impact
in profit in 1996-97. I believe we should think as industrialists and brandbuilders
n
and take a longer-term view of the future.
io
A Global Player
at
Bhaskar Batt, Vice- President, Sales and Marketing, Titan Industries, on Titan’s Global
Strategy: ul
Over the years, the Titan brand itself has become one of our strengths. All elements of
marketing have helped make it a premium brand. Our people, excellent pricing, the Tata
irc
culture, and a willingness to innovate are our other strengths.
The watch market was predominantly mechanical when we entered and quartz watches
C
were smuggled into the country. We were able to convince the consumer that looks are
mattered, while durability and reliability were inherent. Today Titan and Timex together
have 70 per cent of the organised marketshare for watches in India.
d
Our international thrust began with West Asia: the UAE, Bahrain, Oman. We were able to
establish volumes using the product ranges we were selling in India. However, sensing
ite
opportunities in Europe a few years ago, we decided to enter the European market. We now
had to design a whole new collection of watches for Europe using their designers and set up
im
separate manufacturing facilities to make those kinds of watches which are complicated in
design.
Our marketing office in Europe contributes 10 per cent to our turnover from export sales.
rL
We are now poised to enter the American market. The reception has been extremely good. In
West Asia initially, we were bought by expat Indians but increasing Arabs are buying our
watches. We have become a brand to reckon with. In Europe, we have a unique positioning –
Fo
a New World watch. We have been received very well.

The upper-end of the market is still very underdeveloped and awareness of the brand
name is limited. What we offer in solid gold will be offered by others in non-precious metal.
Moreover, their prices are extraordinarily high. We do not perceive them as competition as
we are in different league altogether. Titan’s strengths are its ability to design beautiful
products, an understanding of the Indian market and the ability to retail and build a brand
image. We will bring to the business professionalism and ethics.
Source: Business Barons, Vol.1, No.7, December 31, 1996.
1. What are the environmental issues (regulatory and competitive) that are affecting
Titan’s business?
2. What are the specific supply chain management issues that the Senior Vice President,
y
Marketing, is concerned with?
nl
3. What is the impact of the product variety, batch sizes and sourcing on the inventory
held in the supply chain and can this be reduced?
O
n
io
The analysis for this case needs to view Supply Chain Management as a strategic area, which
Titan can leverage to be a responsive player in the watch industry. The implications of
at
various choices, keeping in view the positioning that Titan has built for itself, need to be
examined. ul
Once the right questions in terms of linking Supply Chain Management with the
strategic concerns are posed, the data in the case can help diagnoses of the same. For
irc
example, where dials are concerned, the sourcing and ordering policy can be questioned.
Stocks as of a given month are over 25% obsolete, and after a further three months, 58%
C
have been classified as ‘not usable in the next three months’.

Diagnostics can be on different dimensions: geographic - for sales, channel-wise,
product/component category wise and geographic—for sourcing
d
ite
im
rL
Fo
CASE 6
y
nl
This case looks at how to develop a marketing strategy for CONCOR, for transporting tea
O
from gardens to the processing plants as containerized cargo by rail. The worldwide trend in
product movement is towards containerization, but only 1% of domestic cargo movement
n
in India is containerised. Tea is very well suited to containerization, in principle, but very
little containerization actually takes place, particularly in the domestic segment in India.
io
The case identifies the reasons behind the preference of tea transporters for road over
railways and the constraints faced by the railways and CONCOR. It also explores possible
at
options that would help stem the shift of tea movement from the railways to road transport.
The focus is on developing a logistics strategy that would attract tea transporters to
ul
containerization. Issues such as pricing, value added services like scheduling, tracking of
consignments, insurance, etc are important.
irc
C
CONCOR: TEA TRANSPORTATION

d
Early one morning in August 2000, Mr. Rambhai, Head of the Eastern Region of Container
ite
Corporation of India (CONCOR), was concerned as he ran through the figures of

CONCOR’s performance for 1998-99, on his excel sheet yet again. There had been signals
from top management that the eastern region, more specifically the northeast, ought to do
im
better. That year, the eastern region had contributed 6 per cent of total CONCOR
throughput, with the northeast (Amingaon) contributing less than half per cent. Almost the
entire outward movement from Amingaon consisted of tea for export. The domestic
rL
segment, however, registered much lesser volume and growth, with road transport competing
and taking the lion’s share. The figures for 1999-2000 were expected to be no different.
Fo
Prepared by G Raghuram, and Anusha Dhasarathy, Preeti Monippally, Rohit Phatak, Malavika Pillaim
and Rohithari Ranjan. Research assistance provided by Premlata Agarwal is acknowledged. Case of the
Indian Institute of Management, Ahmedabad, is prepared as a basis for class discussion. Cases are not
designed to present illustrations of either correct or incorrect handling of administrative problems.
Copyright © 2001 by the Indian Institute of Management, Ahemdabad.
Case 6: CONCOR: Tea Transportation 391
Mr Rambhai was now contemplating how to develop the domestic market and curtail the
shift from rail to road transport. He knew that there was an unexploited opportunity in the
domestic movement of tea from the gardens and auction centres in the northeast to various
consumption centres, especially those in western India. (Western India had a high per capita
consumption of tea.) He wanted to develop a comprehensive strategy to target the domestic
y
tea market and exploit its true potential.
nl
CONCOR
O
CONCOR was set up in 1987 under the Ministry of Railways. Its main objective was to
promote containerization and boost India’s international and domestic trade. It was currently
the only provider of containerized freight transportation by rail. CONCOR’s core business
n
was divided into three main activities: containerized transport, terminal operations, and
container freight station operations. Exhibit 1 gives a brief profile of CONCOR. Table 1 of
io
Exhibit 1 gives the infrastructure of CONCOR.
at
The two basic customer segments it catered to, were export/import traffic and the
domestic segment. Export/import container traffic of CONCOR was 5.8 lakh TEUs
(twenty-foot equivalent units) in 1998-99, and constituted 30 per cent of the total export/
ul
import traffic of 19.2 lakh TEUs handled at the ports (Table 2, Exhibit 1). The CONCOR
traffic was more than the ICD (inland container depot) movement of 3.3 lakh TEUs,
irc
indicating that CONCOR itself handled about 2.5 lakh TEUs at other than ICDs.
CONCOR had been focusing on export/import traffic and consequently the share of
domestic traffic out of the total traffic had declined from 41 per cent in 1995-96 to 27 per
C
cent in 1999-2000 (Table 3, Exhibit 1). In absolute terms, however, the domestic traffic had
remained steady. CONCOR was growing at an impressive rate of 20 per cent per annum
d
since its inception, achieving a throughput of 9.3 lakh TEUs in 1999-2000. However,
although CONCOR itself was doing well, the overall cargo movement split between rail and
ite
road traffic had shifted from 80:20 in 1950-51 to 40:60 in 1998-99. The ratio was expected
to be closer to 35:65 by 1999-2000, out of approximately 800 billion-ton km of freight
traffic. Substantial quantities of high value, small volume traffic that typically needed to be
im
consolidated for movement by rail were being moved by road.

The income of CONCOR for 1999-2000 was Rs. 831.42 crore, registering a growth rate
rL
of 21.41 per cent over the previous year. Exhibit 2 gives the profit and loss account of
CONCOR for the years 1999-2000, 1998-99 and 1997-98.
As mentioned earlier, and as seen in Exhibit 3, the eastern region’s traffic contribution to
Fo
CONCOR’s total was six per cent during 1998-99. Within this, Amingaon, which was the
only ICD in the northeast, contributed less than half per cent of CONCOR’s total. About
2,400 TEUs were brought in empty, to be loaded with tea for exports. (80 per cent of the
containerised tea exports were through one company, with four others accounting for the
rest). On the domestic front, just 50 loaded containers had moved out and less than 200
containers had come in with cargo.
Exhibit 4 gives a map of eastern India, with key locations of CONCOR and the tea
industry.
Potential for Domestic Movement of Tea

A description of the domestic tea market is given in the Appendix. Production in 1998 was
870 million kg, out of which 660 million kg were consumed in the domestic tea market and
210 million kg were exported.
y
Mr. Rambhai felt that the railways were well suited for tea transportation within the
country on two counts: distance and damages. The main routes of the domestic movement of
nl
tea were from Assam and Kolkata to Maharashtra and Gujarat, and to a lesser extent to
Madhya Pradesh and Rajasthan. (Exhibit 5 gives the state-wise share of consumption of tea.
O
Maharashtra and Gujarat accounted for over 24 per cent of tea consumption. Gujarat, with a
population of nearly six crores, consumed nearly 70 million kg). Thus, a particular
consignment might have to cover a distance of more than 2,500 km, for which rail could be
n
economical. Moreover, tea consignments were often moved during the monsoon. Tea was
io
adversely affected by water. In open top trucks, even with tarpaulin covers, it was difficult to
protect the consignment from water seepage during the monsoon season.
at
The Indian Container Leasing Company (ICLC), a subsidiary of the public sector Balmer
Lawrie & Co and by far the country’s biggest container leasing company, had identified tea as
ul
the commodity most easily amenable to containerization for movement within the country.
ICLC had identified certain routes, including Guwahati-Kolkata, Kolkata-Ahmedabad, and
irc
Kolkata-Delhi, on which large quantities of tea moved in open top trucks.
The real advantages of rail-based containerised movement would be protection from
damage, wastage, and pilferage and, perhaps the most important, ease of handling.
C
Complaints of pilferage from sealed containers transported by trucks were quite common.
Since the later part of the nineteenth century, when the railways in the northeast were built
d
primarily to source the tea, transportation of tea by rail had grown both for exports and
domestic movement until the 70s. However, with better road access to the northeast, the
ite
advantages of rail were offset by delays in delivery, extra cost of handling at multiple
locations, and losses owing to pilferage, especially in the non-containerized wagon mode.
Containerization could only hold on to some of the export traffic, and currently accounted
im
for about 12 to 15 per cent of tea exports from the northeast. Over the years, the domestic tea
movement out of the northeast had shifted almost entirely to road.
rL
Channels for Sale and Marketing of Tea

The primary channels for sales and marketing of tea were auctions or garden sales. The recent
Fo
trend had been an increase in garden sales (Exhibit 6). In the recent past, the proportion of
garden sales had been increasing, from a low of 23 per cent in 1985 to 49 per cent in 1998.
Most of the exported tea, however, was bought off the auctions. While traditionally the
Kolkata auction had catered to the export market, the significance of the Guwahati auction
was increasing. The shipper received instructions from the buyer as to the mode of transport,
specific shipping line, etc. The primary source of tea which was containerized from
Amingaon was either direct garden-based exports or export tea sourced from the Guwahati
auction.
Exhibit 7 gives state-wise tea production in the past decade. Tea production in the
northeast (Assam and West Bengal) accounted for 77 per cent of total production of 870
million kg in 1998. Out of the production in the northeast (670 million kg), about 50 per
cent were direct sales from the gardens. The balance sales were accounted for by the
Guwahati (45 per cent), Kolkata (30 per cent), and Siliguri (25 per cent) auctions. The share
y
of the Kolkata auction was declining, while that of the Guwahati and Siliguri auctions was
increasing. Exports were almost entirely through Kolkata and Haldia ports. Out of all the tea
nl
moved from this region, domestic movement accounted for 76 per cent of the total.
In the domestic market, a large number of intermediaries operated between the garden
O
and the retailer. Hence, warehouses were required at each stage: origin, destination, and
transportation nodes. This could translate into an opportunity for CONCOR if it could
provide smooth and coordinated logistic services from the garden to the buyer’s factory.
n
The reasons domestic buyers preferred trucks to the railways were as follows:
io
1. Service inflexibility: No less than container load was allowed. Pooling was not
organised. The domestic segment did not have on-demand service.
at
2. Claims: There was very low leverage with the railways in case of damage or loss.
3. Large volumes: Often, volumes being moved were not large enough to justify a whole
ul
rake (a full train load).
4. Door to door service: This was offered by truckers.
irc
5. Credit: Truckers offered three month credit, while CONCOR required buyers to pay in
advance.
C
Exhibit 8 gives details of train services offered by CONCOR in the eastern region.
Operating Constraints
d
ite
While Mr Rambhai was considering the various options available to him to increase domestic
movement of tea by CONCOR, there were some operating constraints that he perceived.
The railways had little presence in most parts of the northeast. Capacity constraints on
im
movement, both intra-regionally, and with the rest of the country, were a major problem of
the region. Because of capacity constraints, the railways chose to concentrate on the
movement of bulk materials for the core and priority sectors, namely power, steel, cement,
rL
oil, foodgrain, etc. thus losing its clientele in high value non-bulk sectors which often
recorded higher growth rates. Actual demand for movement into the northeast was around
20 to 25 rakes a day. But the railways could move only 7 to 8 rakes a day.
Fo
Militancy was another problem. Road transporters would pay hafta, or protection money,
which the railways or CONCOR (being government agencies) could not do.
Tariff
The tariff policy of the railways where passenger traffic was subsidised by freight meant a
continual increase in freight rates which, in turn, was driving away even some of the long
distance bulk traffic from the railways to road.
On the other hand, road transporters, being largely in the private sector, moved
aggressively to occupy the space vacated by the railways. They were aided by a liberal permit
and regulatory system for national trucking, cheap finance made available by the banking
sector, and an energy pricing policy which subsidised diesel. (The railways paid an
exceptionally high tariff for electricity consumption.)
y
The railways charged freight rates on a per TEU km basis from CONCOR to simplify the
nl
freight charging mechanism for the benefit of the customer as well as to avoid creating a large
organisation and infrastructure (weighbridges, etc.) which would be necessary if the freight
O
was on a per kg or per quintal basis. (Freight rates on TEU basis were calculated by
considering the average weight being hauled for a particular commodity through the
container wagon.) Exhibit 9 gives freight rates for movement of containers by CONCOR in
n
the eastern region.
io
Tea, being a bulky but lightweight commodity, containers would typically be stuffed with
about 6 to 7 tonnes of tea, as opposed to road transporters who could carry 9 to 10 tons. The
at
cost of transportation per ton of tea was therefore higher when it was containerised. The
CONCOR top brass said that freight charges for a round trip from Haldia to Amingaon and
ul
back would soon go up by about Rs. 1500 per TEU. The hike was necessary because
CONCOR would be required to pay higher tariff to the railways. Currently, 90% of the
irc
TEU rates charged by CONCOR went to the railways. The loaded freight charges from
Amingaon to Haldia were expected to go up to Rs. 9800 per TEU (from Rs. 9100). The
empty freight charges from Haldia to Amingaon were expected to go up by Rs. 6800 per
C
TEU (from Rs. 6200). The terminal handling charges of Rs. 3500 per TEU at Amingaon was
also expected to go up.
d
CONCOR did not have much flexibility in lowering the rates as the charges to be paid to
the railways were fixed. However, in the case of bulk movements, quantity discounts were
ite
offered within the margins available to CONCOR after payments to the Indian railways.
High transportation charges were a big issue since many tea shippers argued that ICD
im
transportation had been rendered unattractive vis-a-vis road transportation mainly because
of the high cost of moving empty containers between Kolkata/Haldia and Amingaon. There
was virtually no cargo available for movement by containers between Kolkata/Haldia and
rL
Amingaon. (In 1999-2000, while 2,015 containers of tea were moved from Amingaon, only
22 loaded containers were received at Amingaon.) As a result, empty containers had to be
moved, whatever the cost.
Fo
Although CONCOR did charge differential rates for loaded and empty containers, the
rates were still too high when compared to road. Of course, truck drivers had the flexibility to
cut rates, while railways resorted to a uniform tariff structure, which was not modified
according to supply and demand.
Questions
Keeping all the above issues in mind, Mr Rambhai now needed to develop a strategy to
stimulate growth in the domestic segment. The questions he needed to find answers to, were:
Who should be CONCOR’s key customers? On which point in the value chain should it
focus? For example, should CONCOR attempt to provide door-to-door delivery services
y
from the garden to tea factories? For this, it might be possible to have tie-ups with truckers
nl
and provide multi-modal transport solutions. Exhibit 10 gives a structure of flows of tea. Mr
Rambhai was wondering if he could assess the quantum of flows on each segment to see the
O
potential for containerised movement.
Given that flexibility in pricing was limited, CONCOR would have to decide on the
value-added services it would offer like warehousing, time definite deliveries, tracking, or
n
innovative solutions like organising pooled rake loads. At a more fundamental level, how
io
were changes to be implemented in the organisation in order to change the negative image of
CONCOR and the railways among tea merchants?
at
ul
Exhibit 1: Brief Profile of CONCOR
irc
As a pioneer in the field of containerised transportation, CONCOR had grown impressively
since its establishment in 1987 under the Ministry of Railways. Its main objective was to be
a catalyst for promoting containerisation and to give a boost to India’s international and
C
internal trade and commerce by organising multimodal logistics support.

CONCOR provided logistics services and support for the international and domestic
cargo industry in India. It was the only means by which shippers could obtain containerised
d
freight transportation by rail. CONCOR’s network of facilities consisted of inland container

ite
depots (ICDs), container freight stations (CFSs), port container terminals, port side
container terminals, and domestic container terminals. Table 1 gives a list of CONCOR’s
terminal facilities at various locations. ICDs and the other terminals were consolidation
im
points up to/from which CONCOR handled the transportation. CFSs and other terminals
where cargo-handling facility was provided, enabled customers to stuff and destuff
containers. CONCOR’s core businesses could be divided into three main operations:
rL
containerized freight carrier, terminal operator, and container freight station operator.
Freight Carrier: CONCOR provided rail connectivity as the mainstay of its transportation
service. While some ICDs like Pithampur (Indore), Mulund (Mumbai), Milavittan
Fo
(Tuticorin), Babarpur (Panipat), Daulatabad (Aurangabad) and Malanpur (Gwalior) were

exclusively road-fed ICDs, majority of ICD and terminals in the CONCOR network were
rail linked. As rail was price competitive over road for long leads (almost 20 per cent cost
advantage), CONCOR had a natural competitive advantage.
Terminal Operator: CONCOR’s terminal operations were spread across the country.
CONCOR owned and operated nearly 30 international container terminals and 10 domestic
container terminals. Terminal operations were at a variety of locations, differentiated as deep
and medium hinterland, portside and within port. For international operations, the
intention was to cater to a diversity of clients, irrespective of lead from ports. For domestic
operations, the objective was to increase the coverage and develop the container market.
Container Freight Station Operator: As a CFS operator, CONCOR provided value-added
cargo handling services by offering transit warehousing to export/import cargo. Another
y
segment growing in importance was bonded warehousing which enabled importers to store
nl
cargo and ask for partial releases, thereby deferring duty payment. Consolidation of less than
container load (LCL) cargoes, air-cargo complexes, and hazardous cargo handling were
O
business growth areas.
The two basic customer segments were export/import and domestic. CONCOR believed
that only about 30 per cent of port traffic originated from and terminated at places within
n
300 km from the port. The remaining 70 per cent was to and from the hinterland,
io
representing a potentially large demand for CONCOR’s services. However, as of now, only
17 per cent of export containerised movement and 17 per cent of import containerised
at
movement was from/to ICDs. Table 2 gives the current profile of export/import containers
handled by (i) CONCOR, (ii) ports, and (iii) ICDs. CONCOR’s share of export and import
ul
traffic was 30 per cent in 1998-99. It is interesting to note that ICD handling was less than
total CONCOR handling, indicating the significant role of port side handling by
CONCOR. ICD handling constituted 57 per cent of CONCOR’s export traffic and 55 per
irc
cent of CONCOR’s import traffic. Movement of empty containers of shipping lines was also
undertaken by CONCOR to balance the differential levels of exports and imports.
C
CONCOR was trying to develop the domestic container segment, targeting the piecemeal
cargo movement in the country. CONCOR started Contrack service in 1997, and later on
Conraj service. These were scheduled trains with specified frequencies, running between pre-
d
determined pairs of points with guaranteed transit times. Client-focused customised services
ite
were the backbone of this segment. Large companies with substantial movement
requirements were also being targeted. CONCOR owned and operated over 4,500
containers in the domestic sector, some of which were dedicated to specific industries for
im
close-circuit operations. The share of export/import and domestic traffic over the last five
years is given in Table 3.
rL
Fo
Table 1: CONCOR’s Infrastructure Network
ICD with CFS Port Container Terminal

(Customs-Bonded, Rail Linked, Cargo Handling Facility, (Customs-Bonded, Rail Linked, Rail-Road Transfer, No Cargo
Direct Port Link) Handling Facility, within Port Premises)
y
1. Tughlakabad (Delhi) 8. New Mulund (Mumbai) 1. Harbour of Madras (Chennai)
2. Ajni (Nagpur) 9. Belanganj (Agra) 2. Kandla
nl
3. Whitefield (Bangalore) 10. Guntur 3. Haldia
4. Coimbatore 11. Anaparti
O
Port Side Container Terminal
5. Sanat Nagar (Hyderabad) 12. Moradabad
(Not Customs-Bonded, Rail Linked, Rail-Road Transfer, Cargo
6. Amingaon (Guwahati) 13. Madurai
Handling Facility, within Port Premises)
7. Tondiarpet (Chennai) 14. Sabarmati (Ahmedabad)
n
1. Cossipore Road (Kolkata)
Road ICD with CFS 2. Shalimar (Kolkata)
io
(Customs-Bonded, Cargo Handling Facility, No Rail Link) 3. Wadi Bunder (Mumbai)
1. Pithampur (Indore) 4. Dronagiri Node (Navi Mumbai)
at
2. Mulund (Mumbai) Domestic Container Terminal
3. Milavittan (Tuticorin)
1. Tughlakabad (Delhi) 6. Shalimar (Kolkata)
4. Babarpur (Panipat)
5. Daulatabad (Aurangabad)
ul 2. Dhandari Kalan (Ludhiana)
3. Salem Market
7.
8.
Cossipore Road (Kolkata)
Kankaria (Ahmedabad)
6. Malanpur (Gwalior)
irc
4. Whitefield (Bangalore) 9. Wadi Bunder (Mumbai)
ICD without CFS 5. Tondiarpet (Chennai) 10. Turbhe (Navi Mumbai)
(Customs-Bonded, Rail Linked, Rail-Road Transfer, No Cargo Future Terminals
C
Handling Facility, Direct Port Link)

1. Dadri, New Delhi 7. Miraj
1. Dhandari Kalan (Ludhiana)
2. Rewari 8. Ballabhgarh
2. Chinchwad (Pune)
3. Balasore 9. Rajkot
d
3. Cochin
4. Jaipur 10. Ankleshwar
4. Vadodara
ite
5. Kanpur 11. Bhusawal

5. Raxaul
6. Cossipore Road (Kolkata)
Source: Exim India, ICD-Sabarmati Special, June 2000
im
Table 2: Export/Import Container Traffic

rL
(Lakh TEUs)
Export Import Total

Fo
Loaded Empty Total Loaded Empty Total Export/Import
CONCOR 1999-00 (1) 2.4 0.9 3.2 2.2 1.2 3.4 6.6
CONCOR 1998-99 (1) 2.0 0.8 2.9 1.7 1.2 2.9 5.8
Ports 1998-99 (2) 8.7 0.9 9.6 6.9 2.7 9.7 19.2
ICD Movement 1998-99 (2) 1.7 1.6 3.3
Sources: 1) Exim India, ICD-Sabarmati Special, June 2000
2) Major Ports of India: A Profile 1998-99
Table 3: Export/Import and Domestic Traffic

(Lakh TEUs)
Year 1995–96 1996–97 1997–98 1998–99 1999–00
Export/Import 3.5 4.2 5.0 5.8 6.8
y
Domestic 2.4 2.8 2.3 2.2 2.5
nl
Total 5.9 7.0 7.3 8.0 9.3
Domestic % 40.7 40.0 31.5 27.5 26.9
O
Source: CONCOR Annual Reports and Company Sources
n
Exhibit 2: Profit and Loss Accounts
io
(Rs. lakh)
at
Year Ended Year Ended Year Ended
31.03.2000 31.03.1999 31.03.1998
I INCOME
Terminal and Other Service Charges
ul
83141.6 68477.3 60625.4
irc
Gross Interest on Short Term
Investments/Deposits 2924.2 2845.5 1870.6
Miscellaneous Income 366.1 274.0 196.8
Excess Provision Written Back 56.1 39.2 162.3
C
Total Income 86488.0 71636.0 62855.1

II EXPENDITURE
Terminal and Other Service Charges 51574.0 45280.0 40000.3
d
Employees Remuneration & Benefits 1179.6 1066.6 709.3

Administrative & Other Expenses 4945.2 3263.2 2996.3
ite
Total Expenditure 57698.8 49609.8 43705.9

Profit Before Depreciation 28789.1 22026.2 19149.2
Less: Depreciation 2261.8 1227.4 1154.4
im
Profit After Depreciation 26527.3 20798.8 17994.9

Less: Provision for Taxes 9000.0 6651.2 6326.8
Profit After Tax 17527.3 14147.6 11668.1
rL
Add/Less: Prior Period Adjustments (Net) 231.5 (82.4) (87.8)

Net Profit 17758.9 14065.2 11580.3
III APPROPRIATIONS
Fo
Proposed Dividend 3574.5 2924.6 1299.8

Corporate Tax on Dividend 643.4 292.5 130.0
General Reserve 1775.9 1406.5 11588.0
Balance Carried to Balance Sheet 11765.1 9441.6 8992.5
Total Appropriations 17758.9 14065.2 11580.3
Source: CONCOR Annual Reports 1999-00 and 1998-99
Exhibit 3: CONCOR’s Performance: Eastern Region (1998-99)

Overall
(TEUs)
Eastern Region CONCOR’s Total % of Total
y
Export/Import 12004 576790 2.08
nl
Domestic 36104 225156 16.04
Total 48108 801946 6.00
O
Export/Import Traffic
(TEUs)
n
Export Import Total
io
Loaded Empty Total Loaded Empty Total Export/Import
Cossipore Road 115 2225 2340 594 594 2934
at
Shalimar 9 120 129 3 346 349 478
Amingaon 2412 2412 22 2422 2444 4856
Raxaul
Haldia

265
80
289
ul
80
554
80
2388

353
80
2741
160
3295
irc
Adhoc total 281 281 281
Total 2801 2995 5796 2493 3715 6208 12004
Domestic Traffic
C
(TEUs)
Outward Loaded Inward Loaded Total

d
Cossipore Road 2728 1351 4079

ite
Shalimar 3205 8292 11497

Amingaon 50 189 239
Total 5983 9832 15815
im
Amingaon ICD
The revenues of ICD/Amingaon were approximately Rs 3.95 crore for 1999-2000. The
rL
approximate distribution of traffic between various tea companies at Amingaon was as

follows:
Fo
Company Share (%)
Williamson and Magor 80

Assam Company 10
Andrew Yale, Brook Bond, Tata Tea About 3 each
Expansion Plans for CONCOR, East

CONCOR proposed to have a new ICD at Tinsukia in Assam. A feasibility report, assigned
by CONCOR to Indian Market Research Bureau, was expected to be ready soon. The bulk
of tea that was exported through Amingaon (Guwahati), ICD originated from tea gardens
located around Tinsukia. This was because most of the tea companies participating in the
y
ICD movement for exports had gardens concentrated in that region. If an ICD was installed
nl
close to these gardens, the long-road haulage of the consignments, as was undertaken to reach
Amingaon ICD, would be rendered redundant. There would be substantial savings on
O
transportation and handling costs.
There were some concerns with respect to the Tinsukia ICD. First, the broad-gauge
railway line was yet to be extended to Tinsukia from Guwahati. Next, there were fears in
n
many quarters that once the ICD became operational, the Amingaon ICD would lose its
importance, as many tea shippers having gardens on the south bank of the Brahmaputra
io
would abandon Amingaon in favour of Tinsukia.
at
However, the apprehension, according to many, was unfounded. According to leading
shippers, the Tinsukia ICD, once operational, would cater mainly to the domestic trade - tea,
plywood etc. The Amingaon ICD, according to them, would continue to account for the
ul
bulk of exports. The tea companies having gardens on both sides of the river, it was pointed
out, would prefer Amingaon to Tinsukia. Also, many producer-exporters on the north bank
irc
who had so far shied away from Amingaon could be persuaded to use it.
Connecting Kolkata/Haldia with the Tughlakabad ICD had remained a thorny issue.
C
Recently, American President Line had moved five rakes - three carrying imports from
Kolkata/Haldia to Tughlakabad and two others carrying exports in the opposite direction.
More shipments in both directions were expected to be achieved.
d
For shipments to and from Nepal, a full-fledged ICD on the India-Nepal border was
ite
expected to meet the requirements of the trade with/of the Himalayan kingdom. The present
terminal facility at Raxaul was clearly inadequate. There was a talk of having a full-fledged
ICD to be set up with World Bank assistance at Birgunj on the Nepalese side of the border.
im
The effort would be meaningful if supporting facilities were strengthened on both sides.
Source: LINK – Global Trade and Freight Review, CONCOR, 1999.
rL
Fo
Exhibit 4: Eastern India
y
nl
O
n
io
at
ul
irc
C
Source: www.mapsofindia.com
d
ite
im
rL
Fo
Exhibit 5: Estimated Domestic Tea Consumption
Name of State Share of Total Per Capita

Consumption Consumption
(%) (kg/year)
y
Kg/annum
nl
MAHARASHTRA 12.9 0.96
GUJARAT 11.3 1.18
O
UTTAR PRADESH 8.2 0.34
PUNJAB 8.0 1.49
RAJASTHAN 8.0 0.78
n
WEST BENGAL 7.0 0.42
KERALA 6.1 1.06
io
MADHYA PRADESH 5.9 0.58
ANDHRA PRADESH 4.6 0.40
at
KARNATAKA 4.6 0.71
TAMIL NADU 3.9 0.33
ASSAM
BIHAR
ul 3.4
3.4
0.67
0.19
HARYANA 3.2 0.99
irc
JAMMU & KASHMIR 1.8 1.55
ORISSA 1.8 0.24
OTHERS 5.9
C
TOTAL 100.0 0.70

Sources: 1) CTTA. TEA DIGEST, 1998
d
2) Reddy V N and Bose A, IIM Calcutta, 1995

ite
im
rL
Fo
Exhibit 6: Auction Sales vs Garden Sales

(Million Kgs)
Year Total Auction Garden Share of

Production Sales Sales Garden
y
Sales
nl
1980 589.2 307.0 282.2 48
1981 560.4 376.2 184.2 33
O
1982 560.6 358.8 201.7 36
1983 581.5 338.9 242.6 42
1984 639.9 420.8 219.0 34
n
1985 656.2 505.3 150.9 23
1986 620.8 468.4 152.4 25
io
1987 665.2 472.5 192.8 29
1988 700.0 497.5 202.5 29
at
1989 688.1 477.4 210.7 31
1990 720.3 482.2 238.1 33
1991
1992
754.2
732.3
ul 501.6
448.1
252.6
284.3
33
39
irc
1993 760.8 441.7 319.1 42
1994 752.9 428.3 324.6 43
1995 756.0 428.4 327.6 43
C
1996 780.2 443.1 337.1 43

1997 810.6 459.0 351.6 43
1998 870.4 442.4 428.1 49
d
Source: CMIE Sector Report: TEA, July 2000

ite
im
rL
Fo
y
404
nl
Exhibit 7: Production in Major Tea Producing States
O
(in per cent)

n
1990 1991 1992 1993 1994 1995 1996 1997 1998 1990–98 1996–98
CAGR CAGR
io
Assam 53.89 52.59 56.14 53.95 53.22 53.26 54.34 52.48 53.17 -0.23 -1.08
West Bengal 20.79 21.10 20.63 21.38 21.10 20.84 21.12 20.97 22.48 0.53 3.18
at
North India 75.67 74.64 77.83 76.43 75.43 75.21 76.67 74.63 76.86 0.03 0.12
Tamil Nadu 15.35 15.90 14.11 14.83 15.61 15.60 14.85 16.11 14.37 -0.12 -1.62
ul
Karnataka 8.42 8.86 7.48 8.15 8.39 8.57 7.89 8.59 8.11 -0.14 1.39
Kerala 0.55 0.60 0.58 0.59 0.57 0.62 0.59 0.67 0.65 1.78 5.38
irc
South India 24.33 25.36 22.17 23.57 24.57 24.79 23.33 25.37 23.14 -0.08 -0.41
Total India (Million kg) 720.34 754.19 732.32 760.83 752.90 756.02 780.23 810.61 870.41 1.85 5.62
Total Exports (Million kg) 210.02 202.92 174.96 175.32 150.69 168.00 161.70 203.00 210.34 1.00 1.14
C
Total Export Value (Million Rs) 11133.51 11345.53 9953.31 11612.64 9891.37 12080.16 12468.72 17747.78 23094.36 1.10 1.36
Source: Asopa, V.N., Competitiveness in Global Tea Trade, Centre for Management in Agriculture, IIM, Ahmedabad, 2000
d
ite
im
rL
Fo
Exhibit 8: Scheduled Train Services (Eastern Region)

Export/Import: Fixed Services
CD Amingaon PSCT Cossipore Road Seasonal (2-3 trains per month)
y
CD Amingaon PCT Haldia
nl
Export/Import: On Demand (Bi-Directional) Services
O
Terminal 1 Terminal 2
Mumbai area Kolkata area, Haldia
Chennai area Kolkata area, Haldia
n
Visakhapatnam port Kolkata area
ICD Sabarmati Kolkata area
io
Domestic Services
at
Stream Frequency
Tughlakabad (Delhi) Cossipore Road

ul Weekly
Gandhidam Shalimar Fortnightly
Kanpur Cossipore Road Weekly
irc
Lucknow Cossipore Road Weekly
Moradabad Cossipore Road Weekly
Sanat Nagar (Hyderabad) Cossipore Road Fortnightly
C
Mumbai Area Amingaon On Demand

d
Mumbai area includes JNPT, Mumbai port, Wadi Bunder, Mulund and New Mulund.
Kolkata area includes Kolkata port, PSCT Cossipore Road, and Shalimar.
ite
Source: LINK Global Trade and Freight Review, CONCOR, 1999.

im
rL
Fo
Exhibit 9: Freight Rates for Movement of Containers by CONCOR (Eastern Region)

(Rs/TEU)
Between Vizag Port and Empty Loaded
y
Delhi 12400
Bangalore 7100
nl
Coimbatore 7700
Chennai 5000 6900
O
Hyderabad 4300 6200
Guntur 3000 4400
Mumbai 9500 13500
n
Between Haldia and
io
Delhi 8900 13400
Ludhiana 10400 15400
Cossipore Road/Shalimar/Kolkata 1800 2100
at
Chennai 10200 14800
Amingaon 6200 9100
Nagpur
Mumbai
ul 6700
11800
9600
17900
irc
Vizag 6100
Between Amingaon and
Cossipore Road 6000
C
Vizag 11800
Haldia 6200 9100
Mumbai 15000
d
Between Shalimar/ Kolkata/ Cossipore Road and

Mumbai 11800 17900
ite
Chennai 10700 15500

Sabarmati 6100
Nagpur 12400
im
Vizag 6800
Source: LINK Global Trade and Freight Review, CONCOR, 1999
rL
Truck Rates
The average rate from Guwahati to Ahmedabad was around Rs 12,000 per truck. A truck
Fo
would carry 9-10 tonnes of tea.

Source: Somabhai Tea Processors Ltd.
Exhibit 10: Structure of Flows of Tea
PM Ex
Blending
y
Locations
nl
Tea Gardens PM
PM PM Do
O
Auction Centres
n
io
TM Ex
at
Tea Gardens Blending
P Locations
ul TM
TM Do
irc
P Ex
C
d
P Do
ite
P Producer P Do Producer Sales to Domestic Market

im
PM Producer Merchant P Ex Producer Sales to Export

TM Trader Merchant PM Do Producer Merchant Sales to Domestic Market
PM Ex Producer Merchant Sales to Export
rL
TM Do Trader Merchant Sales to Domestic Market

TM Ex Trader Merchant Sales to Export
Fo
APPENDIX
THE DOMESTIC TEA MARKET

Domestic tea consumption was estimated to have been around 660 million kgs in 1999 and
y
was growing at an average annual rate of 3.5 per cent per annum. Production, on the other
nl
hand, grew at a CAGR of 1.4 per cent during the last decade. The exportable surplus had
therefore been declining over the last few years. Exports in 1998 amounted to 210 million kg.
O
In the domestic market, only 30 per cent of the tea produced was sold in packaged or
branded form while 46 per cent was sold loose. Tea was, by far, the most popular beverage
in India. Its penetration was 89.3 per cent in urban areas and 72.6 per cent in rural areas.
n
Across India, tea penetration averaged 77.2 per cent.
io
Interestingly, tea penetration was higher in medium sized towns1 at 93.3 per cent than
in metros 2 where it was about 90.5 per cent. This could be due to the fact that coffee
at
penetration was significantly higher in metros (24.2 per cent) than in medium sized towns
(16.7 per cent).
Characteristics of Tea
ul
irc
Tea supply depended upon weather conditions. Tea prices fluctuated like those of any
commodity driven by temporary shortages and surpluses. Demand for branded tea was
highly price sensitive, with consumers shifting to relatively cheaper loose tea when prices
C
of branded or packaged tea increased. The aggregate demand for tea, however, was not
price sensitive.
d
Tea was a very sensitive commodity and was prone to damage easily under adverse and
careless handling. The commodity was perishable and demand relatively inelastic to price.
ite
While demand had a secular growth rate, supply could vary depending on climatic
conditions in the major tea growing countries. Unlike other commodities, tea price cycles
had no linkage with the general economic cycles, but varied with agro-climatic conditions.
im
Key Players in Tea Sector

rL
The major players in the tea market could be classified as:

✧ Producers
✧ Merchants
Fo
✧ Producer-Merchants
Producers were basically garden owners who sold their produce directly or through
auctions to merchants.
1
Medium sized towns are defined as those with population 0.5 to 1 million.
2
Metros are cities with population above 1 million.
Merchants bought the primary product and processed it further (blending, etc.) and
either exported this or sold to wholesalers, who then started off the value chain till the tea
reached retailers and consumers.
Producer-merchants were those who both produced tea in their own gardens and
purchased from auctions, and performed blending, etc. The big tea companies fell in this
y
bracket.
nl
The packed tea market was dominated by Hindustan Lever with over 45 per cent
market share followed by Tata Tea, which had an estimated market share of 28 per cent.
O
The rest of the market was highly fragmented. Some leading national players were
Goodricke, Godfrey Phillips and Duncans. HLL procured most of its tea requirements
from outside. Other major tea plantation companies included Harrison Malayalam,
n
Warren Tea, Duncans, Goodricke, Bishnauth, and Eveready. Tata Tea, although the largest
integrated tea producer in the world, accounted for only 7 per cent of tea production in
io
India. A brief profile of some of the players is given below.
at
Tata Tea Limited (TTL): Tata Tea had gardens in both the Northeast and the South,
besides tea estates in Sri Lanka. It owned and operated 54 tea estates (21 in Assam, 4 in
West Bengal, 24 in Kerala and 5 in Tamil Nadu), spread over 25,714 hectares of land. TTL’s
ul
estates yielded 62 million kg of tea in 1998-99. About 75 per cent of this was sold in
packed form, while the balance was sold loose through auctions. TTL sold an estimated
irc
volume of 42 million kg in packed branded form in India. It recorded a total income of Rs
879 crore in 1997-98 (Rs 717 crore the previous year) and net profits were up by 74 per
cent to Rs 102.16 crore.
C
Bishnauth Tea: The Bishnauth Tea Company was one of the largest tea-producing
companies of the Williamson Magor Group. The company had focussed
d
its attention on the export market and had generally recorded exports
ite
aggregating over 35 per cent of its turnover.

Goodricke Group: The Goodricke Group was a leading producer of Darjeeling tea and
the third largest tea producer in the country. The company owned 17 gardens spread over
im
9,500 hectares. Primarily a seller in the auction market, Goodricke also had an equal
proportion of sales in the bulk and packet tea segments, besides a significant presence in
the export market. Besides selling bulk teas in domestic and international markets, the
rL
company also marketed tea in consumer packs. Goodricke sold 18.9 million kg of bulk and
packed tea in 1998. The company set up an export-oriented tea packaging unit near
Mumbai.
Fo
Duncans Industries Limited (DIL): Duncans Industries Limited (DIL) was the third
largest player in the Indian packet tea market. DIL had a market share of approximately 9
per cent and sold about Rs 1.5 billion ($37.5 million) worth of tea in 1996-97. Almost the
entire amount sold was in packets. DIL owned 16 tea gardens, 3 blending units, 6 packing
units, and 22 depots. While tea gardens and blending units were concentrated in the
northeast, the packing units and depots were scattered all over the country. DIL had a
countrywide network of 11,500 distributors through whom the tea ultimately reached
about 325,000 retailers.
Eleven of the gardens had on-site factories that processed tea leaves. The produce of
gardens without factories was transferred to those equipped with processing units. The
manufacturing of tea started when the plucked green leaves were withered (the leaves were
placed on a trough and hot air was blown over them to remove the moisture). The usual
moisture content in raw leaves was about 68 per cent; the process of withering continued
y
for about 8 to 10 hours and took away about 10 to12 per cent of the moisture. The leaves
were then cut and curled in the CTC (crush, tear and curl) machine. CTC tea, which was
nl
still green in colour, was spread out in large trays for some time, depending upon the
temperature and humidity in the room for fermentation. The fermented granular tea,
O
which was still wet, was charged into dryers and subjected to temperatures of 250 to 260
degrees Fahrenheit. The output, now dry and brown in color, was put through the sorter
machine to separate the tea according to the size of the granules and remove fibres from
n
the tea. Samples were then rigorously tested and the manufactured tea graded according to
io
quality.
Tea from the gardens was sent to blending units, where teas of various grades were
at
mixed in fixed proportions to produce blends of stipulated quality. Well known DIL blends
included Sargam, Double Diamond and the prestigious Runglee Rungliot. Tea for
ul
blending, was also often purchased from auction centres. The progress of the Indian tea
industry owed much to the tea auction system, which provided opportunities for sale and
irc
purchase of large quantities of tea of many varieties. The main auction centres were
Kolkata, Siliguri and Guwahati in northern India. Tea was also occasionally sold by DIL at
auctions.
C
Blended tea was transported to packing units. Several gardens had packing facilities.
There were also six off-garden packing units in various parts of the country. In packing
units, tea was placed in packets of different sizes and type. The combination of blend, size
d
and packing type determined a line. Thus, Sargam tea, packed in 500 gram cartons,
ite
constituted a line, while Double Diamond, packed in 100 gram polythene pouches,
constituted another. About 120 lines were produced in all.
DIL had a network of 22 depots in various parts of the country. The depots received
im
stocks of packet tea from different packing units and in turn despatched these to
distributors in the respective jurisdiction. Certain depots had to be supplied from specific
packing units in order to take advantage of sales tax benefits provided by some states.
rL
There were 11,500 distributors through whom the needs of about 325,000 retailers were
met.
Waghbakri Tea: This company was the leading player in Gujarat, with presence in
Fo
Maharashtra and Rajasthan. Sales during 1999-00 was Rs 190 crore. It sold 8 million kg of
its branded packed Good Morning Tea and about one million kg as branded loose tea. The
average price of tea was Rs 170 per kg. Its main factory was in Dholka, about 40 kms from
Ahmedabad. There was competition not only from national players like HLL and TTL, but
other local brands like C Somabhai Tea. Waghbakri Tea had a 15 per cent market share in
Gujarat.
Source: OR/MS Today, April 2000 (for DIL), Annual Reports and Internet.
1. What customer segments should the head of the Eastern region of CONCOR focus
to provoke growth in containerised rail transportation of tea?
2. At what point in the value chain of tea should CONCOR focus, for maximum
y
effectiveness?
nl
3. What value added services and/or partnerships should he consider for an effective
service offer?
O
4. Increased tea traffic will most likely involve further flow of empty containers back to
the eastern region. Would these costs be borne by the intended market?
n
io
at
It would be important to understand the changing trends of the tea industry, in terms of size
of gardens, role of auctions, export versus domestic consumption, and market spread. A
ul
value stream mapping for tea transport through rail containers may be useful. Domestic
movement through containers could be possible in some markets in the country, in addition
irc
to the export markets.
While providing value added services, the delivery competence needs to be considered,
C
along with the institutional arragements. Pricing would become an issue. Value added
services could be an opportunity in increasing contribution, since on rail transport, the
pricing of containers by Indian Railways is a significant cost. Empty flows can be handled in
d
more than one way, by looking at markets at both ends.

ite
im
rL
Fo
CASE 7
y
nl
O
Chilli was the main earner in the Indian spices export basket. Incidents of adulteration of
chilli powder (by carcinogenic dyes) by some unscrupulous and careless exporters had led to
large-scale product recalls in the UK and tainted the image of Indian spices export industry.
n
This resulted in a complete shake up of the sector with serious consequences, leading to the
introduction of mandatory and expensive testing, certification, stringent safety regulations,
io
food recall and threats by EU of banning imports from India.
This case focuses on the events that led to this as well as the status of food safety
at
regulation in our country. The role of Spices Board of India (SBI) has been instrumental in
protecting the brand image of Indian spices exports. The case revolves around SBI, the
ul
choices it had and the possible implications towards damage control and prevention.
irc
C
CHILLI IN SOUP (A)

d
INTRODUCTION
ite
It was a damp morning on June 10, 2003 with overcast sky, raining outside. The monsoon
im
had arrived in Kerala a few days ago. Mr C J Jose, a tall well-built man in his fifties, the
Chairman of the Spices Board of India (SBI), was in his spacious office deeply engrossed in a
rL
Case prepared by Tathagata Bandyopadhyay, G Raghuram, and Neeraj Sisodia, IIM Ahmedabad.
We acknowlwdge the research assistance by Vishal Kashyap. We are thankful to Mr Jose, Chairman and
Fo
other senior executives of the Spices Board of India for their discussion and data support. Some data has
been masked to protect sensitivities. Teaching material of the Indian Institute of a Management,
Institute of Management, Ahmedabad, India, 380015.
Case 7: Chilli in Soup (A) 413
file. It was 10 a.m. in the morning. His phone started ringing. It was a call from the Secretary,
Department of Commerce, Ministry of Commerce and Industry, Government of India. The
matter was of serious concern. The Secretary seemed to be worried and conveyed to him the
following message. “Yesterday evening, the Ministry received a communication from the
Charge d’Affaires, European Union (EU), regarding a notification. It came via the existing
y
Rapid Alert System for Food and Feed (RASFF) of EU. The notification stated that French
scientists have detected traces of Sudan I, a banned carcinogenic dye in a British food
nl
product in the last month. Further investigation revealed that the source of Sudan I was red
chilli powder exported from India. The British firm used it as an ingredient. The Ministry
O
was requested to inform the competent authority in India of these findings in order to help
avoid such problems in the future. My office will be sending a fax of the notification in an
hour”. Closing the conversation, he said, considering the seriousness of the matter, the
n
Ministry would like to be periodically apprised of the actions SBI would be taking in its
io
capacity.
Mr. Jose hung up the phone and brooded for some time over possible actions he could
at
take. He called his Personal Assistant, Mr. Raphael, and asked him to cancel all his meetings
and appointments of the day unless something was exigent. He then discussed the matter
ul
with Mr. S Kannan, Director (Marketing), over phone, and immediately decided to summon
a meeting at 11 a.m. of all the departmental heads and a few other officers whose inputs were
irc
desirable. (Exhibit 1 gives the functions and organisation structure of SBI.) He was expecting
the fax to come before the meeting would start. He did all this in a few minutes after
receiving the call.
C
Mr. Jose had, in his long administrative service (as an officer of the Indian Administrative
Service), faced many a crisis in various capacities and always found these an interesting part
of life. He was posted as the Chairman of SBI in April 2001. Until the previous day, it had
d
been a relatively easy posting, doing routine regulatory work, besides looking for
ite
opportunities to enhance the markets outside India, meeting exporters, attending meetings
with similar international bodies regarding export of Indian spices. But, the call from the
Ministry in the morning looked like a sign of an impending crisis situation. He could smell it
im
from his long experience. He started thinking logically about the stakes and the possible
fallouts of the incident.
The fax came before the meeting. Three Mumbai based exporters had been identified as
rL
the suppliers of the contaminated chilli powder in the RASFF (Exhibit 2). During the
meeting, Mr. Jose summed up the following issues which had to be addressed by SBI as the
licensing authority of exporters of Indian spices.
Fo
(i) Assessing the damage done by the event and formulating strategies to control it.
(ii) Defending the brand image of Indian spices in the international market through
confidence building exercise.
(iii) Formulating strategies to avoid future recurrence of such events.
At the end of the meeting, Mr. Jose requested Mr. Kannan to verify the EU claim and
collect back-up information in the next few days and report to him. He also requested Mr.
KRK Menon, Senior Scientist, in charge of the Quality Evaluation Laboratory (QEL), to
report on the quality parameters and the tests related to chilli powder that were being
presently carried out in the SBI lab.
Mr. Kannan set about his task immediately towards the above. Examining past
correspondence and records, he gathered the following:
y
1. Sudan I is a red dye that is used for colouring solvents, oils, waxes, petrol, and shoe and
nl
floor polishes. Under the colours in Food Regulations 1995 of EU, the red dye is illegal,
and is considered to be a genotoxic carcinogen. Its presence, at any level, is not
O
permitted in foodstuffs for any purpose. The ban came following the experimental
results on rats, which suggested that the chemical could trigger the formation of
malignant tumours. The details about its health risks (Exhibit 3) mention a significant
n
fact that “the risk of cancer in humans from Sudan I has not been proven and any risk
from these foods is likely to be very small indeed.”
io
2. As per the RASFF (Exhibit 2) dated May 9, 2003, issued by the EU, consignments of
at
chilli powder exported by the following firms to East West Spices, UK have been
detained for the presence of Sudan I. All are based in Mumbai. The names are:
(i) Gautam Export Corporation
Flat 11-B, 3rd Floor
ul
irc
Koolbreeze CHS Limited, Plot No. K-72
17th Road, Khar (West)
Mumbai-400 052
C
(ii) Patons Exports Private Limited

10, Koolbreeze CHS Limited,
d
17th Road, Khar (West)

ite
Mumbai-400 052
East West Spices, UK (the importers) provided the details of how they traced back the
im
material to the exporters. In order to verify the correctness of the EU claims that the
material actually went from the two named exporters (Gautam Export Corporation and
Patons Exports Private Limited) to East West Spices, UK, an investigation was
rL
conducted by SBI. The investigation showed that the shipping marks and lot numbers
of the material sampled corresponded to those used by Volga Spices and Masala Mills
Private Limited, at the behest of the named exporters. Volga Spices, also a registered
Fo
exporter, had processed and supplied the material to Gautam Export Corporation and
Patons Exports Private Limited.
Gautam Export Corporation’s senior partner, Jagdish Advani was a big exporter of
spices to Europe. His company used to trade around 5,400 tons of spice annually to
Europe. Volga Spices was run by the Shaikh family. Imran Shaikh was one of its
directors. Patons Exports Private Limited had directors from the Advani family who
managed Gautam Export Corporation.
A consignment of 5 tons had been shipped from Gautam Export Corporation to East
West Spices (UK), who had in turn exported part of the chilli powder, possibly clubbing
with shipments from the other party to an importer in France where the Sudan I was
detected.
3. Chilli is commercially important for two qualities, its red colour by the pigment
y
Capsanthin and its biting pungency by Capsaicin. The price that chilli powder fetches
nl
in the market is determined by its pungency and colour. Sudan I, a cheap red dye, would
brighten up the colour if mixed with crushed chilli (processed from cheap discoloured
chilli) adequately during its processing. If Sudan I was present in chilli powder it should
O
be the result of a deliberate act of adulteration performed to brighten up its colour, and
hence its appearance. Experiments had shown that for good colour enhancement,
n
couple of kgs of Sudan I would be required per ton of chilli powder. This would cost
about US$ 5 and could potentially increase per ton chilli price from US$ 600 to US$
io
1200.
4. Data related to Indian chilli exports and production (Exhibit 4), major importing
at
countries of Indian chilli (Exhibit 5), and India’s share of chilli imports of some major
importing countries (Exhibit 6) gives a perspective on the export market scenario of
ul
chilli from India.
irc
Mr. Menon, with the help of his subordinates, put together the following information
regarding SBI’s testing equipment and processes.
C
1. The SBI set up the QEL in 1990. The lab was ISO 9002 certified by the British
Standards Institution (BSI) in 1994. The QEL adopted ISO 14001 Environmental
Management System for its activities in the lab. The SBI also had two major concepts
d
for quality improvement. The SBI awarded ISB logo to the consumer packs of spices
ite
selectively to the exporters who had certified processing and quality control capability
and maintained a high level of hygiene and sanitation at all stages. For exports of spices
and spice products in bulk, it awarded the Spice House Certificates to those processors/
im
exporters who had a genuine commitment to sustainable quality and export growth and
made investments in developing in-house processing facilities, infrastructure and had
necessary competence to ensure consistent quality and reliability. These facilities should
rL
cover cleaning, grading, processing, packaging and warehousing.

2. The two important quality characteristics that determine the price of red chilli powder
are its colour (measured in ASTA units) and pungency (Capsaicin percentage). Besides
Fo
that, Aflatoxin (a toxic metabolite produced by certain moulds in food and feed)
content is a critical parameter that should not exceed a threshold level (though different
for different countries) in the exported chilli powder. Several other parameters could be
tested in the SBI lab. (Exhibit 7 provides the analytical services offered by SBI including
the list of tests, charges and the time taken for getting the report.)
3. The lab had five High Performance Liquid Chromatographs (HPLC) machines for
detecting Aflatoxin content. HPLC was used to find the amount of a chemical
compound within a mixture of other chemicals. It could detect up to the level of 5 ppm
(5 gm in a ton). An HPLC machine could also be used for detecting Sudan I. Each
machine could analyse 20 samples a day.
4. Usually, the big exporters had their own testing facilities. Most of the requests for
testing were received from small exporters who did not have their own facilities. On an
y
average, around 30 samples were analysed each day by HPLC machines for various tests
nl
on different spices.
On June 13, 2003, Mr Jose, Mr Kannan and Mr Menon met to discuss the way forward.
O
Mr Jose asked Mr Kannan for his views. Mr Kannan said “I was aware of the unregulated use
of banned red dyes to brighten up the colour of red chilli powder in the domestic market in
order to fetch good prices. But, this is the first time it has happened in the export market.
n
May be, it has happened in the past too, but was not detected. The quantum of Sudan I
added by the alleged exporters at 3,890 ppm was a deliberate act that needed stern action.”
io
Mr. Kannan further added, “the events after 9/11 had changed the world. Since then, the
at
threats of bioterrorism had been lurking in the minds of people in the USA and Europe.
Consequently, the food regulatory authorities and the consumers had become more sensitive
about any kind of adulteration. Checking had become more frequent and thus increased the
ul
chance of detection”.
Mr. Jose mentioned that certain sources in Europe had informed him that EU had been
irc
seriously contemplating about imposition of some kind of restrictions to ensure import of
Sudan I free chilli powder and chilli products to its member countries. It could lead to the
C
requirement of an analytical report from well established labs certifying the export
consignment to be Sudan I free. He continued, “If such a situation actually arises,
✧ How should the SBI act?
d
✧ Should the SBI consider mandatory testing of all the export consignments before
ite
shipping? In that case, the immediate question would be,

✧ Is the SBI (and the QEL) equipped to handle the task efficiently and in a reasonably
short period of time? or
im
✧ Should the SBI license other laboratories which could do the testing and issue the
analytical reports? or
rL
✧ Should the SBI require the exporters to submit analytical reports by equipping their
own quality control lab or through other labs? And also,
✧ How should the SBI monitor the whole process starting from taking the sample to the
Fo
shipment of the consignment?

✧ Should they price the test? If so, what should be the price?”
Mr. Kannan informed that during 2002-03, about 8,000 consignments had been shipped
out by nearly 150 exporters. While the largest exporter did about 4,000 tons, many exporters
did even under 100 tons. Testing of all consignments would increase the work load of the SBI
significantly and may have to be done using various other accredited laboratories in the
country.
Mr Kannan also mentioned, “there were also a few private laboratories used by the
exporters, but their quality and capacity was not up to the mark. Further, the trade felt that
their charges for testing for Sudan I, at over Rs 4,000 per sample, are quite high. It would be
better if SBI could make the testing mandatory and thus take control of the process. The SBI
staff at its offices in the port cities could be given the responsibility of taking samples and
y
sending them to SBI for testing.”
nl
Mr. Jose was very concerned about the fact that EU norms would require no trace of
Sudan I, for which testing precision may have to be at the parts per billion (ppb) level.
O
Responding to Mr. Jose’s concern, Mr. Menon specified that available HPLC machines
were not equipped to test for Sudan I at ppb level. A superior machine called LC MS MS
would be required for this. It would have a capacity of testing 40 samples per day and would
n
cost around US$ 200,000. The variable cost (material and labour for sample preparation) per
test would be about Rs. 700 excluding the collection charges. Supervisory costs would be in
io
addition and would require one qualified technician per shift.
at
Mr. Kannan alerted that even though 8,000 consignments had been shipped over a year,
the shipments are not uniform and are subject to high seasonality. Arrivals on a given day
could be as high as three times the average.
ul
Mr. Jose emphasised, “in case mandatory testing was to be brought in, it had to be done in
a manner that the export time schedules are not affected. Almost all exporters were located at
irc
field office locations. Collection, courier of sample and return courier of the analytical report
could cost about Rs. 300 per sample. While mandatory testing by SBI may not require
C
additional human resources at the field offices, it would definitely be required at the Head
Office, both for sample preparation and supervision.” He was also aware that SBI being a
government organisation, recruitment was banned and hence, not an option. Recruiting
d
students with the appropriate educational background as “trainees” was a possible way out.
ite
Though he had raised various questions, Mr. Jose wondered whether they were over-
reacting. Mr. Kannan responded, “It seems that the alleged exporters get their orders through
overseas middlemen. It is, thus, possible that parts of the same shipment could have gone to
im
other processors/customers and the same could be true with other shipments.”
Mr. Jose knew that a clear message had to be sent to the industry. He also had the inkling
that some of the alleged exporters had political connections. He wanted to make sure that the
rL
entire trade must not suffer due to the greed of a few. Further, in terms of regulatory powers,
SBI could only suspend (pending investigation) and revoke (after investigation) the
exporter’s registration for violation of conditions listed in the registration rules. (Exhibit 8
Fo
gives excerpts from the Spices Board Act, 1986 and the Spices Board Rules, 1987.) However,
violation of food safety norms of the importing country or of the Prevention of Food
Adulteration Act, was not a listed condition for which registration could be suspended or
revoked by the SBI. The regulations also did not empower SBI to draw samples on a
compulsory basis from the export consignment. To impose mandatory testing, the
regulations would need to be amended.
Exhibit 1: Spices Board of India: Functions and Organisation Structure
SBI is governed by a 32 member governing body consisting of representatives of the Lok

Sabha, Rajya Sabha, State Governments, certain Central Government Ministries, Planning
Commission, plantation labour, spice growers, processors and exporters, and specialist
y
organisations such as Central Food Technology Research Institute (CFTRI) and Indian
nl
Institute of Packaging (IIP).
Offices
O
The Head Office of SBI is located at Cochin. SBI has 22 Regional Offices, 13 Zonal Offices
and 35 Field Offices. A central Quality Evaluation Laboratory (QEL) is located at the Head
n
Office. A Biotechnology Lab also functions at the Head Office. Indian Cardamom Research
Institute, a research wing of the Spices Board has its main station at Myladumpara (Idukki,
io
Kerala) with Regional Stations located at Thadiankudissai (Tamil Nadu), Saklespur
(Karnataka) and Gangtok (Sikkim).
at
Main Functions ul
1. Export promotion of all spices through support for:-
irc
a) Technology upgradation
b) Quality upgradation
c) Brand promotion
C
d) Research and product development.

2. Research, development and regulation of domestic marketing of small and large
d
cardamom
ite
3. Research and production development of vanilla

4. Post harvest improvement of all spices
im
5. Promotion of organic production, processing and certification of spices

6. Development of spices in the North East
7. Provision of quality evaluation services
rL
Other Responsibilities Related to Export Promotion of Spices

Fo
1. Registration of exporters
2. Quality certification
3. Quality control
4. Collection and documentation of trade information
5. Provision of inputs to the Central Government on policy matters relating to import and
export of spices.
y
nl
Organisation Structure
O
n
io
at
ul
irc
C
Case 7: Chilli in Soup (A)

d
ite
im
Remark: JT. Dir-Joint Director, Dy. Dir-Deputy Director, Asst. Dir- Assistant Director
rL
(Source: www.indianspices.com)
419
Fo
Regulatory Functions
1. Registration of exporters of spices.
2. Licensing of cardamom auctioneers
3. Licensing of cardamom dealers
y
4. Sampling, testing and analysing of samples of all the spices designated for export out of
nl
the country.
5. Suspending/revoking the certificates of registration of exporters for violation of the
O
conditions of registration relating to food safety.
n
io
Exhibit 2: Rapid Alert System for Food and Feed
at
No. 1/22/2003-EP (Agri-V)
ul
Government of India
irc
Ministry of Commerce and Industry
Department of Commerce
C
Udyog Bhavan, New Delhi

Dated, 10th June, 2003
d
Chairman
ite
Spices Board
Cochin
im
Subject: - Rapid Alert Notification
Sir,
rL
1. Kindly find enclosed Rapid Alert Notifications No 2003/111 dated 9th May, 2003 for
presence of colour Sudan I in hot red chilli powder exported by Gautam Export
Corporation, Mumbai and M/s Patons Exports Private Limited, Mumbai.
Fo
Spices Board is requested to take necessary action in the matter.

Yours faithfully,
(KK Singh)
Section Officer
Encl: A/A
EUROPEAN UNION
DELEGATION OF THE EUROPEAN COMMISSION
TO INDIA, BHUTAN, MALDIVES, NEPAL AND SRI LANKA
Minister-Counsellor, Political Affairs and Coordination
Charge d’Affairs a.i.
y
CK/ck/222
nl
09.06.2003
O
Mr YYY
Additional Secretary
Department of Commerce
n
Ministry of Commerce and Industry
io
Udyog Bhavan
New Delhi-110001
at
Dear Mr YYY,
ul
According to Regulation (EC) No 178/2002 laying down the general principles and
requirements of food law, establishing the European Food Safety Authority and laying down
irc
procedures in matters of food safety, I would like to bring to your attention the following
notification received via the Rapid Alert System for Food and Feed (RASFF):
C
Notification: 2003.111 Date: 09/05/2003

Concerning: hot red chilli powder
d
Despatched from: INDIA

Via UK
ite
Having been the subject of: market control

Reason for notification: colour Sudan I
im
Detailed information is attached.
I would be grateful if you could inform the competent authorities in your country of these
rL
findings in order to help avoiding such problems in the future.

The RASFF contact point (fax number +32.2.222.77.77 or e-mail: [email protected]) is
at your disposal should the authorities of your country have questions about the above
Fo
mentioned information.
With kind regards,
Yours Sincerely
Dr ZZZ, Charge d’Affairs a.i.
EUROPEAN COMMISSION
HEALTH and CONSUMER PROTECTION DIRECTORATE-GENERAL
Directorate D – Food Safety: production and distribution chain
D3 – Chemical and Physical risks; surveillance
Brussels, 9 May 2003
y
nl
RAPID
O
ALERT
n
SYSTEM
io
FOR
at
FOODS
ul
irc
FOOD
C
VERY URGENT – TRES URGENT

d
ALERT NOTIFICATION: 2003/111

ite
ORIGINAL NOTIFIACTION
SUBJECT: COLOUR SUDAN I IN HOT RED CHILLI POWDER
im
FROM UK
rL
PAGE: Cover Pages (1) + 10

Fo
FAXNUMBER: ++32-2-222-77-77
EMAIL: [email protected]
http://forum.europa.eu.int
Market control product recall/Product withdrawal

Product distributed to France
y
Manufacturer: East West Spices (UK)
nl
The contact point from UK is kindly requested to provide the possible distribution
and details about origin of the incriminated ingredient
O
The contact point from France has communicated to the Commission the following
information:
n
RAPID ALERT SYSTEM FOR FOOD AND FEED
io
REGULATION (EC) No: 178/2002 – Art 50/GENERAL INFORMATION
at
1* NOTIFICATION TYP FOOD
2* CONTROL TYPE MARKET
3*
4*
NOTIFYING COUNTRY
DATE OF NOTIFICATION
ul
FRANCE
9.05.2003
irc
HAZARD
C
5* NATURE OF HAZARD UNAUTHORISED COLOUR

6* RESULTS OF THE TESTS SUDAN I 3.89 G/KG
7* COUNTER ANALYSIS
d
8* SAMPLING DATES 30.04.2003

ite
No OF SAMPLES TAKING A SAMPLE (2 PACKAGES OF 30 G EACH)

METHOD
PLACE RETAILER
im
9* LABORATORY
10* METHODS OF ANALYSIS USED GC/MS
11* PERSONS AFFECTED
rL
12* TYPE OF THE ILLNESS
PRODUCT
Fo
13* PRODUCT CATEGORY SPICES

14* PRODUCT NAME HOT RED CHILLI POWDER
15* BRAND / TRADE NAME
IDENTIFICATION OF THE LOT(S)
16* CONSIGNMENT / LOT NUMBER NOT AVAILABLE

17* PUBLIC HEALTH
CERTIFICATE NUMBER
DATE
y
CVED No
18* DURABILITY DATES USE-BY DATE
nl
BEST BEFORE DATE
SELL-BY DATE
O
19* DESCRIPTION OF PRODUCT HOT RED CHILLI POWDER
THE PRODUCT ASPECT
No OF UNITS
n
TOTAL NET WEIGHT
io
ORIGIN
20* COUNTRY OF ORIGIN UK
at
21* MANUFACTURER NAME EAST WEST SPICES, UK
ADDRESS
22* DISPATCHER/
VET AP-No ul
EXPORTER NAME
irc
ADDRESS
DISTRIBUTION
C
23* DISTRIBUTED BY IMPORTER

WHOLESALER
d
RETAILER
24* DISTRIBUTION TO MEMBER STATES
ite
DISTRIBUTION LIST ATTACHED NO

25* EXPORTED TO THIRD COUNTRIES NOT BY THE FIRM
DISTRIBUTION LIST ATTACHED NO
im
IN CASE OF A REJECTION AT THE BORDER

26* POINT OF ENTRY
rL
27* TYPE OF CHECK

28* COUNTRY OF DISPATCH
29* COUNTRY OF DESTINATION
Fo
30* CONSIGNEE NAME

ADDRESS
31* CONTAINER NUMBER
32* MEANS OF TRANSPORT
MEASURES ADOPTED
33* VOLUNTARY MEASURES DESTRUCTION
34* COMPULSORY MEASURES MONITORING OF THE DESTRUCTION
35* JUSTIFICATION INFRINGEMENT OF § 11-LID 1 OF THE FOOD
AND COMMODITY ACT
36* SCOPE NATIONAL
y
37* DATE OF ENTRY INTO FORCE
nl
38* DURATION
O
OTHER INFORMATION
39* MINISTRY
40* PERSON TO CONTACT NAME:
n
TEL:
FAX:
io
E-MAIL:
41* OTHER INFORMATION
at
42* CONFIDENTIAL NO
43* IF YES, WHICH BOXES (NUMBERS)
44* IF YES, REASON ul
Attached were consignment notes identifying the following exporters from India:
irc
1. Gautam Export Corporation
2. M/S Patons Exports Private Limited
C
As sold to East West Spices, UK

(Source: Spices Board of India)
d
ite
Exhibit 3: Health Risks of Sudan I

im
Illegal dye: What is the health risk?

There is concern that the dye, Sudan I, has the potential to cause cancer.
rL
However, the Food Standards Agency has said the risk is very low.
Professor Alan Boobis, an expert in toxicology at Imperial College London says there is
little reason for the public to be alarmed.
Fo
Sudan I was banned from use in food products following experiments on rats, which
suggested that the chemical could trigger the formation of malignant tumours.
However, Professor Boobis told the BBC News website that the levels of the chemical fed
to the rats bore no relation to the kind of levels that people would be exposed to, if they ate
contaminated products.
The rats who showed signs of developing tumours were given a daily dose of Sudan I of
around 30 milligrams/kg of bodyweight for two years.
Animals given a lower dose – 15 milligrams/kg of bodyweight - showed no signs of cancer-
related changes.
The contaminated products would contain Sudan I doses of a much smaller magnitude -
y
micrograms, rather than milligrams.
nl
The animals that did show signs of cancer - known scientifically as tumourigenic changes
- started to do so only after many months.
O
In human terms, this would suggest that if the dye was to have any effect, the symptoms
would not start to become apparent for around 20 years.
n
Genetic damage
io
The compound triggers cancer growth through what is known as a DNA reactive mechanism.
at
Within the body it is converted into a form which attacks the DNA of cells, causing
damage which can be passed on to the next generation of cells in the affected tissue,
ultimately leading to cancer. ul
However, Professor Boobis said that the risk of eating just one or two contaminated items
irc
was trivial.
While not an exact comparison, he likened it to the cancer risk associated with smoking
just one cigarette in a lifetime.
C
Even if a person was to eat a contaminated product every day for several years, the risk of
cancer, although higher, was still likely to be very low.
d
“This compound is not a very potent carcinogen in animals,” he said. “People should not
be unduly concerned about the health effects.”
ite
“It is a good idea to remove this substance from the food chain, but this is being done
simply as a precaution, not because there is an immediate impact on health.”
im
(Source: newsvote.bbc.co.uk)
Federal Institute of Risk Assessment (BfR)

rL
“Dyes Sudan I to IV in food”

BfR Opinion of 19 November 2003
Fo
Sudan I was classified as a category 3 carcinogen and as a category 3 mutagen (Annex I of the
Directive 67/548/EEC) according to EU criteria. Substances in category 3 give rise to
concern because of a possible carcinogenic effect in man but cannot be definitively assessed
because of the lack of information.
For Sudan I, the possible intake amounts can be compared in the following way with the
dose which led in animal experiments to neoplastic liver nodules.
In the case of an assumed high dye level (3500 mg/kg) and a large consumption amount of
chilli powder (up to 500 mg/day), 1750 µg Sudan I can be taken in per day in the worst-case
scenario. This corresponds to 29.2 µg/kg bodyweight (at a bodyweight of 60 kg). This
amount is below the dose of 30 mg/kg bodyweight by a factor of 1 × 103 at which a
statistically significant increase in the incidence of neoplastic liver nodules (NTP, 1982) was
y
observed in animal experiments in rats after chronic administration with Sudan I in feed. In
other words, the difference between the amount of Sudan I which can be taken in based on
nl
these assumptions in the worst case scenario per day (29.2 µg/kg bodyweight) and the
amount at which a statistically significant increase in the incidence of neoplastic liver nodules
O
was observed in animal experiments (30 mg/kg bodyweight) amounts to three orders of
magnitude.
n
Assuming a lower dye level (e.g. 10 mg/kg) and a large consumption amount of chilli
powder (up to 500 mg/day), 5 µg Sudan I can be taken in per day in the worst case scenario.
io
This corresponds to 0.083 µg/kg bodyweight (at a bodyweight of 60 kg). This amount is
below the dose of 30 mg/kg bodyweight by a factor of 3.6 × 105 at which a statistically
at
significant increase in the incidence of neoplastic liver nodules (NTP, 1982) was observed in
animal experiments in rats after chronic administration of Sudan I in the feed. In other
ul
words, the difference between the amount of Sudan I which can be taken in based on these
assumptions in the worst case scenario per day (0.083 µg/kg bodyweight) and the amount at
irc
which a statistically significant increase in the incidence of neoplastic liver nodules was
observed in animal experiments (30 mg/kg bodyweight) amounts to six orders of magnitude.
C
Conclusions
From the above assessments, the conclusion can be drawn that in the case of one-off or
d
occasional consumption of foods which are contaminated with Sudan dyes in concentrations
of a few milligrams per kilogram, the risk of cancer is probably very low. This is because the
ite
difference between the dye or amine amount, which can be taken in per day in conjunction
with a high consumed amount in the worst-case scenario and the dose at which carcinogenic
effects were observed in animal experiments is roughly five to seven orders of magnitude.
im
However, this can no longer be assumed for concentrations of several thousand milligrams
per kilogram because the difference between the dye or amine amount, which can be taken in
rL
per day in conjunction with a high consumed amount in the worst-case scenario and the dose
at which carcinogenic effects were observed in animal experiments, is now only two to three
orders of magnitude.
Fo
The estimation that in the case of one-off or occasional consumption of only a few foods
contaminated with Sudan dyes, the risk of cancer is probably very low but does not mean
that there is no risk at all. For carcinogenic substances in category 2, whose action constitutes
a clear cancer risk for human beings according to the current state of knowledge, no
concentration can be given which could still be considered to be safe (DFG, 2003).
Furthermore, the risk of course increases in the case of frequent or ongoing consumption. For
precautionary reasons, the intake of substances of this kind should, therefore, be kept as low
as possible and any exposure, if at all possible, avoided. This applies even more since intake of
carcinogenic amines may also result from other applications like hair dyes (Platzek et al,
1999; SCCNFP, 2002) and the action of several different carcinogenic amines can lead to
additive effects.
Furthermore, it should be borne in mind that risks which may be tolerable in the safety at
y
work range must be assessed differently in the food sector if these are avoidable risks as they
nl
are in this case.
(Source: www.bfr.bund.de)
O
n
Exhibit 4: Indian Chilli Exports and Production
io
Year Exports Production Annual Annual
at
Average Spot Average
Prices of Prices of Export
ul Indian Chilli in Chilli Share of
Value Qty Qty New York in India Production
(Rs m) (‘000 t) (‘000 t) (US$/t) (Rs/t) % (of t)
irc
199899 2,529 68 1,043 1698 4134 6.5
199900 2,547 64 1,056 1587 3300 6.0
200001 2,297 62 1,046 1345 2393 6.0
C
200102 2,524 70 1,113 1301 2660 6.3

200203 3,151 81 846 800 3358 9.6
d

ite
im
rL
Fo
Exhibit 5: Major Importing Countries of Indian Chilli
y
nl
Country 1998–99 1999–00 2000–01 2001-02 2002-03 Five Year
Value Qty Value Qty Value Qty Value Qty Value Qty Average
Qty
O
(Rs m) (‘000 t) (Rs m) (‘000 t) (Rs m) (‘000 t) (Rs m) (‘000 t) (Rs m) (‘000 t) (‘000 t)
Sri Lanka 730 20 599 18 618 21 700 25 785 22 21
n
USA 441 7 649 12 682 13 757 15 868 17 13
Bangladesh 445 11 183 4 33 1 86 3 460 16 7
io
EU 1 147 3 205 4 243 5 232 4 247 5 4
Malaysia 61 1 213 5 74 2 107 3 150 4 3
at
Pakistan 320 15 199 8 87 3 68 3 17 1 6
Total of Above 2144 57 2048 51 1737 45 1950 53 2527 65 54
ul
Overall Total 2529 68 2547 64 2297 62 2524 70 3151 81 69
% Share of Above countries 85 84 80 80 76 73 77 76 80 80 79
irc
1.The top ten importing countries of EU, which accounted for almost the entire EU imports are elaborated below.
C
Country 1998–99 1999–00 2000–01 2001-02 2002-03 Five Year
Case 7: Chilli in Soup (A)

Value Qty Value Qty Value Qty Value Qty Value Qty Average
Qty
d (Rs m) (‘000 t) (Rs m) (‘000 t) (Rs m) (‘000 t) (Rs m) (‘000 t) (Rs m) (‘000 t) (‘000 t)
UK 80.0 1.40 87.0 1.42 92.8 1.67 104.7 1.92 98.0 1.80 1.64
ite
Netherlands 16.0 0.30 29.0 0.57 47.1 1.08 33.1 0.75 39.0 0.82 0.70
Italy 22.0 0.39 43.0 0.75 41.3 0.79 36.1 0.62 41.0 0.69 0.65
France 8.5 0.20 14.0 0.30 14.9 0.37 16.0 0.38 13.0 0.29 0.31
im
Germany 3.5 0.09 11.0 0.34 10.0 0.19 9.5 0.22 11.0 0.24 0.21
Spain 3.5 0.06 4.6 0.07 17.6 0.27 17.5 0.26 20.0 0.34 0.20
Poland 2.8 0.10 2.6 0.11 5.6 0.21 2.4 0.09 4.7 0.16 0.13
rL
Greece 3.7 0.08 12.0 0.24 5.8 0.13 5.0 0.11 5.1 0.10 0.13
Belgium 4.1 0.10 2.4 0.05 3.5 0.06 2.2 0.04 6.9 0.11 0.07
Sweden 3.2 0.02 0.1 0.00 4.2 0.06 5.0 0.05 8.7 0.08 0.04
429
Fo
EU Total 147 3 206 4 243 5 232 4 247 5 4

y
430
nl
Exhibit 6: India’s Share of Chilli Imports of Some Major Importing Countries
O
Country 1998–99 1999–00 2000–01 2001-02 2002-03

n
Value Qty Value Qty Value Qty Value Qty Value Qty
(US$ m) (‘000 t) (US$ m) (‘000 t) (US$ m) (‘000 t) (US$ m) (‘000 t) (US$ m) (‘000 t)
io
USA (Total) 89 54 90 57 96 59 99 66 125 85
USA (India) 13 9 19 12 18 12 18 14 21 17
at
Indias Share (%) 14.9 15.9 21.0 21.0 18.7 21.1 18.1 20.6 16.4 19.6
ul
EU (Total) 72 75 96
EU (India) 5.7 5.9 6.1
irc
Indias Share (%) 7.9 7.9 6.4
Malaysia (Total) 22 23 35 29 32 29 30 33 35 40
C
Malaysia (India) 0.2 0.4 4.8 5.0 4.5 5.0 2.0 2.7 4.8 6.5
Indias Share (%)
d 0.8 1.5 13.7 17.4 14.3 17.5 6.8 8.3 13.8 16.1
UK (Total) 16 5.3 12 4.7 12 5.1 14 7.5 14 6.2

UK (India) 2.1 1.4 2.5 1.7 2.4 1.7 2.4 1.9 2.6 2.2
ite
Indias Share (%) 13.3 25.8 21.3 36.7 20.8 33.8 17.5 25.2 18.9 35
Sri Lanka (Total) 18 20 17 18 25

im
Sri Lanka (India) 18 20 17 17 24

Indias Share (%) 96.3 96.1 99.5 96.2 96.6
(Source: unstats.un.org)
rL
Fo
Exhibit 7: Analytical Services of Spices Board of India
The QEL offers analytical services to the exporters, traders, farmers and research organisations for
analysis of spices and spice products for various parameters listed below.
y
Sl No Name of Minimum Quantity Time
Analysis Required for Analysis Charges Required
nl
(gram) (Rs) (No. of days)
O
1. Agmark specifications 500 * %
2. ASTA specifications 500/250 ´ 10 Nos ** * 1
3. Acid Insoluble ash 250 200 6
n
4. Aflatoxin (B1,B2,G1,G2) (HPLC-METHOD) 250@ 1200 6
5. Alcohol soluble extract 250 100 4
io
6. Bacillus Cereus 100 300 6
7. Bulk density/litre weight of spices 1000 50 1
at
8. Calcium as CaO 250 100 5
9. Capsaicin - HPLC method (%SHU) 100@ 280 4
10. Capsaicin - Hv difference method (%SHU) ul 100@ 200 4
11. Chromate test (Qualitative) 100 80 2
12. Clostridium Perfringens 100 400 6
irc
13. Cold water soluble extract 100 100 4
14. Coliforms 100 180 5
15. Colour value (ASTA METHOD) 100@ 150 4
C
16. Common salt 100 80 5

17. Crocine 10 150 3
18. Crude fibre 100 250 6
d
19. Curcumin (ASTA METHOD) 100@ 210 4

ite
20. Excreta, mamalin 500/250 50 1

21. Excreta, others 500/250 50 1
22. Extraneous / foreign matter 500/250 50 1
im
23. E. Coli 100 150 4

24. Filth, heavy 100 240 4
25. Filth, light 100 240 5
rL
26. Heavy metals (Cd,Cr,Cu,Fe,Mg,Mo,Pb & Zn) 100@ 100& 5

27. Insect defiled / infested 500/250 50 1
28. Light berries 250 70 1
Fo
29. Moisture (ASTA METHOD) 250 70 2

30. Mould (Microbiological) 100 100 5
31. Mould (Physical) 250 50 1
32. Non - volatile ether extract 100 120 6
33. Oleoresin (EDC extractables) 100 120 5
(Continued)
(Continued)
Sl No Name of Minimum Quantity Time

Analysis Required for Analysis Charges Required
(gram) (Rs) (No. of days)
34. Pesticide residues. Organochlorine Pesticides (Isomers of
y
BHC, Endosulfan, DDT, Heptachlor, Aldrin, Dieldrin, Endrin
and Endrin aldehyde) 200@ 1200 6
nl
35. Pesticide residues. Organophosphorous Pesticides
(Chlorpyriphos,Dimethoate, Disulfoton,Ethion, Methyl
Parathion, Phorate, Parathion, and Quinalphos) 200@ 1200 6
O
36. Pesticide residues. Pyrethroid Pesticides
(Cypermethrin and Fenvalerate) 200@ 800 6
37. Pesticide residues. Single compound (any of the above) 200@ 800 6
n
38. Picrocrocine 10 150 3
io
39. Piperine (ASTA METHOD) 100@ 200 4
40. Residual solvent 200 400 3
at
41. Safranal 10 150 3
42. Salmonella 100 350 6
43. Starch ul 100 230 6
44. Staphylococcus aureus 100 350 6
45. Sulphur dioxide 200 180 4
irc
46. Total ash 100 120 4
47. Total plate count 100 120 3
48. USFDA specifications 500/250 ´ 10 nos ** * 6
C
49. Volatile oil 200@ 70 3

50. Vanillin 100 200 3
51. Water soluble ash 100 150 6
d
52. Whole insects, dead (by count) 250 50 1

ite
53. Yeast (Microbiological) 100 100 5

54. Yeast and mould (microbiological) 100 180 5
Authorised by : CMS Issue No. 8 Issue Date 15/05/2001

im
Notes
% Depends on the number of parameters for individual spices and number of samples.
rL
* Charges based on the parameter specified for each spice.

** In the case of analysis for ASTA/USDFA parameters, sample size should be a minimum of 500 g for heavier items like pepper, ginger,
turmeric etc and 250 g sample for low density items like chilli and seed spices.
Fo
@ In the case of analysis of oleoresin samples of spices for the above parameters, a minimum of 50g sample size is required and in case of
coriander and herbs, a sample size of 300 g is required.
& Analytical charges Rs 100/- per element.

The samples may be sent to the Senior Scientist (QC), Quality Evaluation Laboratory, Spices Board, Sugandha Bhavan, Palarivattom,
Cochin-682 025, along with analytical charges in the form of cash/DD drawn in favour of Secretary, Spices Board.
(Source: www.vigyanprasar.com)
Exhibit 8: Excerpts from the Spices Board Act, 1986 and

the Spices Board Rules, 1987
Excerpts from the Act
y
nl
Chapter II: THE SPICES BOARD
O
7. Functions of the board
(1) The board may—
n
(i) develop, promote and regulate export of spices;
(ii) grant certificate for export of spices and register brokers therefor;
io
(iii) undertake programmes and projects for promotion of export of spices;
at
(iv) assist and encourage studies and research for improvement of processing, quality,
techniques of grading and packaging of spices;
ul
(v) strive towards stabilisation of prices of spices for export;
(vi) evolve suitable quality standards and introduce certification of quality through
irc
“Quality Marking” for spices for export;
(vii) control quality of spices for export;
C
(viii) give licences, subject to such terms and conditions as may be prescribed, to the
manufacturers of spices for export;
d
(ix) market any spice, if it considers necessary, in the interest of promotion of export;
(x) provide warehousing facilities abroad for spices;
ite
(xi) collect statistics with regard to spices for compilation and publication;
(xii) import, with the previous approval of the Central Government, any spice for sale;
im
and
(xiii) advise the Central Government on matters relating to import and export of spices.
rL
(2) The board may also—

(i) promote co-operative efforts among growers of cardamom;
(ii) ensure remunerative returns to growers of cardamom;
Fo
(iii) provide financial or other assistance for improved methods of cultivation and
processing of cardamom, for replanting cardamom and for extension of cardamom
growing areas;
(iv) regulate the sale of cardamom and stabilisation of prices of cardamom;
(v) provide training in cardamom testing and fixing grade standards of cardamom;
(vi) increase the consumption of cardamom and carry on propaganda for that purpose;
(vii) register and licence brokers (including auctioneers) of cardamom and persons
engaged in the business of cardamom;
(viii) improve the marketing of cardamom;
(ix) collect statistics from growers, dealers and such other persons as may be prescribed
on any matter relating to the cardamom industry; publish statistics so collected or
y
portions thereof or extracts therefrom;
nl
(x) secure better working conditions and the provision and improvement of amenities
and incentives for workers; and
O
(xi) undertake, assist or encourage scientific, technological and economic research.
Chapter IV: CERTIFICATE FOR EXPORT OF SPICES
n
io
11. No person to export spices without certificate
Save as otherwise provided in this act, no person shall, after the commencement of this act,
at
commence or carry on the business of export of any spice except under and in accordance
with a certificate.
ul
Provided that a person carrying on the business of export of spices immediately before the
commencement of this act, may continue to do so for a period of three months from such
irc
commencement; and if he has made an application for such certificate within the said period
of three months till the disposal of such application.
Explanation- The reference in this section to the commencement of this act shall be
C
construed in relation to any spice added to the Schedule by notification under the provision
to clause (n) of section 2 as reference to the date with effect from which such spice is added to
d
the Schedule.
ite
12. Grant of certificate

(1) An application for grant of certificate shall be made to the board in such form and shall
im
contain such particulars as may be prescribed and shall be accompanied by a receipt

evidencing the payment of the prescribed fee.
(2) On receipt of such application, the board shall,
rL
(a) if the application is not in the prescribed form, or does not contain any of the
prescribed particulars, return the application to the applicant; or
Fo
(b) if the application is in the prescribed form and contains the prescribed particulars,
grant the certificate subject to such terms and conditions as may be determined by
regulations.
13. Cancellation, suspension, etc. of certificate

(1) The board may cancel any certificate on any one or more of the following grounds,
namely:
(a) that the holder of the certificate has violated any of the terms and conditions of the
certificate; and
(b) that in the opinion of the Central Government it is necessary in the interests of
general public to cancel the certificate.
(2) Where the board, for reasons to be recorded in writing, is satisfied that pending
y
consideration of the question of canceling the certificate on any grounds mentioned in
nl
sub-section (1), it is necessary so to do, the board may, by order in writing, suspend the
operation of the certificate for such period not exceeding forty-five days as may be
O
specified in the order and require the holder of the certificate to show cause, within
fifteen days from the date of receipt of such an order, as to why the suspension of the
certificate should not be extended till the determination of the question as to whether
n
the registration should be cancelled.
io
(3) No order of cancellation of registration under this section shall be made unless the
person concerned has been given a reasonable opportunity of being heard in respect of
at
the grounds for such cancellation.
14. Appeal ul
(1) Any person aggrieved by an order made under section 13 may prefer an appeal to the
irc
Central Government within such period as may be prescribed.
(2) No appeal shall be admitted if it is preferred after the expiry of the period prescribed
therefore:
C
Provided that an appeal may be admitted after the expiry of the period prescribed
therefore, if the appellant satisfies the Central Government that he had sufficient cause
d
for not preferring the appeal within the prescribed period.

(3) Every appeal made under this section shall be made in such form and shall be
ite
accompanied by a copy of order appealed against and by such fees as may be prescribed.
(4) The procedure for disposing of an appeal shall be such as may be prescribed:
im
Provided that before disposing of an appeal, the appellant shall be given a reasonable
opportunity of being heard.
rL
(5) The Central Government may confirm, modify or reverse the order appealed against.
15. Power to permit export without certificate

Fo
The Central Government may, if satisfied that it is necessary or expedient, so to do, in

public interest, by notification in the Official Gazette and subject to such conditions, if
any, as may be specified therein, permit any body or other agency to commence or carry
on the business of export of spices without a certificate.
(Source: www.legalpundits.com)
Spice House Certificate

The certificate is issued to those processors/exporters who have a genuine commitment to
quality, and whose long-term objective is sustained export growth. The Spice House
Certificate seeks to identify and recognise processors who have made investments in in-house
processing facilities and infrastructure, and have the necessary competence to ensure
y
consistent quality and reliability. These facilities cover all critical areas—cleaning, grading,
nl
processing, packaging and warehousing. A foolproof system of quality assurance should be
employed at all stages of processing - from raw material selection to final shipping. The
O
processors are also expected to maintain a high degree of sanitation in the plant, while the
workers must observe absolute cleanliness and personnel hygiene.
n
Spice House Certificate holders have quality upgradation as their ultimate objective, but
with a basic difference in focus.
io
The certification programme aims at exporters of spices and spice products in bulk
packing. The products covered by the certification programme include whole spices as well as
at
value-added products like spice mixes, ground spices, curry blends, spice oils oleoresins and
sterilized, dehydrated, pickled and candied spices.
ul
(Source: The Spices Board Act, 1986, www.indianspices.com)
irc
FORM VII
C
(See Rule 15A)

SPICE HOUSE CERTIFICATE
d
SPICES BOARD
ite
(Ministry of Commerce, Government of India, Cochin – 25)
Number……………..
im
M/s………………………………………………exporter of spices having Exporter

Registration No. are hereby granted this Spice House Certificate on the basis of their
rL
facilities for cleaning, processing, grading, warehousing, packaging of spices/spice products.

This Certificate is valid upto the ………………day of …………..
Fo
Place: Cochin Secretary

Date: (Seal)
SCHEDULE
FACILITIES TO BE MAINTAINED BY AN EXPORTER OF SPICES FOR GRANT OF
A SPICE HOUSE CERTIFICATE
Processing Unit:
y
A Cleaning Shall have facilities for washing, removal of extraneous matters, stone, dust manual or
nl
automatic
B Processing Facilities for processing spices, spice mix, spice powders, oils, oleoresins or any
other value added item.
O
Drying Shall have facilities for drying spices. The drying yards shall have cemented floors
without crevices, provided with skirting all around and nets for preventing entry of birds.
Details of other A laboratory capable of analysing raw materials and the finished products.
n
processing technology
and equipments
io
C Grading Shall have facilities for grading spices using sieves mechanically or manually operated
or for grading the spices by sorting machines or by manual means.
at
D Warehousing Shall have store houses/storage area separate for a raw materials and finished
products. Storage premises shall have cemented floors without crevices provided with
ceiling and doors to prevent entry of rodents and birds shall be provided with wooden
ul
pallets/wire mesh and kept clean to avoid entry of rodents, insects, spiders etc.
E Packaging Shall have facilities for packaging spices (manual or automatic)
irc
F Other Facilities The unit shall be provided with washing facilities for hands and feet at the entrance of
the unit. Toilets facilities separate for ladies and gents, washing facilities with soap.
Head gears for workers. The surroundings of the unit shall be maintained free of weeds
and dumped waste materials have facilities for disposal of waste material.
C
Excerpts from the Rules
d
ite
Chapter IV: CERTIFICATE OF REGISTRATION

15. (1) An application for grant of certificate under section 12 shall be made to the
Secretary of the Board in Form I.
im
(2) Every application for grant of certificate of registration shall be accompanied

by a receipt evidencing the payment of fees of Rupees two thousand for a
rL
block period of three years ending on the 31st August or part thereof. Renewal
fee for such period shall be Rupees one thousand.
15 (A) An exporter of spices who has his own or taken on rent or leased premises
Fo
having facilities for cleaning, grading, processing, warehousing and packing as

given in the Schedule annexed to these rules, may apply to the Secretary in
Form IV for grant of a Spice House Certificate. The Secretary, on being
satisfied as to the facilities available in the premises owned or taken on rent or
lease for a period of not less than three years, shall issue a certificate in Form
VII. The certificate shall be valid for the block period of three years for which
the Exporter Registration Certificate is valid and it shall be renewed if he
maintains such facilities. If, on inspection at any time, it is found that the
exporter does not have such facilities, the Spice House Certificate issued to
him shall be cancelled.
16. (1) Any person aggrieved by an order of the Board made under section 13 may,
within sixty days from the date of making of such order, appeal to the Central
y
Government.
nl
(2) Every appeal made under sub-section (1) shall be made in Form II and shall be
accompanied by a copy of order appealed against and a receipt evidencing the
O
payment of Rupees twenty-five.
(3) On receipt of appeal under sub-rule (1), the Central Government shall, after
giving a reasonable opportunity to the appellant of being heard, pass such order
n
as it may deem fit.
io
Source: The Spices Board Rules, 1987, Spices Board, Ministry of Commerce and Industry,
Government of India, Cochin.
at
ul
irc
Putting oneself in the shoes of Mr Jose, the following questions can be raised:
1. What are the consequences that this event could trigger? Could it lead to a crisis?
2. What should be done immediately to control the damage caused by the event?
C
3. What course of actions should be considered to prevent the future recurrence of

such an event?
d
ite
im
As described in the case, the consequences of the event are quite far reaching and potentially
severe, being related to a food item used in other products in turn. The events that could be
rL
impacted by this crisis can be mapped using tools from quality management and process
flow charting.
Both short-term and long-term actions could be considered, keeping in view various
Fo
stakeholders. The options of (a) sampling and testing by different agencies, and (b) pricing of
testing services could impact the effectiveness of the system that is designed. Scheduling of
activities of agencies involved in this process could also affect the viability of the new system.
The facilities for testing need to be planned for the required turnaround times and volumes,
and the required accuracy.
CASE 8
y
nl
O
Bayer CropScience is a world leader in the crop protection business. Significant
challenges exist in ensuring availability of pesticides to farmers spread all over India.
Indian agriculture is predominantly rain fed and crucially depends on the seasonal
n
monsoon. Different pests could attack the standing crop during different phases of the
crop life cycle. Forecasting the type and intensity of pest attacks across geographically
io
dispersed territories is a challenge. The case describes the existing supply chain structure
and focuses on the design of an efficient and effective supply chain that can meet such
at
challenges.
ul
irc
BAYER CROPSCIENCE: Science for a Better Supply Chain
C
September 14, 2005

Jadcherla, Andhra Pradesh
d
Suresh Babu, Territory Manager for Jadcherla, was a worried person. Standing in the
middle of a cotton field, he could clearly see the telltale signs of an extensive bollworm
ite
attack. This was the second year that farmers in Andhra Pradesh had planted BT cotton,
supposedly resistant to bollworm attacks. Acreage under BT cotton had increased
substantially since last year. Owing to premium pricing, a major section of farmers had
im
opted for BT cotton seeds from illegal third-party sources. The quality of these seeds was
always under doubt. Suresh Babu had taken a big risk by giving a sales plan of Spintor that
matched last year’s sales volume. Spintor was highly effective in controlling the bollworm
rL
pest and was a premium brand in the market. The big risk taken seemed to be turning in
his favour a few days ago as bollworm attacks were reported to be at the same level as last
year. Yet, Suresh could not believe his bad luck. A few feet away, the farmer was filling up
Fo
the spraying machine with a mix of water and a cola-based soft drink. A rumour had spread
Written by Saral Mukherjee and G. Raghuram. Cases of the Indian Institute of Management,
Ahmedabad, are prepared as a basis for class discussion. Cases are not designed to present illustrations of
Institute of Management, Ahmedabad.
through the district that cola-based soft drinks were particularly effective in resisting
bollworm attacks. Within a few days, cola-based beverages were selling at a premium in the
market. This meant that Spintor sales would not be anywhere near the quantities that
Suresh had stocked at dealers and distributors.
Kaithal, Haryana
y
nl
Sampat Singh, Regional Manager, had just finished calling the Logistics and Distribution
Department (LDD) of Bayer CropScience at the Mumbai headquarters. He felt a growing
O
anxiety as he recalled the day’s events. Earlier in the day, he had talked to major dealers
offering them a 10 per cent discount if they lifted a specified quantity of Sherpa Alpha.
Sherpa Alpha was a generic product sold in a highly competitive market. By offering this
n
discount, Singh was sure about capturing at least 25 per cent of the market. He had planned
io
his moves well in advance. The discounts were announced only Wednesday morning and he
planned to deliver the quantities by next Monday. This way the competitors would not have
enough time to react even if they came to know of the development through the grapevine.
at
As an experienced sales manager, he knew that speed of execution was a prime factor in the
success of such moves. It was absolutely essential that the shelf space at dealers be blocked
ul
before others could do so. The amount of Sherpa Alpha lying in the godowns was not enough
to meet the huge demand. Realising this, Singh had already been in touch with LDD and
irc
arrangements had been made to divert excess stock from nearby regions. Arrangement had
also been made with the plant in Chandigarh to produce the excess requirement on a war
footing. LDD was reasonably sure that all demand could be met by Monday morning. “All’s
C
well that ends well,” thought Sampat Singh as he uttered a silent prayer.
Background
d
ite
Bayer CropScience Limited was a leading firm in the fields of crop protection, non-
agricultural pest control, seeds, and plant biotechnology. For the year ended December
2004, its annual turnover of Rs. 7,865 million constituted a major portion of the annual
im
turnover of Rs. 19,816 million of the Bayer group in India. (Exhibit 1 provides details about
different Bayer group companies in India.) Bayer CropScience had a market share of 20.2 per
cent in 2004, down from 22.8 per cent in the previous year. Its main competitors were
rL
established players like Syngenta, Dow Chemicals, Du Pont, Monsanto, and Rallis.
The firm converted six zonal sales offices into four zonal profit centres in a recent
restructuring exercise. The objective was to inculcate a spirit of entrepreneurship through a
Fo
decentralised structure, empowerment, and accountability. (The organisation structure is

shown in Exhibit 2.) The general managers at the zonal level enjoyed substantial decision
making powers and had complete responsibility over all zonal functions. At the zonal level,
the distribution and finance functions reported to the general manager with dotted
relationship reporting to the head office (Exhibit 3).
Case 8: Buyer Cropscience 441
Pesticide Market
The pesticide market was made up of insecticide, herbicide, and fungicide segments. Sales
in all these segments were highly seasonal (Exhibit 4) and depended crucially on a good
monsoon. A good monsoon resulted in well-nourished crops, which in turn, attracted a
variety of pests. Different pests could attack the same crop during different stages of a crop
y
life cycle (Exhibit 5). Thus different types of pesticides were effective during different stages
nl
of crop growth. Certain pesticides were prescribed as prophylactic applications to ensure that
the crop was resistant to pest attacks. Curative application of pesticides focused on saving the
O
crop once a pest attack had taken place.
Bayer CropScience sold 53 different pesticides in various pack sizes. Most of these
products were established brands having high recall among farmers. (Details about two such
n
products are given in Exhibit 6.) The technical superiority of its products allowed Bayer
io
CropScience to command a premium in the market.
at
Planning Process
The planning cycle in a year was divided into the kharif season (April to November) and the
ul
rabi season (December to March). The planning for the kharif season was done in April when
sales managers forecasted the demand for a particular product. This was arrived at by first
irc
forecasting the acreage under cultivation for different crops. Then, the percentage of total
acreage to be covered by a particular product was determined. The forecasted sales of the
product was arrived at by considering several factors like recommended dosage, number of
C
sprayings, etc. Sales targets could be modified to reflect region-specific targets for market
share, stock availability, and strategic considerations. Bayer CropScience divided the 53
d
products into four segments according to their gross margins (see Exhibit 7). The sales plan
reflected the increasing focus on A and B categories.
ite
Once finalised, the sales plan was converted to a monthly sales plan using the liquidation
trend of sales of the previous year and uploaded in the SAP system. At any point in time,
im
rolling sales forecasts (RFCs) were available in the SAP system for m (current month), (m +
1), (m + 2) and (m + 3) months. RFCs could be modified by generating a mid RFC around
the fifteenth of a month. Additional RFCs could be generated according to the need (Exhibit
rL
8). The SAP system determined the master production schedule (MPS), which was further
fine-tuned following discussions with distribution. Once firmed up, MRP was run to
determine the production plan for different production facilities and purchase requirements.
Fo
(Different monthly planning activities undertaken by the Industrial Operations Planning

(IOP) Department are given in Exhibit 9.)
Production
The production process involved two distinct stages. In the first stage, technicals (active
ingredients) were produced in batches in large integrated plants at Thane in Maharashtra and
Ankleshwar in Gujarat. The Ankleshwar plant was the sole production hub for worldwide
sales of several technicals. Certain technicals were sourced from the associate firm Bilag
Industries and some were imported. Technicals were stored in bulk in warehouses
constructed following strict norms for storage of hazardous materials.
In the second stage, the end products were formulated by dissolving the active ingredients,
solvents, and emulsifiers, and packaged in different sizes. Almost 40 per cent of the volume
y
was formulated at Ankleshwar. Another 40 per cent was subcontracted to a firm at Kharar in
nl
Chandigarh. The rest was formulated at Himmatnagar in Gujarat and other subcontracting
locations, referred to as tolling locations. In January 2004, the state of Jammu and Kashmir
announced its industrial policy which offered excise duty waiver (currently 16 per cent) and
O
a tax holiday of 10 years for new industrial units. Bayer CropScience was considering shifting
a major portion of its formulation activities from different locations to Jammu to avail the
n
benefits of the tax cut.
There were four production units at Ankleshwar (see Exhibit 10). The EC unit contained
io
five vessels for dilution of technicals, one having 5 kilolitre (kl) capacity, three having 6 kl
capacity, and one having 12 kl capacity. One vessel having 6 kl capacity was dedicated to
at
herbicide production. Hence, at any point, a maximum of five different products could be
formulated. After formulation, the material from any vessel could be directed to one
ul
of seven filling stations. The filling stations were dedicated according to pack sizes (see
Exhibit 11).
irc
Depending on vessel capacity, a minimum batch size of 2.5—5 kl was required for
effective stirring. A smaller batch was considered uneconomical as the cost of running the
motor used for stirring was invariant of the actual amount inside the vessel. Changeover from
C
one product to another required flushing the entire system with a solvent. The solvent could
be reused either in a future flushing or as an ingredient to the formulation process. Strict
d
norms were enforced for checking the quality of flushing. A presence of no more than 800
parts per million of an active ingredient was essential for changing over from one insecticide
ite
to another or from an insecticide to an herbicide. For changing from herbicide to insecticide,

the norm was zero parts per million. Changeovers took two shifts on average. The
formulation units worked on a single shift basis in general and three shift basis during peak
im
season (April to September).

The tolling unit at Kharar was operated by Punjab Pesticides Industrial Cooperative
rL
Society Limited (PPIC) and had 14 formulation lines. It supplied formulated products to
several manufacturers apart from Bayer CropScience. Several formulation lines were
dedicated to Bayer CropScience products during peak season. Working on a single shift basis
Fo
it could formulate and pack 150 – 200 kl in a month. The plant could produce a maximum
of 750 kl in a month working on a two-shift basis.
Distribution
Bayer CropScience did not have retail presence; instead relied solely on a network of dealers,
distributors, and preferred dealers to service the demand. The sales organisation was made up
of four zones, 25 regions and 169 territories. The South Zone, headquartered in Hyderabad,
was by far the biggest (Exhibit 12) and looked after the sales in Andhra Pradesh, Karnataka,
Tamil Nadu, and Kerala. Andhra Pradesh accounted for almost 25 per cent of national sales.
SKUs were transported from any one of the 13 manufacturing/tolling locations to 25
warehouses of which 17 were managed by carrying and forwarding agents (C&FA). (Some of
the prominent manufacturing locations and headquarters are shown in Exhibit 13.) South
y
Zone warehouses were located in Hyderabad, Guntur, and Kurnool in Andhra Pradesh,
Hubli and Bangalore in Karnataka, Trichy and Coimbatore in Tamil Nadu, and Kottayam in
nl
Kerala. Seventy-five per cent of transportation from the manufacturing location to a
warehouse used single source-single destination-multi product-full truckloads. The rest
O
involved movement from a single source to multiple destinations on a full truckload basis.
From the warehouses, 80 per cent of the material was moved to the distributors in full
truckload milk runs. The rest was sent as parcels to a single destination. Distributors
n
managed the movement of material to dealers using auto-rickshaws and public transport. In
io
Andhra Pradesh, for example, the dealer telephoned the distributor and placed his order. The
orders received during the day were dispatched between 9.30 pm and 3.30 am. A preferred
at
dealer (PD) was appointed in certain satellite markets. A PD was primarily a dealer who was
serviced directly from the warehouses. PD could also act as a distributor and sell to other
small dealers and big farmers who would in turn supply to co-farmers. In all, Bayer
ul
CropScience sold through a network of 2,500 distributors and 35,000 dealers.
irc
Trade Promotions
Individual farmers bought their pesticide requirements from dealers located in nearby
C
markets. Such dealers generally stocked other agricultural inputs like seeds and fertilizers
besides pesticides. Trade promotions constituted an important marketing tool in stimulating
d
demand in the market. The promotional schemes were territory-specific and could be
categorised into two distinct classes. Liquidation discounts involved SKU-specific discounts
ite
offered to liquidate stocks with dealers. Such discounts were floated during the season and
were intended to create a “hungama” (excitement) in the market. In contrast, placement
discounts were based on the quantity of pesticide stock that a dealer ordered before the start
im
of the season. For example, a dealer may be offered a cash discount of 10 per cent on demand
draft and 9.5 per cent on cheques in January for stocks that were likely to be sold during the
kharif season. The cash discount would reduce by 1.5 per cent each month. The discount
rL
could also be in the form of promotional schemes like gold coins. Such promotional schemes
were commonly employed by almost all major players in the pesticide market.
The main objective of trade promotions was to garner shelf space. An effective way to
Fo
“book” shelf space was to induce the dealer to park a part of his working capital with Bayer
CropScience. A dealer could transfer a lumpsum amount to Bayer CropScience in January
without specifying the SKU-wise requirement. The actual order was then placed at the
beginning of the season in April and stocks delivered subsequently. The attractiveness of a
promotional scheme depended on whether the dealer paid cash or took material on credit. In
2005, dealers in Punjab and Haryana had surplus cash and could afford to commit funds well
in advance. In contrast, Andhra Pradesh was primarily a credit market.
Inventory Management
Availability of the right pesticide in the right market at the right time was key to profitability.
The nature and quantum of pest attack was difficult to predict. The extent of a pest attack
could be significantly affected by temperature and rainfall. Indian agriculture relied heavily
on monsoon rains. Monsoon activity was difficult to predict in aggregate, more so at local
y
level. (The sales forecast accuracy for the Jadcherla territory for 2004 is given in Exhibit 14.)
nl
The inherent uncertainty meant that stock built up at a particular location in anticipation of
a pest attack could suddenly become redundant. Stocks returned from distributors to
O
warehouses were termed sales returns. Sales returns averaged 3.4 per cent in 2004, down
from 6-7 per cent in earlier years. Such stocks could then be moved to another location and
were classified as intra-zonal or inter-zonal transfers (Exhibit 12) depending on the
n
destination. The company did not separately track the intra and inter-zonal transfers arising
io
from a sales return from those arising out of other reasons. Stocks returned by a zone from its
warehouses to manufacturing units constituted branch returns. These stocks were either near
at
expiry or expired stocks. Sales returns exhibited significant variability from one location to
another and also from one season to another.
ul
Bayer CropScience tracked the inventory lying in different manufacturing and warehouse
locations using the SAP platform on a real-time basis. However, stock level information was
available only till the distributor level. This information was used in deciding to call back
irc
stocks from one location for redeployment in another location. Redeployment entailed
certain risks. In one example, the company decided to redeploy stocks following less than
C
adequate rainfall in a particular territory. Five days after redeploying Rs. 41 million of stocks,
substantial rainfall changed the odds of a pest attack. A decision was taken the day after the
rains to rebuild the inventory in that territory. Of the Rs 47 million worth of stocks rebuilt,
d
only Rs. 10 million was the same product as earlier stocked. Redeployment of stocks was
possible to the extent that the crop sowing date differed from one territory to another.
ite
Redeployments were mostly to nearby territories and within a zone. Inter-zonal

redeployments were rare.
im
Most pesticides had a shelf life of two years. The company tracked the expiry date of
stocks located in different locations and classified stock as “near expiry” if the expiry date was
less than three months away. After a season was over, dealers showed an aversion to keeping
rL
unsold stocks even if the stocks were not classified as near expiry. Unsold stocks were
returned by dealers and could only be repositioned at the beginning of the next season, that
too after offering a discount. Old stocks could be reprocessed any time before the expiry date.
Fo
Reprocessing allowed the shelf life to be extended by two more years and entailed a
reprocessing cost amounting to 28 per cent of the original manufacturing cost. The company
employed first-in-first-out (FIFO) rule in all inventory related decisions. In a recent
initiative, the zones were being asked to return all unsold stocks after a season for
reprocessing. All expired stocks had to be incinerated. In 2004, the company had to write off
Rs 85 million of stocks. A further Rs 70 million constituted material loss during
reprocessing. It did not track the revenue loss arising from non-availability of stocks.
A further complication involved availability of different pack sizes for a particular

product. On an average, each of the 53 products had four pack sizes, resulting in more than
200 SKUs. The pesticide price per ml depended on the pack size and could vary considerably
(Exhibit 6). The IOP Department tried to schedule the production of bigger pack sizes
during the season and smaller pack sizes during the off season. This was done since, for a
y
specific production quantity, the filling time for smaller pack sizes were more than that for a
bigger pack size.
nl
Inventory planning was based on determination of the stock coverage representing the
ratio of stocks to sales. The coverage ratio was tracked on a zonal and national basis and could
O
be tracked on an aggregate basis and also on a specific SKU basis. A target inventory norm of
1.5 months coverage was considered ideal. In practice, the stock coverage varied from 1.5 to
n
3 months and could be set depending on the product and market conditions.
io
Science for a Better Supply Chain
at
September 14, 2005
Mumbai, Maharashtra
ul
Susan D’Costa, planning executive at LDD, was updating a report on the forecast accuracy
for presentation to top management. Forecast accuracy had been hotly debated in the
irc
headquarters. Earlier in the day, the report seemed to be almost complete when she received
data regarding the sudden demand for Sherpa Alpha which was well above the sales forecast.
More disturbing was the news about Spintor that came from the Hyderabad office. For a long
C
time, D’Costa had been emphatically stating, “We need a better forecast”. She felt time had
come to reinforce the need for a better forecast.
d
As she finished updating the forecast data, D’Costa wondered whether there were
alternative forecasting techniques for the pesticide market. Could a better forecast be a
ite
competitive tool? Could Bayer CropScience align its supply chain to be responsive to market
requirements? Such a move could actually reduce the reliance on a sales forecast. But it may
need substantial changes in the manufacturing and logistics processes apart from a change in
im
the organisational mindset. Similarly, promotion schemes needed to be aligned with the
supply chain strategy. On her desk, was a company brochure prominently displaying the
rL
mission statement adopted by Bayer worldwide: “Bayer: Science for a better life.” Perhaps
the need of the hour in the Indian pesticide market was a science for a better supply chain.
Fo
Exhibit 1: Bayer Group of Companies in India
Bayer CropScience Limited, erstwhile Bayer (India) Limited, constituted the core
cropscience company and had production facilities at Thane, Himmatnagar, and
Ankleshwar.
y
Bayer Diagnostics India Limited, with headquarters and production facilities in Vadodara,
nl
had been active in India since 1976 and had a leading position in the market for medical
diagnostic devices.
O
Bayer Pharmaceuticals Private Limited was established in July 2000 and was a wholly
owned pharmaceutical subsidiary of Bayer Industries Limited. The company handled
pharmaceutical sales and marketing operations.
n
Bayer Material Science Private Limited was one of the largest producers of polymers and
io
high-performance plastics. The company had a plant in Cuddalore, near Chennai, which
manufactured thermoplastic polyurethanes.
at
Proagro Seed Company Private Limited, headquartered in Hyderabad, was involved in
ul
activities in bioscience to develop hybrid seeds for crops catering to specific farmers’ needs
across India. The company, incorporated in 1977, had emerged as a national player having
the advantage of technology transfer, excellent R&D facilities, and presence in hybrid as well
irc
as field crops.
Bilag Industries Private Limited was one of the largest manufacturers of synthetic
C
pyrethroids in the world.

Bayer Polychem (India) Limited, was a 100 per cent subsidiary of Bayer CropScience
d
Limited, and handled non-cropscience businesses which were earlier part of Bayer (India)
Limited.
ite
im
rL
Fo
y
nl
Exhibit 2: Organization Structure of Bayer CropScience
O
Governmental Affairs/
Managing Director / Sustainable
= Executive
n
Country Speaker Development
Committee member
io
at
Chief Operating IOP / Tolling / Procurement / Logistics /
Chief Financial Officer
Officer Prod. Planning Distribution
ul
Procurement Human Resources /
Auditing North Zone Ankleshwar
irc
RM/PM/T&S Payroll
Corp.
Legal / C.S. West Zone Thane Dist. North
Case 8: Buyer Cropscience

C
Communication
Site Services /
Finance East Zone Himatnagar Dist. South Ind. Relat. /
d
Health
ite
Formulation Engineering &
Information Mgt. South Zone Dist. West
/Tolling Energies
im
Marketing Safety / Environ.

Controlling Dist. East
& Portfolio Mgmt. & Quality
rL
Development /
EART
Regulatory Affairs
447
Fo
y
448
nl
Exhibit 3: Zonal Organization Structure of Bayer CropScience
O
Chief Operating

Officer
n
io
General
Manager ZonalSupport
at
Staff
ul
Finance and Zonal Marketing Regional Manager Zonal Zonal
Development
irc
Accounts Manager Region 1 Distribution
Field
Manager Field Marketing Regional Manager Regional
Development
C
Reporting Manager Region 2 Distribution
Manager
Manager Field Marketing Regional Manager

d
Regional
Accounting Manager Region 3 Distribution
ite
Manager
Manager Regional Manager
Credit Control Region 4
Administration
im
Manager
Manager
Accounting, Leagal
and Banking
rL
Fo
Exhibit 4: Seasonality
y
nl
O
n
io
at
ul
irc
C
Exhibit 5: Pest Attacks on Cotton

d
Days Pest Application Bayer Product

ite
0 30 Sucking Pest Prophylactic (50%) Confidor

30 45 Sucking Pest Prophylactic (50%) Confidor-Metasystox
45 60 Bollworm Prophylactic (20%) Thiodan-Hostathion-Sherpa
im
alpha-Bilcyp-Decis-Cybil
60 75 Sucking Pest + Bollworm Thiodan-Hostathion-Spark
Decis-2.8ec-Larvin-Spintor
Decis100-Confidor-Metasystox
rL
75 90 Mixed infestation Curative (90%) Thiodan-Hostathion-Spark

Decis-2.8ec-Larvin-Spintor
Decis100-Confidor-Metasystox
Fo
90 120 Mixed infestation Curative (90%) Thiodan-Hostathion

Spark-Decis-2.8ec-Larvin
Spintor-Decis100-Confidor-Metasystox
Exhibit 6: Product Details
Product Spintor Sherpa Alpha
Type Insecticide Insecticide
y
Technical Spinocyte Alphamethrin (Synthetic pyrethroid)
Crop Cotton Cotton
nl
Pests Bollworm Initial Stage of Bollworm
Pack Sizes and MRP 75 ml Rs 888 250ml- Rs 103
O
250 ml Rs 2766 500ml- Rs 193
1ltr- Rs 377
5ltr- Rs 1850
n
Distributor Margin 2530% of MRP 2530% of MRP
Bayer Gross Margin 16% of MRP 32% of MRP
io
at
Exhibit 7: Categorisation of Products
Category
ul
Gross Margin Objective
irc
A 30-40 % Invest in branding
B 10-20 % Milking
C
C 5-9 % Phase out

D <5% Phase out
d
ite
im
rL
Fo
Exhibit 8: Monthwise Value of Projected Sales Plan as per First RFC and Variations: 2004
y
(Rs million)
nl
East Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total
O
Plan 111.0 87.8 93.7 53.2 28.2 49.9 82.1 102.3 77.3 47.7 49.1 133.4 915.7
Additional -3.8 13.2 8.2 42.2 22.3 6.8 21.4 6.0 5.6 3.4 0.8 35.8 162.0
Total 107.2 101.0 101.9 95.4 50.5 56.7 103.5 108.3 82.9 51.0 49.9 169.3 1077.7
n
% variation -3.4 15.0 8.8 79.3 79.1 13.5 26.1 5.9 7.3 7.1 1.6 26.9 17.7
io
West Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total
Plan 43.5 20.2 12.6 72.3 119.9 275.4 314.8 366.3 129.3 89.5 86.4 74.5 1604.7
at
Additional 12.8 0.2 0.7 3.2 23.4 25.3 -20.5 -2.8 69.0 12.0 5.3 28.9 157.1
Total 56.3 20.0 13.2 75.5 143.3 300.7 294.3 363.5 198.3 101.5 91.7 103.3 1761.8
ul
% variation 29.5 0.9 5.2 4.4 19.5 9.2 -6.5 -0.8 53.4 13.4 6.2 38.7 9.8
irc
North Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total
Plan 31.4 50.5 44.6 71.0 131.2 328.8 211.7 258.6 58.1 52.0 111.6 123.9 1473.4
Case 8: Buyer Cropscience

Additional -3.7 42.5 -5.3 23.9 31.2 4.0 110.7 -82.1 68.6 -21.5 3.3 21.5 108.1
C
Total 27.7 8.0 39.3 94.9 162.4 332.9 322.5 176.5 126.7 30.5 114.9 145.4 1581.5
% variation -11.6 84.2 -12.0 33.6 23.8 1.2 52.3 -31.7 118.1 -41.3 2.9 17.3 7.3
d
South Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total
ite
Plan 165.1 157.8 17.5 50.4 68.1 150.1 267.6 458.1 501.6 531.1 227.7 217.4 2812.5
Additional 21.3 -6.5 10.6 15.2 8.7 6.4 57.0 37.2 165.7 183.4 0.8 131.8 631.5
Total 186.4 151.3 28.0 65.6 76.8 156.5 324.6 495.4 667.3 714.5 228.5 349.2 3444.0
im
% variation 12.9 4.1 60.5 30.2 12.8 4.3 21.3 8.1 33.0 34.5 0.3 60.6 22.5
All 4 Zones Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total
rL
Plan 350.9 316.3 168.4 246.9 347.4 804.3 876.2 1185.3 766.3 720.2 474.8 549.3 6806.2
Additional 26.7 36.0 14.1 84.5 85.6 42.5 168.6 -41.6 308.9 177.3 10.2 218.0 1058.8
451
Fo
Total 377.6 280.3 182.4 331.4 433.0 846.8 1044.8 1143.7 1075.2 897.5 485.0 767.2 7865.0
% variation 7.6 11.4 8.4 34.2 24.6 5.3 19.2 -3.5 40.3 24.6 2.2 39.7 15.6
Exhibit 9: Monthly Planning Cycle
Day of Month Activity
4 Receipt of RFC
y
46 Generate MPS
nl
6 Meeting between IOP and Distribution
67 Release Production Plan
O
89 Release Requisition for Procurement
12 Release Import Plan
16 17 Mid Month Review
n
28 Production Plan for next 10 days
io
at
Exhibit 10: Formulation Units
Unit Type
ul Material Monthly Capacity
irc
WP I Herbicide Powder 100 t
WP II Insecticide Powder 90 t
C
Ezeetab Insecticide Tablets 1.2 m

EC Insecticide/Herbicide Liquid 750 kl
d
ite
Exhibit 11: Filling Stations

im
Filling Station Container Type Pack Size Capacity per Shift

Units
Bulk Flexible 20-200 ltr 300

rL
Flexible 5 ltr 1300

Tin 5 ltr 1700
Small pack PET 250/500 ml 9000
Fo
PET 500 ml 8000

PET 250 1000 ml 9000
PET/ Co-EX 250 1000 ml 9000
Exhibit 12: Zonal Transfers (2004)

(Rs million)
East West North South
Sales 1077.7 1761.8 1581.5 3444.0
y
Transfer within zone 46.74 26.35 37.44 37.15
nl
Transfers outside zone 6.68 29.56 87.14 5.78
Branch returns 4.40 0.91 3.75 1.40
O
n
Exhibit 13: Prominent Locations
io
at
ul
irc
C
d
ite
im
rL
Fo
Exhibit 14: Sales Forecast Accuracy for Jadcherla Market (2004)
Product Pack Forecast Sales Product Pack Forecast Sales

(Kg. or L) (Kg. or L) (Kg. or L) (Kg. or L)
y
Antracol 100 g 56 34.8 Metasystox 100 ml 56 12.6
nl
250 g 77 35.5 250 ml 84 26
500 g 203 25.0 500 ml 91 38.5
O
1 Kg 175 10.0 1L 42 51
Bayrusil 500 ml 35 98.5 Planofix 100 ml 112 65.4
1L 175 178.0 250 ml 294 107.5
n
5L 210 125.0 500 ml 122.5 119
io
Bilcyp 100 ml 7 28.8 1L 35 148
250 ml 60 51.5 Quintal 100 g 39.2 22.2
at
500 ml 210 57.0 250 g 39.2 30
1L 179 106.0 500 g 21 8.5
Bilzeb 500 g
1 Kg
70
70
13.8
18.0
ul Raft
1 Kg
250 ml
28
385
4
250
irc
Confidor 50 ml 49 19.8 500 ml 280 58.5
100 ml 28 12.6 1L 70 16
C
Decis 100 ml 7 16.2 Sevin 100 g 21 6

250 ml 28 33.0 500 g 28 31.5
500 ml 14 49.5 Sherpa Alpha 250 ml 21 8.5
d
Derosal 100 g 154 34.8 500 ml 70 36.5

ite
25 Kg 450 225.0 1L 98 44
Folidol Dust 5 Kg 2100 1955.0 Spark 100 ml 21 29.4
im
10 Kg 4200 2280.0 250 ml 56 57.5

25 Kg 35000 26200.0 500 ml 63 86.5
Hostathion 100 ml 14 72.0 1L 28 112
rL
250 ml 56 725.5 Spintor 75 ml 140 7.5

500 ml 126 1451.5 250 ml 126 8.5
1L 8561 5235.0 Thiodan 100 ml 280 45.6
Fo
5L 14145 4355.0 250 ml 14 90

20 L 1120 2020.0 500 ml 1050 945
Metacid 100 ml 4 43.2 1L 6510 6861
250 ml 14 74.5 5L 9625 6765
500 ml 35 118.5 Whip Super 250 ml 350 604.5
1L 140 111.0 500 ml 2660 3585
1. What are the supply chain challenges faced by Bayer CropScience?

2. Would Bayer CropScience be able to meet these challenges?
y
3. What would be the right inventories to hold at the beginning of each crop cycle?
nl
4. What would be the changes to be implemented in a redesigned supply chain? This
should be accompanied by an implementation plan.
O
n
io
Different performance measures can be used to diagnose whether the Bayer CropScience
supply chain requires a re-structuring.
at
Uniform production would require some actions to deploy the output. One is to
maintain inventory at the factories/warehouses and deploy them at short notice at the
ul
locations where they are demanded. Another is to push inventories to dealers in
anticipation of demand, through a price promotional policy. If uniform production is not
irc
sacrosanct, restructuring the production process to blend chemicals appropriately to make
the right products closer to time of use can be an option. These options can be evaluated
from the data in the case.
C
The problem of stocking optimal amounts to avoid an excess of overstocking or an excess

of shortages can be addressed through the tools of probabilistic inventory, in particular, the
newsvendor logic. The probability of pests attacking can be estimated using the given data.
d
ite
im
rL
Fo
CASE 9
y
nl
FoodWorld was a division of Spencer’s, the retailing company under RPG Enterprises
O
(RPG). This case focuses on the supply chain strategy and throws up some fundamental
questions to do with retailing, in the introduction. At the time of the case, FoodWorld
n
operated with over 6,000 stock keeping units (SKU’s), but were planning to rationalise to a
figure closer to 3,000. The thrust on sourcing was to source directly from a point as far
io
upstream as possible. This was also expected to reduce time and cost of procurement.
The above issues were rendered more significant due to FoodWorld seeing growth
at
opportunities in (a) the perishables and non-branded groceries and (b) an increase in the
number of outlets from the current 19 to 50 in two years, and to even more ambitious
ul
figures in the future.
irc
C
FOODWORLD B: SUPPLY CHAIN STRATEGY
Introduction
d
ite
“The key to our success, given where we are, is effective supply chain management,” was the
sentiment echoed by Mr Shiv Murti, the portly affable Vice President (Merchandising) of
FoodWorld. Having established itself as a long term, large and “here to stay” player in the
im
organised food retailing industry, the primary supply chain management concern for
FoodWorld was how to negotiate best terms for supply of products and source them from as
upstream as possible in the supply chain.
rL
Fo
Case prepared by G. Raghuram, Bibek Banerjee and Abraham Koshy, Indian Institute of Management,
Ahmedabad as a basis for class discussion. Research assistance of Parvathy Raman and Anita
Basalingappa is acknowledged. The case is prepared as a basis for class discussion rather than to illustrate
either effective or ineffective handling of an administrative situation. The authors wish to thank Mr. PK
Mohapatra, Mr. Shiv Murti, Mr. Ganesh Chella and other top managers of FoodWorld for their
generous cooperation and financial support in preparing this case. © Copyright Indian Institute of
Management, Ahmedabad, February 1999. Revised December 2004.
Case 9: Foodworld B: Supply Chain Strategy 457
FoodWorld operated with over 6,000 stock keeping units (SKU’s) of which 80% were
from the organised sector. The remaining, which were from the unorganised sector,
consisted of two supply channels. One was the perishables, which needed to be sourced
fresh, with a focus on quality and direct delivery to stores. The other was the non
perishables, which were repacked and sold under the FoodWorld brand. The challenge in the
y
organised sector SKU’s (which were essentially branded non perishables) was to be able to
deal with the principal directly and obtain as much of a price advantage as possible. The
nl
challenge in the unorganised sector SKU’s (which were growing in number and volume as per
customer expectations, and also had the potential of high margins) was to identify right
O
sources and have good quality management processes.
In a functional sense, the key issues to be dealt with were (a) what to stock, (b) whom and
n
where to source from, (c) how to reduce total delivery time from vendor to store, and (d) how
to reduce the cost of procurement.
io
Company Background
at
FoodWorld was a division of Spencer’s, the retailing company under RPG Enterprises
ul
(RPG). RPG was among the top five business houses in India, with sales of about 65 billion
rupees in 1996-97. Its asset base was over 75 billion rupees in 1997. RPG’s business interests
spanned a variety of sectors including power, tyres, agribusiness, telecommunications,
irc
retailing and others including financial services. RPG got into retailing with the acquisition
of Spencer’s & Co in 1989. It had a large number of partnerships with international
C
companies, including from among the Fortune 500 companies. For FoodWorld, there was a
partnership with Dairy Farm International, a large retail house in Hong Kong.
Spencer’s & Co had been founded in Madras (now renamed as Chennai) in 1865 offering
d
imported items to the large British expatriate and military population. By 1897, it had
ite
grown to be the largest store in India with 65,000 square feet of shopping space. At its peak
in 1940, it had 50 stores in most of the major cities in India. It also had backward integrated
into making some of the products that it sold, like soft drinks, cosmetics etc. After India
im
gained independence from the British in 1947, sales dropped significantly, though Spencer’s
some how survived and continued to offer food, clothing, cosmetics and other high priced
speciality items. The customers were primarily the expatriate community.
rL
Due to its deteriorating sales, Spencer’s had been open for acquisition. In the early 70’s,
ownership changed once. In 1989, RPG purchased Spencer’s, and established it as a separate
division under the leadership of P K Mohapatra, a senior RPG executive. The primary
Fo
motive for this acquisition was the undervalued real estate.

At that time, Spencer’s had nine stores and was the largest retail chain in India. Though
one of the options was to simply focus on the development of the real estate offered by
Spencer’s, the RPG executives looking after Spencer’s felt that the retail business potential
should not be given up easily. It was decided to experiment with one store to test its
potential. If the experiment failed, RPG would close the retail operations. The departmental
store in Bangalore was modernised in 1991, retaining its product profile of hardware, food,
kitchen appliances and clothing. When the store opened, sales increased to four times the
previous levels and returned a healthy contribution. Shiv Murti, who had joined Spencer’s to
head merchandising, had played a significant role in this turnaround. This settled the issue
in favour of continuing with the more important Spencer’s activities including retailing, the
airline general sales agency (GSA) and pharmaceuticals. From Rs. 25 crores at the time of
y
acquisition, turnover had risen to Rs. 100 crores by 1994 through a careful process of
nurturing the three activities, while dropping over 20 others. The airline GSA accounted for
nl
over 80% of the turnover.
During 1994-95, the RPG group went through a reassessment of its portfolio of activities
O
with the help of a large international consulting firm. It was recommended that retail
business development should be one of the key thrusts (along with telecommunication and
n
financial services), since it offered a lot of growth potential. There was an emerging middle
class who had barely had a glimpse of modern retailing. Retailing had traditionally been in
io
the “non formal” small sector, and had remained unchanged for over a century.
The issue then was whether to build on the Spencer’s image or consider a fresh start.
at
Leveraging on Spencer’s for retail business development had its risks. While the “Spencer’s”
brand name was widely associated with quality, it also had a connotation of high prices. The
ul
popularity of this was reflected in the expression, “paying the Spencer’s price,” which was
commonly used to suggest the payment of high prices. Further, the existing Spencer’s
irc
employees were both poorly qualified and underpaid. It was decided that a new retailing
format, with no link to Spencer’s, would be developed.
Based on a study of retail chains world over, it was felt that the retail format should be
C
‘mass’ based rather than ‘niche’ based, since it offered greater growth potential. In order to
develop a mass base, it was decided that the retailing should focus on the daily livelihood of
d
people. Food, clothing and health care products were considered. Clothing was dropped,
since RPG had no background in either the fashion industry or textile manufacturing,
ite
considered essential for success in this area. Food was selected as the first area of entry, to be
followed by health care products.
im
FoodWorld
After a study of customer preferences in early 1996, the supermarket format was selected
rL
to offer (a) value based on price and quality, (b) choice through self service from a spread of
merchandise and (c) a better shopping environment. In terms of location, three choices were
considered (in decreasing order of property prices), (a) commercial high street, (b) residential
Fo
high street and (c) out of town. Since the supermarket was to be positioned as a one stop
shop for all food related items during a shopping outing, the commercial high street was not
that important. Further, with the mass based positioning, the middle income group was
considered the primary target. This group would not have access to their own transport or
would not fancy spending time on transportation for supermarket shopping. Hence, the out
of town location concept was dropped and the residential high street was decided on.
Various ‘ideal’ sizes were considered, keeping in mind likely costs and availability of
properties, ranging from 1,200 sq ft to 6,000 sq ft. Given the self service format and the
merchandising choice, a minimum size of 3,000 sq ft was considered essential, while 4,500
sq ft would be the preferred size. All stores would be air-conditioned, and offer extensive
assortments of groceries, personal care and the cleaning products, kitchenware, and
y
tableware. There would also be a fast food and bakery section. The attempt would be to
provide a pleasant ambience and offer outstanding customer service. With this concept in
nl
view, a name had to be selected for the supermarket retail chain. After brainstorming for a
suitable name, and some market research, the FoodWorld name was selected, along with logo
O
and the bright red and orange signature colour.
The first FoodWorld store was launched on May 9, 1996 in Chennai (Exhibit 1). This
n
was followed by a store in Bangalore (August 20, 1996), and then by two more in Chennai
(September 1, 1996). The M G Road, Bangalore and part of the Mount Road, Chennai
io
Spencer’s stores were converted as FoodWorld stores in December 1996. New stores were
opened at regular intervals, until the current tally of 19 (six in Chennai, eight in Bangalore
at
and five in Hyderabad). In the FoodWorld chain, an investment of Rs 65 crores had been
made so far. The turnover during 1996-97 was Rs 21 crores, 1997-98 was Rs 42 crores, with
ul
a projection of Rs 87 crores in 1998-99 (of which Rs 52 crores had been achieved during
April-November, 1998). Exhibit 2 gives the performance trends for 1996-97, 1997-98 and
irc
April-November, 1998, classified and aggregated according to the stores opened during the
three periods.
The gross margin on this turnover was 16.7% in 1996-97, 18.4% in 1997-98 and
C
projected at 20.5% in 1998-99. (The actuals during April-November 1998 was 18.4%.)
Every one of the stores had broken even within a few months of start and made a
contribution towards regional and corporate expenses. The store operating expenses as a
d
percentage of sales reduced over time as seen in Exhibit 2. This figure during the first year of
ite
operations was higher for later stores, reflecting higher startup costs. The M G Road store in
Bangalore had a gross margin and store operating expenses (as a percentage of sales) of 13.0%
and 7.2% respectively in 1996-97, 18.7% and 7.4% respectively in 1997-98 and 18.3% and
im
6.8% respectively for April-November 1998. Within store operating expenses, salaries and
wages accounted for about 2.5% of sales, rent about 2.2%, shrinkage about 1.6% and
depreciation about 0.7%. In the future, salaries and wages, rent and depreciation was
rL
expected to increase as newer and more expensive properties were to be acquired.

FoodWorld as a business enterprise was still in the red. This was typical of large retail startup
businesses where a critical number of outlets were necessary before the bottom line was
Fo
positive.
Apart from increasing turnover in each of the stores by leveraging a large store network, it
would be imperative to increase margins to ensure sustainability and growth of the
enterprise. This would require appropriate sourcing and negotiating with suppliers for better
margins, and achieving efficiencies in the regional distribution system that FoodWorld had
in place. In this context, the merchandising function (what specific products/brands to offer,
whom to source from) and the distribution strategy (how to organize the logistics of supply
to the stores) gained significance as critical success factors.
Merchandising Function
The merchandising function was carried both at the centralised level (in the corporate office
y
of FoodWorld at Chennai) and the regional levels. Exhibit 3 gives the organisation structure
nl
of FoodWorld, showing the merchandising function along with other key “line” and support
functions, including the regional setup. Exhibit 4 gives the organisation structure of the
O
regional setup. The position of FoodWorld along with other profit centres, and corporate
level staff functions under Mr P K Mohapatra is given in exhibit 5. Each of the six units
(each consisting of profit centres and/or corporate level staff functions) was headed either by
n
a President or a General Manager.
io
The merchandising strategy began with the customer preference study mentioned earlier.
This study highlighted the following customer expectations in terms of considering the store
at
for food purchase.
Product range
ul
Variety for choice in a given product
Freshness
irc
Availability
Reasonable price (not more than neighbourhood stores)
C
In response to this, FoodWorld executives decided that the core range of products would
consist of everything that a household shopped for on a daily/weekly/monthly basis. Insights
d
into this were obtained based on customers’ shopping lists and budgets. A merchandising
offer consisting of seven major groups, namely, staples, processed foods, beverages, non-food,
ite
health & beauty, perishables and hard ware and home appliances was finalised. These were
divided into 49 categories, with a category descriptor as destination, strategic (routine),
convenience and speciality (occasional), depending upon essentiality in the customers’
im
purchase basket and frequency of purchase. Exhibit 6 gives a categorywise description as of

December, 1998. These categories were then further divided into about 270 subcategories, to
rL
enable sourcing, keeping in mind range of products available from suppliers. The total
number of SKU’s in the original list was more than 6,000 (4,500 in six categories, 1,700 in
hardware & home appliances, also called general merchandise. However, most stores carried
about 150 basic items in general merchandise, unless there was an explicit section for this.)
Fo
The various categories, depending on their attributes, were to have the following strategies
regarding the width (brand choice), depth (number of variants), price and tactical usage (for
drawing customers to the store). In any category/sub-category, a minimum of two brands
would be on offer.
Destination Strategic Convenience Speciality
Width High High High High/Med

Depth High Low Low Low
Price Aggressive Aggressive MRP/KVI MRP
Key SKUs Value Added Price
y
Best in City Range
for Margin
nl
Tactical Usage High Medium Medium Low
O
Most of the store sales came from branded non-perishables. However, the perishables and
non-branded repack SKU’s were expected to witness the highest growth with higher than
average margins. The revenue share and margin spread of the main items were as given
n
below.
io
Item Revenue Margin Spread
Perishables 15%
at
Branded
Frozen 10-20%
Dairy
Unbranded
ul 410%
irc
Bakery 30%
Fruits & Vegetables 25%
Repack 10% 520%
C
All others 75% 15%
Almost all of the ‘repack’ items belonged to the staples group and carried a store branding.
However, the total staples group had other companies’ branded products categorised under
d
‘all others’. While the repack items constituted 10% of revenue, the staples group
ite
constituted 28% of revenue. The top 1000 SKUs accounted for 70% of the revenues. Within
this, the growth was expected to be in the repack and perishable items, which constituted
about 100 SKUs.
im
Interior design and layout of a store was a fairly involved subject, taking into account basic
ergonomics, purchase preferences, items on promotion, basic layout of the property etc.
rL
This also involved design of appropriate display hardware like gondolas (an open rack,
typically eight feet in length and five feet in height, with five or six shelves), gondola ends
(for special promotional displays), coolers and freezers, special display hardware for increased
attention, channels on the gondola shelves for stickers, hardware for promotional material
Fo
and signages, etc. It was also important to design a bay (a set of gondolas) keeping in mind
the alignment of shelves etc. The number of checkout counters (tills), staff assignment,
assignment of tills for different types of customers (large/small volume, cash/credit), security
features were also decisions of significance.
In terms of product display, the destination categories (which were an essential
component of the purchase basket) would typically be in the far end of the store. This would
attract customers right into the store. The strategic and convenience categories were
displayed so as to provide good visibility and access, like in the eye level shelves rather than in
the top or bottom shelves. The various procedures followed for display in a store, signages,
shelving and stacking were determined and documented with a lot of “science” behind them.
These procedures were to be followed by the customer service supervisors and
representatives. The daily routines of the key store staff including the customer service
y
supervisors and representatives, cashiers at the checkout counters (tills), goods receiving staff,
etc were developed to a great level of detail and constantly reinforced through training and
nl
periodic meetings including one before every shift.
The typical number of bays and SKU’s, based on the type of the store was as given below.
O
Store Trading Area No of Example Location Typical No of
Type (Sq feet) Stores No of Bays SKU’s
n
A 6000+ 2 MG Road (Bangalore)
io
Mount Road (Chennai) 120+ 6000
B 3000+ 9 Himayatnagar (Hyderabad)
at
Gandhi Nagar (Chennai) 80-95 4500
C 2000+ 8 Jaynagar (Bangalore)
Marredpally (Hydreabad)
ul 65-75 3000
Each bay was 3 feet in width, with a height of 5 or 6 feet. Correspondingly, number of
irc
shelves per bay was either six or seven. All the shelves had a depth of 15 inches except the
bottom one which had 18 inches for increased visibility. A bay with six shelves thus had 18
running feet and could accommodate 45-70 facings of SKUs, depending on the SKU
C
dimensions. An SKU may have one to three facings depending upon the sales volume and
desired visibility. A description of SKU’s in two bays is given in exhibit 7. (The “MBQ” and
“SUF” attributes are described later under the indenting process.)
d
ite
Distribution (Supply Chain) Strategy

The key elements of FoodWorld’s distribution strategy, which actually addressed the supply
im
chain management, were as follows.

i) Minimum Suppliers – This would have benefits due to (a) economies of scale, both in
purchasing and supply logistics, (b) reduced overheads and control requirements and (c)
rL
easier vendor development

ii) Creation of Regional Hubs – Three regional offices (at Chennai, Bangalore and
Hyderabad) were set up to address the statewise requirements. This facilitated over
Fo
90% central distribution (a level desired by FoodWorld. The remaining 10% were
supplied direct to store (mostly perishable items like fruits and vegetables, bakery etc).
iii) Replenishment Frequency – The desired servicing of stores from the warehouses was
daily, while the supply frequency for any specific SKU was twice a week. The desired
ordering and servicing frequency from suppliers to the warehouses was weekly.
Hardware and general merchandise items were treated as exceptions, to be indented and
ordered as required.
iv) Sourcing with Minimum Intermediaries – The idea was to source from as “upstream” as
possible in the supply chain. This would help reduce losses and increase margins for
FoodWorld.
Indenting Process
y
This referred to the request for stocks made by the different stores on their servicing
nl
warehouse/direct supply. Most SKU’s could be indented twice a week on nominated days,
for which supply would be made from the warehouse after a gap of a day, again on nominated
O
days (Exhibit 8). The nominated days enabled synchronisation of indent processing at the
warehouse. Direct supply products were indented daily, directly from the suppliers.
Hardware items were indented based on need, and using the information of the ordering
n
schedule from suppliers by the warehouse (Exhibit 9).
io
Each SKU in a store had two specific attributes that helped indenting, namely minimum
base quantity (MBQ) and supply unit factor (SUF). The MBQ was determined as follows.
at
MBQ = 120% of highest sales achieved per month
ul
4
The rationale behind this was there should be enough stocks even if there was (a) a surge
irc
in demand and (b) short/no supply from the warehouse until the next indent, thus providing
for a week. The indent quantity was determined after a physical verification of store stocks
C
on nominated indent days, as follows. The average stock turns per year was 12 in 1997 and
expected to be higher in 1998.
d
Indent quantity = (MBQ – Physical stock) in multiples of SUF

ite
The quantity was in multiples of SUF to enable convenient repacking of suppliers’ stocks
at the warehouse. For most SKUs, the shelf space volume was less than the MBQ. This was
by choice, to enable frequent replenishment and to provide a sense of fullness in the shelves.
im
The indenting process was being automated, so that stock levels would be obtained based on
point of sale data. As a result, indent quantities would be automatically generated. Key
indicators like stock out percentage etc would also then get generated.
rL
A random examination of certain records gave a 20% stock out figure. As explained by a
store manager, the true stock out was only 10%. The rest was accounted for by items which
had been discontinued, but not yet deleted from the indent records. The indent fill (no of
Fo
SKUs) rates from the warehouse was typically 60%. The case fill (SKU quantity) rates were
85%. For certain SKUs, they were even more than 100%, especially if other sizes of the same
brand or brands of the same item were not available. If the case fill rate was less than 75%, it
was considered as the indent not having been serviced. Exhibit 10 gives a sample of two
successive indents from the same store for the tea/coffee category. The stock in the store, the
indented quantity and the picked quantity for supply from the warehouse are given.
Ordering Process
The category managers in the warehouse monitored the stock position of the SKUs under
their control and placed orders from the suppliers. Exhibit 11 gives an example of three SKUs
for which the base quantity stocks have been defined. For example, Cadbury’s 5 Star (38 gms)
had a base quantity (MBQ for A, B and C category stores were 240, 96 and 48 respectively)
y
of 420 units in the warehouse (D). The actual stocks being 1046, no order was yet due.
nl
However, an order would be due for Surf Excel since actual stocks were less than the
warehouse base quantity. Exhibits 12 and 13 give sample purchase orders placed on vendors
O
and the goods receipt note from the vendors, respectively. In all the three cases, supplies were
received within the third day.
For almost all of the SKUs, the supplies were made from the nearest vendor, who was the
n
dealer appointed by the supplying company. Dealer margins were anywhere up to 10%.
io
Most dealers were willing to operate with a weekly frequency of supplies to the warehouse, by
consolidating all SKUs to be supplied by them. The average number of SKU’s per vendor
at
(supplying company) was about 20. Exhibit 14 gives the list of top vendors from whom the
cumulative purchase value in a month was 75% of the total purchase, in which 63 vendors
ul
accounted for 4106 SKU’s. (The top five companies accounted for about 1400 SKUs.) The
order fill rate was typically in the range of 60% to 75%. More than one category manager
could be dealing with the top few vendors for raising orders. For example, five category
irc
managers dealt with Hex Trading.
Generally, the products were received through LCVs (one to three tonners). On receipt at
C
the warehouse, a quantity and quality check was undertaken before accepting the goods. The
next step was to paste bar coded stickers as per FoodWorld’s code, to enable easy processing at
the point of sale. However, there were certain items like some of the vegetable oils, tetrapack
d
drinks, etc for which sticker sheets were only enclosed with the cartons and left as a task for
ite
the stores to complete before display. This was due to the fact that such SKUs were sensitive
to additional handling.
Indents received from stores during a given working day (say day 1) were processed on day
im
2 and then despatched to the stores on day 3. Non-branded staples were repacked using the
FoodWorld brand, in a separate “repack” section in the warehouse, equipped with weighing
and packet sealing machines. Most of the processing in the warehouse and in the store was
rL
computerised. However, integration had not yet been achieved. The store representatives and
the warehouse staff consisted of both permanent and temporary personnel. The ratio was
roughly even.
Fo
Vendor Development
The key elements of the vendor development process were as follows.
(i) Identification of the supplying company’s one point contact.
(ii) Driving towards standardised trading terms across all three regions on the following
dimensions.
(a) Credit
(b) Promotion
(c) Single point sources of supply across SKU’s/categories, preferably direct from the
company’s depot/CFA
(d) Margins (over and above product retail margins) for turnover, distributors’
y
allowance, new store opening, bar coding (for data) and trade schemes
nl
The above strategy would also be driven by when a supplying company was willing to take
notice of FoodWorld. This would depend on the volume of sales and perceived value of the
O
level of services offered by FoodWorld. For FoodWorld, with an annual turnover of Rs 100
crores, dealing with a company like Hindustan Lever Ltd, with an annual turnover of Rs
7,800 crores (Exhibit 15) would be a challenge. In terms of branded food processing, the top
n
five companies (led by Hindustan Lever Ltd. at Rs 2,000 crores) accounted for an annual
io
turnover of Rs 6,000 crores. The top 20 companies accounted for Rs 10,000 crores. Certain
consumer goods companies had been willing to provide better terms and services to
FoodWorld, including direct supply from depots or at least having a specially selected dealer
at
service the warehouse. The possibility of better information processing through electronic
data interchange was also being considered. One of the arguments that suppliers had against
ul
direct supply was the implied reduced earning potential for their distributors.
The different supply chain structures in operation at FoodWorld are given below.
irc
i. Supplier ———— CFA ———— Distributor ————— W/H ————— Store
C
(Branded products from consumer goods companies)

d
ii. Supplier ————— CFA ————————— W/H ————————— Store

ite
(Experimental)
im
iii. Supplier—————Repack Section —————————— W/H————— Store

(Non-branded staples)
rL
iv. Supplier ————————— Consolidation Point ———————————Store

(Some Fruits & Vegetables)
Fo
v. Supplier ——————————————————————————— Store

(Bakery, Diary, Some Fruits and Vegetables)
FoodWorld saw a potential of increasing their margins by 4% to 10%, by negotiating and
sourcing directly from producing locations for branded products. The scope was even more
in non-branded staples and perishables (like fruits and vegetables), where there could be
many intermediaries even between producing locations and the visible suppliers. As seen in
Exhibit 16, intermediaries, in comparison to a developed country like the United States,
accounted for a larger share of the margins, while at the same time contributing less in terms
of value addition.
As an example, the cost of rice procurement for the Madras region and its variation since
y
January 1997 is given in exhibit 17. The selling price of the rice and its impact on sales,
nl
percentage and absolute margins is also provided. In the early months, rice of the appropriate
quality used to be procured from the wholesale market in Madras. Later on, procurement
from the markets near the millers in Andhra Pradesh was organised. The next step was to
O
attempt direct procurement from mills, and then possibly from paddy markets after which
milling would be done on a contract basis. While the scope for increasing margins would
n
increase, FoodWorld might have to get into activities not necessarily within its realm of
expertise.
io
Conclusion
at
FoodWorld felt that the number of SKU’s they were dealing with was too large. A
ul
rationalisation effort focused on reducing the high end subcategories/SKUs, low volume
SKUs and the depth (number of variants) was considered essential, to reduce the maximum
SKU count to 3000. Exhibit 18 gives the expected profile of “value added” food
irc
consumption in India by 2005. Exhibit 19 gives the economics of a typical store, for which
the breakeven would be Rs 20.8 lakhs/month at a 17% store margin.
C
The VP (Merchandising) was aware that over the next five years, an additional investment
of Rs 300 crores was planned to increase the number of outlets from the present 19, to 26 by
mid 1999, 50 by mid 2000, 120 by 2002, and 300 by 2004. The challenges for supply chain
d
management in this context were going to be significant and needed a proper strategic
ite
response. As Mr. P K Mohapatra said, “Organised retailing pre-supposes retailer’s ability to

be able to influence or more importantly manage a set of supply chains to be able to deliver
value to the consumer; in a commercially viable and sustainable way”.
im
rL
Fo
Exhibit 1: Storewise Description
Store Date of Total Trading Store Store Oct.98 Oct.98 Oprtg PBIT PBT
Location Opening Area Area Type Type Sale GM Exp
Sq Ft Sq Ft Based on Based Rs % Rs 000 Rs 000 Rs 000
y
Trading on 000
Area Sale
nl
Madras
Store 1 Dec 96 14065 7257 A A 9156 22.41 694 1358.09 1005
O
Store 2 09.05.96 3000 2404 C B 4415 19.43 351 506.66 428
Store 3 01.09.96 4494 3094 B B 4619 20.12 392 537.34 398
Store 4 01.09.96 4231 3177 B B 5431 20.39 399 708.33 538
n
Store 5 09.05.98 5000 3305 B B 4151 19.89 485 340.80 137
io
Store 6 23.10.98 3500 2700 C B 1269 19.89 212 40.45 -60
at
Bangalore
Store 1 Dec 96 9500 6492 A ul A 11487 19.04 754 1433.13 1265
Store 2 20.8.96 5720 3976 B B 4072 16.61 494 182.36 127
Store 3 5.6.98 2565 2472 C B 4910 17.18 420 423.54 285
irc
Store 4* 29.12.96 4362 3729 B B 4242 17.40 480 258.11 61
Store 5 5.4.97 3650 3421 B C 2180 15.75 324 19.35 -44
Store 6 4.5.97 2862 2437 C B 2910 16.84 342 148.04 95
C
Store 7 30.7.98 3024 2494 C B 4825 16.81 388 423.08 65

Store 8* 25.4.98 3000 2656 C C 2569 16.40 324 97.32 -4
d
Hyderabad
ite
Store 1 26.9.97 4500 2971 C B 3117 16.78 349 174.033 150

Store 2 13.12.97 4000 3278 B A 4690 16.20 426 333.78 301
Store 3* 17.4.98 3000 2168 C C 2520 16.71 264 157.09 137
im
Store 4 1.10.98 4084 3500 B B 3445 16.63 426 146.90 120

Store 5* 31.10.98 3500 3000 B B 179 14.22 143 -117.55 -190
* Some of the data are estimated
rL
Source: Company Records, December,1998

Fo
Exhibit 2: Performance Trends
y
468
(Rs lakhs)
nl
Year 1996-97 1997-98 YTD Apr ’98-Nov ‘98 1998-99 1999-00
O
Budget % of Actuals % of Budget % of Actuals % of Budget % of Actuals % of Proj % of Proj % of
sales Sales Sales Sales Sales Sales Sales Sales
Stores

n
New Stores96-97 8 7
New Stores 97-98 13 4
io
New Stores 98-98 10 8
Total No of Stores 8 7 20 11 21 19
at
Sales
New Stores96-97 2577 2088 3366 3654 2830 3148
New Stores 97-98 1118 579 764 898
ul
New Stores 98-98 1241 1106
Subtotal 2577 2088 4484 4233 4835 5152 8665 100% 12957 100%
irc
Gross Margin
New Stores96-97 382 14.8% 349 16.7% 633 18.8% 671 18.4% 530 18.7% 587 18.6%
New Stores 97-98 152 13.6% 107 18.5% 143 18.7% 147 16.4%
New Stores 98-98 234 18.9% 212 19.2%
C
Subtotal 382 14.8% 349 16.7% 785 17.5% 778 18.4% 907 18.8% 946 18.4% 1774 20.5% 2786 21.5%
Store Operation
New Stores96-97 346 13.4% 266 12.7% 480 14.3% 444 12.2% 286 10.1% 286 9.1%
d
New Stores 97-98 171 15.3% 126 21.8% 115 15.1% 112 12.5%
ite
New Stores 98-98 284 22.9% 165 14.9%
Subtotal 346 13.4% 266 12.7% 651 14.5% 570 13.5% 685 14.2% 563 10.9% 1408 16.2% 2119 16.4%
Warehouse/
Regional Office
im
New Stores96-97 136 5.3% 134 6.4% 219 6.5% 268 7.3% 143 5.1% 167 5.3%
New Stores 97-98 73 6.5% 42 7.3% 39 5.1% 48 5.3%
New Stores 98-98 63 5.1% 58 5.2%
rL
Subtotal 136 5.3% 134 6.4% 292 6.5% 310 7.3% 245 5.1% 273 5.3% 437 24.6% 494 17.7%
Corporate Opex
New Stores96-97 112 4.3% 96 4.6% 57 1.7% 76 2.1% 32 1.1% 40 1.3%
New Stores 97-98 19 1.7% 12 2.1% 9 1.2% 11 1.2%
Fo
New Stores 98-98 13 1.0% 14 1.3%

Subtotal 112 4.3% 96 4.6% 76 1.7% 88 2.1% 54 1.1% 65 1.3% 95 1.1% 104 0.8%
(Continued)
y
nl
(Continued)
Year 1996-97 1997-98 YTD Apr ’98-Nov ‘98 1998-99 1999-00
O
Budget % of Actuals % of Budget % of Actuals % of Budget % of Actuals % of Proj % of Proj % of
Sales Sales Sales Sales Sales Sales Sales Sales
n
EBIT
New Stores96-97 212 8.2% 147 7.0% 123 3.7% 117 3.2% 69 2.4% 94 3.0%
io
New Stores 97-98 111 9.9% 73 12.6% -20 2.6% 24 2.7%
New Stores 98-98 126 10.2% 25 2.3%
at
Subtotal 212 8.2% 147 7.0% 234 5.2% 190 4.5% 77 1.6% 45 0.9% 166 1.9% 69 0.5%
Notes: 1. Figures exclude notional rent on owned properties
Case 9: Foodworld B: Supply Chain Strategy

2. POS Maintenance included under Maintenance
ul
3. Warehouse & Corporate Opex apportioned to Existing and New Stores on the basis of sales
Source: Company Records, December, 1998
irc
C
d
ite
im
rL
469
Fo
Exhibit 3: FoodWorld Organisation Structure
y
nl
O
n
io
at
ul
Exhibit 4: Regional Organisation Structure
irc
C
d
ite
im
rL
Fo

Exhibit 5: Retail and Entertainment Organisation Structure
Retail and Entertainment Sector
y
Retail 1 Retail 2 Entertainment Sector Human Sector Information Future Projects
Technology Cash and
nl
GCIL (HMV) Carry
FoodWorld Travel GSA
O
MusicWorld Health & Glow Music World
RMIL International
and Electronic
Spencers
Properties GRAMCO films Commerce
Super
n
Stores
Mount Road GRAMCO publishing
io
Kodaikanal
Mercara
Thiru-
at
vanathapura
Source: Company Records, December, 1998 ul

irc
C
d
ite
im
rL
Fo
Exhibit 6: Categorywise Description
Category Main Group Sub SKUs Turn Reve- Ave- Hash Category
Cate- over nue rage Sale Description
gories (oct 98) Margin Qty
y
# # Rs 000 % % #
nl
CEREALS 01 STAPLES 9 72 5909 7.47 10.95 69347 DESTINATION
PULSES 02 STAPLES 4 75 3432 4.34 16.61 143070 DESTINATION
O
FLOURS 03 STAPLES 5 70 2475 3.13 18.54 116394 DESTINATION
SPICES 04 STAPLES 8 250 5007 6.33 23.18 259619 ROUTINE
EDIBLE OILS 05 STAPLES 9 56 5632 7.12 8.63 84976 DESTINATION
n
SAVOURIES 06 PROCESSED FOODS 5 187 1959 2.48 29.91 132050 CONVENIENCE
BISCUITS 07 PROCESSED FOODS 7 98 1660 2.10 16.05 153810 ROUTINE
io
NOODLES 08 PROCESSED FOODS 4 64 838 1.06 16.71 52123 CONVENIENCE
CONFECTIONARY 09 PROCESSED FOODS 10 74 3519 4.45 16.49 195028 CONVENIENCE
at
CAN NON-VEG 10 PROCESSED FOODS 2 15 51 0.06 18.25 1182 CONVENIENCE
CAN FRUIT 11 PROCESSED FOODS 3 15 25 0.03 21.94 475 CONVENIENCE
CAN VEG. 12 PROCESSED FOODS 3 ul3 78 0.10 16.72 2071 CONVENIENCE
DESSERTS 13 PROCESSED FOODS 6 90 582 0.74 15.75 23242 CONVENIENCE
JAMS 14 PROCESSED FOODS 4 58 580 0.73 15.27 15938 ROUTINE
irc
TINNED MILK 15 PROCESSED FOODS 2 20 279 0.35 9.12 5495 CONVENIENCE
SAUCES 16 PROCESSED FOODS 7 29 619 0.78 16.34 20678 ROUTINE
PICKLES 17 PROCESSED FOODS 4 106 538 0.68 17.29 14696 ROUTINE
C
READY TO FRY 18 PROCESSED FOODS 3 99 569 0.72 24.50 39357 ROUTINE

SOUPS 19 PROCESSED FOODS 2 26 462 0.58 14.37 20272 CONVENIENCE
BREAKFAST FOODS 20 PROCESSED FOODS 4 37 841 1.06 12.71 18099 ROUTINE
d
BABY FOODS 21 PROCESSED FOODS 4 33 234 0.30 9.27 2475 ROUTINE

ite
DRINKS 22 BEVERAGES 9 94 2071 2.62 22.44 83500 CONVENIENCE

TEA/COFFEE 23 BEVERAGES 4 90 2443 3.09 9.84 50649 ROUTINE
HEALTH DRINKS 24 BEVERAGES 5 37 1432 1.81 10.25 19568 ROUTINE
im
LIQUOR 25 BEVERAGES 8 98 4482 5.67 20.05 30899 ROUTINE

DETERGENTS 26 NON-FOOD 4 70 2425 3.07 10.09 92479 ROUTINE
HOUSE CLEANING 27 NON-FOOD 6 114 2319 2.93 16.80 104996 ROUTINE
rL
PAPER GOODS 28 NON-FOOD 5 40 463 0.59 40.32 14995 OCCASIONAL

HOUSEHOLD NEEDS 29 NON-FOOD 9 170 1253 1.58 28.75 69973 ROUTINE
CIGARETTES 30 NON-FOOD 3 20 348 0.44 9.34 11156 CONVENIENCE
Fo
PET FOOD 31 NON-FOOD 1 35 62 0.08 30.00 131 OCCASIONAL

SKIN CARE 32 HEALTH & BEAUTY 9 250 3281 4.15 12.81 143241 ROUTINE
HAIR CARE 33 HEALTH & BEAUTY 6 160 1551 1.96 14.16 32011 ROUTINE
ORAL CARE 34 HEALTH & BEAUTY 5 85 1378 1.74 16.28 58231 ROUTINE
SANITARY GOODS 35 HEALTH & BEAUTY 4 30 615 0.78 12.52 14398 ROUTINE
SHAVING NEEDS 36 HEALTH & BEAUTY 4 85 1299 1.64 15.45 21386 ROUTINE
O.T.C. 37 HEALTH & BEAUTY 5 70 369 0.47 22.71 14249 CONVENIENCE
BABY NEEDS 38 HEALTH & BEAUTY 3 55 546 0.69 15.84 10202 ROUTINE
(Continued)
(Continued)
Category Main Group Sub SKUs Turn Reve- Ave- Hash Category
Cate- over nue rage Sale Description
gories (oct 98) Margin Qty
# # Rs 000 % % #
COSMETICS 39 HEALTH & BEAUTY 10 94 197 0.25 19.50 2695 CONVENIENCE
y
HERBAL 40 HEALTH & BEAUTY 7 240 331 0.42 24.77 4262 CONVENIENCE
nl
DAIRY PRODUCTS 41 PERISHABLES 7 60 2008 2.54 12.00 104150 DESTINATION
FROZEN FOODS 42 PERISHABLES 7 75 1610 2.04 17.00 41704 CONVENIENCE
O
BAKERY 43 &44 PERISHABLES 3 65 4497 5.69 27.78 157159 CONVENIENCE
VEGETABLES 45 & 46 PERISHABLES 2 110 4365 5.52 17.28 407030 DESTINATION
BATTERIES 47 H/WARE & HOME APP. 4 48 306 0.39 23.98 11967 OCCASIONAL
n
MISC. 48 H/WARE & HOME APP. 5 210 1770 2.24 20.21 10116 OCCASIONAL
HARDWARE 49 H/WARE & HOME APP. 24 1435 2350 2.97 25.42 51446 CONVENIENCE
io
TOTAL 264 5317 79060 100.00 17.17 2932990
at
ul
irc
C
d
ite
im
rL
Fo
Exhibit 7: Description of SKU’s in Two Bays

Bay No: 35
Section: Tea (Ht 5 Feet, 6 Shelves)
No SKU SKU Description Company Shelf Depth MBQ Facings SUF Cost MRP
y
1 191148 TAJ MAHAL TEA BAGS 25S HLL 1 15" 12 2 6 63.8 69
nl
2 191146 GREEN LABEL 100 GMS HLL 1 12 2 6 23.44 25
3 191087 YELLOW LABEL 100 GMS HLL 1 12 2 6 15.89 17
O
4 191154 KOTADA TEA 100 GMS Kotada 1 8 1 3 11.90 14
5 191052 TOP STAR 100 GMS HLL 1 12 2 6 17.46 18.25
6 191057 TAJ MAHAL 100 GMS HLL 2 15" 18 3 6 17.75 18.5
n
7 191027 RED LABEL 100 GMS HLL 2 18 3 6 14.2 15
io
8 191147 GREEN LABEL TEA
W.B. 250 GMS HLL 2 4 1 3 94.12 110
9 191086 YELLOW LABEL TEA
at
250 GMS HLL 2 12 2 6 64.15 68
10 191059 TAJ MAHAL TEA 250 GMS HLL 3 15" 12 3 6 45.5 76.5
11
12
191026
191064
RED LABEL 250 GMS
GREEN LABEL TEA 250 GMS
HLL
HLL
ul 3
3
12
8
2
2
6
6
38.25
64.15
43
68
irc
13 191053 TOP STAR 250 GMS HLL 3 8 2 3 39.86 76.5
15 191025 RED LABEL 500 GMS HLL 4 8 2 3 70 73
C
16 191085 YELLOW LABEL 500 GMS HLL 4 8 2 3 79.44 85

17 191058 TAJ MAHAL TEA 500 GMS HLL 5 15" 8 2 3 78 82
18 191000 RED LABEL 500 JAR HLL 5 6 2 3 83.58 88
d
19 191020 SARGAM TEA 250 GMS JAR Duncan 5 6 1 3 28.98 30

20 191153 KOTADA TEA 250 GMS Kotada 5 12 1 6 24.23 28.5
ite
21 191060 TOP STAR 500 GMS HLL 5 6 2 3 73.91 76.5

23 191101 SARGAM TEA 250 REFIL Duncan 6 6 1 3 23.93 25
im
24 191021 RED LABEL 250 JAR HLL 6 12 2 6 41.75 44

25 191065 LIPTON GREEN LABEL
500 GMS HLL 6 12 1 6 128.3 136
rL
26 191152 KOTADA TEA 500 GMS Kotada 6 6 1 3 56.1 66

27 191035 SARGAM TEA JAR 500 Duncan 6 12 2 6 76.36 83
Bay No: 36
Fo
Section: Coffee (Ht 5 Feet, 6 Shelves)

1 190015 GREEN LABEL COFFEE
50 GMS HLL 1 15" 36 3 12 7.28 7.75
2 190012 BRU 50 GMS HLL 1 36 3 12 24.3 28
3 190053 NESCAFE 25 GMS Nestle 1 18 3 6 26.1 28
4 190088 SUNRISE PREMIUM 50 GMS Nestle 2 15" 36 4 12 26.9 30
(Continued)
(Continued)
No SKU SKU Description Company Shelf Depth MBQ Facings SUF Cost MRP
5 190014 BRU SACHET STRONGM

50 GMS HLL 2 24 2 12 24.3 26
6 190060 NESCAFE SELECT 50 GMS Nestle 2 12 2 6 52.92 59
y
7 190092 SUNRISE EXTRA 50 GMS Nestle 3 15" 36 3 12 23.35 25
8 190009 TATA KAPI 50 GMS Tata Tea 3 24 2 12 24.23 28
nl
10 190056 NESCAFE SELECT 100 GMS JAR Nestle 3 24 2 12 105 115
O
11 190091 SUNRISE PREMIUM 100 GMS Nestle 4 15" 12 1 6 54.25 58
12 190007 BRU 100 GMS HLL 4 12 1 6 50.49 54
13 190029 COORG PURE FILTER 100 GMS Tata Tea 4 12 1 6 12.75 15
n
14 190016 GREEN LABEL 100 GMS HLL 4 12 1 6 14.09 15
io
15 190034 BROOKE BOND CAFE
100 GMS HLL 4 12 1 6 10.33 11
16 190042 COORG DOUBLE ROAST
at
100 GMS Tata Tea 4 12 1 6 10.84 12.75
17 190090 SUNRISE PREMIUM
18 190084
100 GMS JAR
SUNRISE PREMIUM 200 GMS
Nestle
Nestle
ul 4
5 15"
8
12
1
1
3
6
51.4
101.87
55
115
19 190008 BRU 200 GMS HLL 5 12 1 6 104.67 112
irc
20 190031 COORG PURE FILTER
200 GMS Tata Tea 5 12 1 6 25.16 29.6
C
200 GMS Tata Tea 5 12 1 6 21.25 25

22 190010 TATA KAPI 200 GMS Tata Tea 5 12 1 6 89.25 105
23 190065 FOOD WORLD PURE
d
FILTER COFFEE 200 GMS 5 12 1 6 30 36

ite

25 190080 NESTLE SUNRISE EXTRA
200 GMS Nestle 6 18" 12 1 6 84.15 90
im
200 GMS HLL 6 12 1 6 28.57 30

500 GMS HLL 6 9 1 3 70.48 74
rL
28 190027 COORG PURE FILTER

COFFEE 500 GMS Tata Tea 6 9 1 3 62.05 73
500 GMS Tata Tea 6 9 1 3 60 65
Fo
30 190066 FOOD WORLD COFFEE

500 GMS 6 9 1 3 75 90
500 GMS Tata Tea 6 9 1 3 67.38 73
32 190076 NESCAFE SELECT
500 GMS Nestle 6 9 1 3 475.75 520
Note: HLL Hindustan Lever Limited
Exhibit 8: Schedule of Indents and Despatches for a Region
Store Indent Schedule Warehouse Despatch Schedule
SUNDAY
y
DESSERT MIXES BABY FOOD
B F CEREALS TIN MILK
nl
BISCUITS TEA/COFFEE
JAMS CONFECTIONARY
O
H DRINKS H H CLEANING
BRANDED FLOURS PAPER GOODS
H H NEEDS DETERGENTS
n
HAIR CARE ORAL CARE
io
LIQUOR OTC
PICKLES SPICES AND DAILY INDENT
AMUL PRODUCTS
at
RICE FLOOR AND DAILY INDENT
MONDAY
ul
SOUPS CANNED FOOD
DRINKS READY TO FRY
irc
NOODLES/VERMACELLI B R SAVOURIES
SAUCES SANITARY
OILS BATTERIES/ELECTRICAL
C
BRANDED SPICES DAL AND DAILY INDENT

BRANDED RICE
d
SKIN CARE
SHAVING NEEDS
ite
HERBAL COSMETICS
BABY NEEDS
CHEESE
im
SPICES AND DAILY INDENT
TUESDAY
rL
TEA/COFFEE DESSERT MIXES

CONFECTIONARY B F CEREALS
H H CLEANING BISCUITS
PAPER GOODS JAMS
Fo
ORAL CARE H DRINKS

DETERGENTS BRANDED FLOURS
PLEASE DO TOP UP INDENTING TO DAY H H NEEDS
HAIR CARE
CIGARETTES
RICE FLOUR AND DAILY INDENT
(Continued)
(Continued)
WEDNESDAY
JAMS SOUPS
H DRINKS DRINKS
BISCUITS NOODLES/VERMECELLI
B F CEREALS SAUCES
DESSERT MIXES OILS
y
BR SAVOURIES BRANDED SPICES
HAIR CARE BRANDED RICE
nl
SANITARY SKIN CARE
H H NEEDS SHAVING NEEDS
O
PICKLES BABY NEEDS
DAL AND DAILY INDENT PICKLES
CHEESE
n
SPICES AND DAILY INDENT
THURSDAY
io
DRINKS TEA/COFFEE
NOODLES/VERMECELLI CONFECTIONARY
at
SAUCES H H CLEANING
OILS PAPER GOODS
BRANDED SPICES DETERGENTS
ul
SKIN CARE ORAL CARE
SHAVING NEEDS ONLY DESPATCH OF TOP UP ITEMS
irc
BABY NEEDS
CHEESE
C
FRIDAY
TEA/COFFEE DESSERT MIXES
CONFECTIONARY B F CEREALS
d
H H CLEANING PAPER GOODS BISCUITS

ORAL CARE JAMS
ite
DETERGENTS H DRINKS
BABY FOOD BRANDED FLOURS
TIN MILK H H NEEDS
im
OTC HAIR CARE

SPICES AND DAILY INDENT SANITARY
ERBAL COSMETICS
rL
DAL AND DAILY INDENT
SATURDAY
CANNED FOOD DRINKS
READY TO FRY NOODLES/VERMECELLI
Fo
B R SAVOURIES SAUCES
SANITARY OILS
BATTERIES/ELECTRICAL BRANDED SPICES
CIGARETTES SKIN CARE
DAL AND DAILY INDENT SHAVING NEEDS
BABY NEEDS
PICKLES
CHEESE
Source: Company Records, December, 1998.
Exhibit 9: Hardware Indent Schedule for a Region
MONDAY
BHARAT/KLASSIK
y
nl
TUESDAY
YERA GLASSWARE
O
ANJALI KITCHENWARE
WALT DISNEY/FISKARS
n
HALLMARK MUGS
io
WEDNESDAY
at
CLEARLINE
LA OPALA
ul
PHILIPS
irc
SS CUTLERY
DECOPRIDE
RAMSONS CUTLERY
C
FRIDAY
d
ARCHANA PRODUCTS
ite
AB PLASTIC CONTAINERS
im
SATURDAY
SHRIPET PLASTICWARE
rL
PRESTIGE COOKERS
BUTLERS/ANUPAM
Fo
SUNDAY
ROSEWOOD ITEMS
HIP FLASK
M/W PROOF CONTAINERS

Exhibit 10: Store Indent on a Warehouse

Report No: 0F8 Indent Date: /11/98 (Sunday)
Store No: 863 Store Name: AAA
No Item Category :TEA/COFFEE Stock Qty Indent Picked
y
Code Qty Qty
nl
Sub Category : 2301 GROUND COFFEE
Indent Number : [630486]
O
1 190046 BROOKEBOND GREEN LABEL COFFEE 278 6 6
2 190024 BROOKE BOND GREEN LABEL COFFEE 244 12 12
3 190027 COORG FILTER COFFEE 500 GMS 74 9 9
n
4 190040 COORG DOUBLE ROAST COFFEE 500 GMS 83 12 12
5 190043 COORG D.R. 500 GM JAR 0 12
io
6 190065 F.W. PURE FILTER COFFEE 200 G 0 12
7 190066 F.W. PURE FILTER COFFEE 500 G 0 12
at
Sub Category: 2302 INSTANT COFFEE
Indent Number : [630487] ul
1 190007 BROOKEBOND BRU INSTANT COFFEE 0 12
2 190014 BRU SUPER STRONG 50 GM 0 12
irc
3 190053 NESCAFE 25 GM JAR 0 12
4 190056 NESCAFE INST COFFEE 100 GM JAR 0 12
5 190064 NESCAFE SELECT INST COFFEE 100 0 12
C
6 190090 NESTLE SUNRISE PREMIUM 100 GM 0 12

7 190091 NESTLE SUNRISE PREMIUM 100 GM R 0 12
d
Sub Category: 2303 DUST TEA

ite
1 191037 SARGAM DUST TEA 500 GM 0 12

2 191039 DOUBLE DIAMOND DUST TEA 250 GM 0 12
3 191147 LIPTON GREEN LABEL TEA 250 W.B 0 12
im
Sub Category: 2304 CTC TEA

Indent Number: [630489]
1 191000 BROOKE BOND RED LABEL 500 GM J 18 6 6
rL
2 191006 BROOKE BOND 3 ROSES 250 GM JA 0 6

3 191021 BROOKE BOND RED LABEL 25O GM JA 0 12
4 191025 BROOKE BOND RED LABEL TEA 500 134 12 12
Fo

7 191028 CHEERS DARJEELING LEAF TEA 250 21 12 12
8 191043 DEVAN TEA 100 GM GREEN 208 6 6
9 191047 KANNAN DEVAN LEAF 250 GM 75 12 12
10 191111 RUNGLEE RUNGLIOT DARJEELING TEA 0 12
11 191129 DUNCON DOUBLE DIAMOND LEAF TEA 0 6
(Continued)
(Continued)
No Item Category :TEA/COFFEE Stock Qty Indent Picked
Code Qty Qty
12 191146 LIPTON GREEN LABEL TEA 100 GM 207 6 6

13 191148 LIPTON TAJMAHAL TEA BAGS 25S 368 24 24
y
14 191150 PRIME TEA 25 BAGS 0 24
15 191151 KOTADA TEA BEARER 250 GM BRAND 0 6
nl
16 191152 KOTADA TEA BEARER BRAND 500 GM 0 6
17 191153 KOTADA CTC SPL DUST 250 GM 0 6
O
18 191154 KOTADA BEARER BRAND TEA 100 GM 0 6
Report No: 0F8 Indent Date: /11/98 (Thursday)
Store No: 863 Store Name: AAA
n
Sub Category : 2301 GROUND COFFEE
io
at
4 190027 COORG FILTER COFFEE 500 GM ul 47 12 12
5 190029 COORG PURE FILTER COFFEE 100 GM 66 18 18
6 190031 COORG PURE FILTER COFFEE 200 GM 83 24 24
irc
7 190034 BROOKE BOND CAFE 100 GM 125 12 12
8 190040 COORG DOUBLE ROAST COFFEE 500 GM 68 12 12
C

11 190043 COORG DOUBLE ROAST COFFEE 500 GM JAR 0 3
12 190065 F.W. PURE FILTER COFFEE 200 GM 0 12
d
13 190066 F.W. PURE FILTER COFFEE 500 GM 0 12

ite
Sub Category: 2302 INSTANT COFFEE

1 190007 BROOKE BOND BRU INSTANT COFFEE 0 18
im
2 190008 BROOKE BOND BRU INSTANT COFFEE 20 271 24 24

3 190010 TATA KAAPI 200 GM 58 6 6
4 190012 BROOKE BOND BRU INST COFFEE 50 381 12 12
rL
5 190014 BRU SUPER STRONG 50 GM 0 24

6 190053 NESCAFE 25G JAR 0 24
7 190056 NESCAFE INST COFFEE 100 GMS JAR 0 24
Fo
8 190060 NESCAFE CLASSIC 50 GMS 254 12 12

9 190064 NESCAFE SELECT INST COFFEE 100 0 12
10 190084 NESCAFE SUNRISE PREMIUM INST 2 291 12 12
11 190088 NESCAFE SUNRISE PREMIUM INST CO 323 12
12 190090 NESTLE SUNRISE PREMIUM 100 GM 0 12
13 190091 NESTLE SUNRISE PREMIUM 100 GM R 0 12
14 500207 NESCAFE 3-IN-ONE 180 GM 0 12
(Continued)
(Continued)
Sub Category: 2303 DUST TEA
1 191020 BROOKE BOND SUPER DUST JAR 500 0 3
2 191037 SARGAM DUST TEA 500 GM 0 3
3 191039 DOUBLE DIAMOND DUST TEA 250 GM 0 3
y
4 191101 SARGAM TEA 250 GM REFIL 0 3
nl
5 191147 LIPTON GREEN LABEL TEA 250 W.B. 0 3
Sub Category: 2304 CTC TEA
O
1 191000 BROOKE BOND RED LABEL 500 GM J 6 6
2 191006 BROOKE BOND 3 ROSES 250 GM J 0 6
n
3 191009 BROOKE BOND 3 ROSE TEA 500 GM 69 6 6
4 191021 BROOKE BOND RED LABEL 250 GM JA 0 6
io
6 191026 BROOKE BOND RED LABEL 250 262 6 6
at
7 191027 BROOKE BOND RED LABEL 100 190 12 12
8 191029 CHAKARAGOLD 100 GM 55 12 12
9
10
191030
191035
CHAKARAGOLD 250 GM
DUNCANS SARGAM TEA JAR GARDEN
ul 59
0
6
6
6
irc
11 191043 KANNAN DEVAN TEA 100 GM GREEN 190 6 6
12 191047 KANNAN DEVAN LEAF 250 GM 51 12 9
13 191052 LIPTON TOPSTAR TEA 100 GM 186 12 12
C
14 191053 LIPTON TOPSTAR TEA 250 GM 168 6 6

15 191057 LIPTON TAJ MAHAL TEA 100 GM 112 12 12
d

ite
19 191065 LIPTON GREEN LABEL TEA 500 GR 1 6

20 191085 LIPTON YELLOW LABEL TEA 500 GR 86 6 6
21 191087 LIPTON YELLOW LABEL TEA 100 GR 190 12 12
im
22 191111 RUNGLEE RUNGLIOT DARJEELING TEA 0 3

23 191129 DUNCON DOUBLE DIAMOND LEAF TEA 0 3
rL
25 191148 LIPTON TAJMAHAL TEA BAGS 25S 242 12 12

26 191150 PRIME TEA 25 BAGS 0 12
27 191151 KOTADA TEA BEARER 250 GM BRAND 0 6
Fo
28 191152 KOTADA TEA BEARER BRAND 500 GM 0 3

29 191153 KOTADA CTC SPL DUST 250 GM 0 6
30 191154 KOTADA BEARER BRAND TEA 100 GM 0 3
31 500218 A V T ASSAM TEA 250 GM JAR 6 6
32 500219 A V T PREMIUM TEA 500 GM 92 6 6
33 500220 A V T PREMIUM TEA 250 GM 92 6 6

Exhibit 11: Ordering Process
FoodWorld Warehouse Product Master (Master Maintenance Menu)

Product Name Cadburys 5 Star Cadburys Crackle Surf Excel Refill
38 gms 35 gms 500 gms
y
Product Code 263607 263573 901039
nl
UOM Gms gms gms
Packout 36 24 6
O
Description Confectionery count lines Confectionery block chocolate Detergent powder
Dept/Group Code 906 905 2602
Tax Code 88 88 88
n
Stock Movement Flag
io
Indent Flag 0 0 0
Prod-Company Code 50 Cadburys 50 Cadburys 125 Hindustan Lever
Vendor Code CRA024 Ashwini Distributors CRA024 Ashwini Distributors CRU034 Upmarkets
at
ABC Code A A A
Power Item N Y N
Base Quantity
Buyer Code
A:240 B:96 C:48 D:420
003
ul A:60 B:36 C:24 D:120
033
A:48 B:24 C:18 D:420
007
irc
Cost Price Rs 8.83 from 24.9.98
WIP: 8.83 Rs 12.37 from 24.9.98 WIP:12.37 Rs 59.28 from 3.12.98
WIP: 59.28
C
MRP 10.00 from 11.2.98 14.00 from 3.9.98 65.00 from 9.5.98
Prev Cost Price 8.89 from 15.9.98 12.46 from 15.9.98 59.58 from 21.11.98
Prev MRP 10.00 from 11.2.98 12.00 from 2.3.98 60.00 from 9.2.98
d
Stock Qty at Warehouse 1046 664 13

ite

im
rL
Fo
Exhibit 12: Sample Purchase Orders
No Product Name Code MRP Qty Cost Value
Hex Trading PO No: 37337
y
1 Brooke Bond 3 Roses 500 gms Jar 191007 101.50 24 96.97 2327.28
nl
2 Brooke Bond Inst Coffee 200 gms 190008 124.00 180 115.89 20860.20
3 Brooke Bond Bru Inst Coffee 50 gms 190012 26.00 120 24.30 2916.00
O
4 Brooke Bond Green Label Coffee 50 gms 190015 8.15 80 7.65 612.00
6 Brooke Bond Red Label 500 gms Jar 191000 86.00 48 81.50 3912.00
n
7 Brooke Bond Red Label Tea 250 gms 191026 43.00 96 41.25 3960.00
8 Brooke Bond Red Label Tea 250 gms Jar 191021 43.00 72 41.25 2970.00
io
10 Lipton Green Label Tea 250 gms 191064 88.25 144 84.38 12150.72
at
11 Lipton Green Label Tea 100 gms 191146 31.50 120 29.98 3597.60
12 Lipton Tajmahal Tea Bags 200s 191145 130.00 18 118.18 2127.24
Dhanalakshmy Agencies
1 Nescafe Select Inst Coffee 50 gms
ul
190060 61.00 224 55.80
PO No 37554
12499.20
irc
2 Nescafe Sunrise Extra 200 gms 190080 100.00 72 93.50 6732.00
3 Nescafe Sunrise Inst Extra 50 gms 190092 27.00 256 25.25 6464.00
C
M S Vel & Company PO No 37810

1 Coorg Double Roast Coffee 500 gms 190040 74.00 96.00 74.00 7104.00
d
3 Coorg Filter Coffee 500 gms 190027 83.00 72.00 83.00 5976.00
ite
4 Coorg Pure Filter Coffee 100 gms 190029 17.00 180.00 17.00 3060.00

im
rL
Fo
Exhibit 13: Sample Goods Receipt Notes
No Product Name Code MRP Qty Cost Value
Hex Trading 8201 DC No: 8201
y
1 Brooke Bond Bru Inst Coffee 200 gms 190008 124.00 180.00 115.89 20860.20
nl
2 Brooke Bond Bru Inst Coffee 50 gms 190012 26.00 120.00 24.30 2916.00
3 Brooke Bond Green Label Coffee 50 gms 190015 8.15 80.00 7.65 612.00
O
5 Brooke Bond Red Label 500 gms Jar 191000 86.00 48.00 81.50 3912.00
6 Brooke Bond Red Label Tea 250 gms 191026 43.00 96.00 41.25 3960.00
n
7 Brooke Bond Red Label Tea 250 gms Jar 191021 43.00 72.00 41.25 2970.00
io
9 Lipton Green Label Tea 250 gms 191064 88.25 144.00 84.38 12150.72
10 Lipton Green Label Tea 100 gms 191146 31.50 120.00 29.98 3597.60
at
11 Lipton Tajmahal Tea Bags 200s 191145 130.00 18.00 118.18 2127.24
Dhanalakshmi Agencies 8205 DC No 8205
1
2
Nescafe Select Inst Coffee 50 gms
Nescafe Sunrise Extra 200 gms
ul
190060
190080
61.00
100.00
224.00
72.00
55.80
93.50
12499.20
6732.00
irc
3 Nescafe Sunrise Inst Extra 50 gms 190092 27.00 256.00 25.25 6464.00
M S Vel & Company 8207 DC No 8207

C
3 Coorg Filter Coffee 500 gms 190027 83.00 72.00 83.00 5976.00
d
4 Coorg Pure Filter Coffee 100 gms 190029 17.00 180.00 17.00 3060.00
ite

im
rL
Fo
Exhibit 14: List of Top Vendors for a Sample Month
y
nl
No Vendor Name No of Total Purchase Company Name Purchase Cumu-
Skus Hash Value from Value lative
Qty Vendor % %
O
1 HEX TRADING 1076 94381 2235908.67 HLL 10.43% 10.43%
2 TAMIL NADU STATE MARKETING CORPN LTD 112 17220 1564409.00 TASMAC 7.30% 17.72%
n
3 VENKATESWARA AGENCIES 61 49461 716526.40 CADBURYS 3.34% 21.07%
4 DHANALAKSHMI AGENCIES 96 14541 524534.02 NESTLE CULNERY 2.45% 23.51%
io
5 ARIHANT AGENCIES 128 52775 518428.91 BRITANNIA 2.42% 25.93%
6 GAJALAKSHMI AGENCIES 34 33737 479216.90 NESTLE CONFECTIONERY 2.24% 28.17%
at
7 SATHYA FRUITS 18 18 474801.80 FRUITS 2.21% 30.38%
8 FRESH N GREEN 37 37 398832.05 VEGETABLE 1.86% 32.24%

9 MAHANTH ENTERPRISES 22 5395 396753.23 ITC 1.85% 34.09%
ul
10 DEW CONSUMER PRODUCTS & SERVICES LTD 54 54 377733.51 BAKERY 1.76% 35.85%
11 SRI GANESH AGENCIES 26 13190 377540.90 AAVIN 1.76% 37.61%
irc
12 BHAWAR SALES CORPORATION 59 6470 353170.10 P&G 1.65% 39.26%
13 J J UDYOG 4 12100 326025.00 DAL, JALGAON 1.52% 40.78%
14 V R MUTHU & BROTHERS 11 5092 301768.00 V V S IDAYAM 1.41% 42.19%
C
15 BHIKSHU MARKETING 87 11513 299039.89 RCI 1.39% 43.58%
16 E I D PARRY (INDIA) LTD 4 20000 276000.00 PARRY - SUGAR 1.29% 44.87%
17 SAKTHI SOFT DRINKS LTD 17 4425 248985.05 COCA COLA INDIA 1.16% 46.03%
d
18 AKASH MARKETING CENTRE 41 14132 242436.90 GUJARAT CO-OP. MILK AMUL 1.13% 47.16%
ite
19 MAC MARKETING COMPANY 44 4797 237024.90 NESTLE - IMPORTED 1.11% 48.27%
20 SRI JEET TRADERS 65 19503 229731.90 DRY FRUITS, MASALAS, WHOLESALE 1.07% 49.34%
21 SURESH ENTERPRISES 31 8993 218623.37 KELLOGGS 1.02% 50.36%
im
22 SANDHYA AGENCIES 54 4037 218378.52 ISPL 1.02% 51.38%

23 SRI SAI ENTERPRISES 118 7351 218004.92 JOHNSON & JOHNSON 1.02% 52.39%
24 GODREJ PILLSBURY LTD 25 8540 216546.90 GODREJ PILLSBURY 1.01% 53.40%
rL
25 K AHAMED HUSSAIN & SONS 40 9282 214779.84 SKB 1.00% 54.41%

26 ASSOCIATED AGENCIES 320 5935 202409.19 HARDWARE 0.94% 55.35%
27 SHREE MANISH & CO 79 10368 195430.74 COLGATE PALMOLIVE 0.91% 56.26%
485
Fo
28 CHANG FOODS PVT LTD 40 40 188287.55 OLD CHANG KEE 0.88% 57.14%
29 SRIJEET ENTERPRISES 47 14488 180802.85 DALS, DRY FRUITS 0.84% 57.98%
(Continued)
(Continued)
y
486
nl
Qty Vendor % %
30 K NARAYANASAMI 12 12 165758.40 VEGETABLE 0.77% 58.76%
O
31 PANKAJ TRADERS 31 10071 157127.50 DALS 0.73% 59.49%
32 AGRI FLORA 24 24 151176.30 EXOTIC F&V 0.71% 60.19%

33 M S VEL & COMPANY 47 4363 147066.10 TATA TEA 0.69% 60.88%
n
34 AMUDHAN AGENCIES & SERVICES PVT LTD 150 1834 146564.95 MODI REVLON 0.68% 61.56%
io
35 FATHIMA AGENCIES PVT LTD 84 7933 135679.05 PONDICHERRY MINERAL WATER 0.63% 62.20%
36 SERVICE UNLIMITED 41 12186 133302.15 CHILLY, DANIYA 0.62% 62.82%
at
39 AMALGAM FOODS LIMITED 113 1523 126132.75 SUMERU AMALGAM FOODS LTD 0.59% 64.60%
40 SALECHA INDUSTRIES 1 4900 124950.00 DALS, 0.58% 65.18%
41 RAJAS MARKETING CO 38 6098 121382.00 HENKEL SPIC 0.57% 65.74%
ul
42 SREE SUBHA MILK PRODUCTS 75 7949 118424.00 DAIRY PRODUCTS 0.55% 66.30%
43 VENKAT 1 3000 112050.00 FRUITS & VEG. 0.52% 66.82%
irc
44 THE NILGIRI DAIRY FARM LTD 15 15 111299.60 DAIRY PRODUCTS, NILGIRIS 0.52% 67.34%
45 SRI SINGHI SPICES PVT LTD 12 2084 110940.00 SATNAM OVERSEAS, KOHINOOR BAS 0.52% 67.86%
46 BULLWORK TRADERS 29 9942 105224.40 JAGGERY, TAMARIND, 0.49% 68.35%
C
47 NAVEEN INDIA 50 6129 99579.34 BEST FOOD INTERNATIONAL, CPC 0.46% 68.81%
48 FIVE STAR AGENCIES 41 13859 98799.65 NUTRINE 0.46% 69.27%
49 BALAJEE ASSOCIATES 23 23 98188.60 BAKERY 0.46% 69.73%
d
50 CORAL ENTERPRISES 19 8190 91231.00 CHILLY, DANIA, PEPPER, DALS 0.43% 70.16%
ite
51 BAWAR SALES CORPORATION 15 2154 91190.95 P&G 0.43% 70.58%
52 SIL AGENCIES 67 2607 90081.39 KAYTIS, COSTAS, PRUTINA-
TIN VEG & NON,VEG, SUZAANE 0.42% 71.00%
im
53 KAMAL STORES 103 8100 89891.00 SAVORIES, UNBRANDED 0.42% 71.42%

54 PAZHA MUDIR CHOLAI 6 6 88077.20 F&V 0.41% 71.83%
55 TOBACCO CENTRE 43 3170 83941.63 ITC 0.39% 72.22%
rL
56 RAAJ TRADE LINGHS 89 2911 82943.15 PARK AVENUE, J K HELEN CURTIS 0.39% 72.61%
57 ASHOK M.LULLA 33 33 81501.15 NEW GANGODHRI, CHAT ITEMS 0.38% 72.99%
58 NEW BHARATH ENTERPRISES 29 1964 80697.57 DABUR INDIA 0.38% 73.37%
59 JAIN MARKETING 70 3673 78293.68 HALDIRAM NAGPUR 0.37% 73.73%
Fo
60 SAROJ ENTERPRISES 24 6134 76111.66 M T R FOODS 0.35% 74.09%

y
nl
(Continued)
O
Qty Vendor % %
n
61 SRI SARAVAN ENTERPRISES 11 1586 75693.20 GOODNIGHT., GODREJ HI CARE 0.35% 74.44%
io
62 N. SURESH BABU 2 7500 74962.50 IDLI RICE 0.35% 74.79%
63 S K SWAMY ENTERPRISES 27 5029 74962.39 KARNATAKA SOAPS, MYSORE SOAPS 0.35% 75.14%
Total of Top 63 4106 596847 16110360.74
at
Others 3697 283168 5330701.33

Grand Total 7803* 880015 21441062.07
ul
* The total number of SKUs reflect many which have been discontinued, but have not yet been taken off from the database
irc
C
d
ite
im
rL
487
Fo
Exhibit 15: Hindustan Lever Limited: Sales Profile – 1997
(Rs Crores)
Sl No Items Amount
y
1 Soaps & Detergents 3360
nl
2 Beverages 1544
3 Personal Products 884
O
4 Others 613
5 Processed Triglycerides Oils and Vanaspathi 542
6 Animal Feeding Stuffs 272
n
7 Speciality Chemicals 155
8 Ice Cream & Frozen Desserts 153
io
9 Branded Staple Foods 112
10 Canned & Processed Fruits & Vegetables 97
at
11 Diary Products 87
Total 7820
ul
Source: The Economic Times, December 30, 1999.
Branded Food Processing (1996)

irc
(Rs Crores)
Company Amount
C
Hindustan Lever Limited 2,000

NDDB 1,200
d
Nestle 1,000
Siel 850
ite
Britannia 650
Total (top five) 6,000
Cadbury 250
im
Total (top twenty) 10,000
Source: Compiled FAIDA report, CII & McKinsey and Co, 1997.
rL
Fo
Exhibit 16: Role of Intermediaries: Comparison between India and United States of America
India United States of America
y
Intermediaries between Farmer and 6 (Consolidator, Commission agent, Trader, 2 (Wholesaler, Retailer)
Consumer (Fruits and Vegetable) Commission agent, Wholesaler, Retailer)
nl
Intermediaries between Farmer 2 (Commission agent, Grain trader) 1 (Grain trader)
and Mill (Wheat)
O
Markup 3350% 9%
Farm-gate prices as share of consumer prices
Apples 30% 40%
Tomatoes 25% 41%
n
Milk 90% NA
io
Share of intermediary margin contributing to 50% 80%
value addition costs
at
Milk co-operatives in India have demonstrated that by reducing the number of intermediaries, the farmers share of revenues can be increased
from 50% to over 90% of the processor price.
Source: Compiled from FAIDA Report, CII & McKinsey and Co, 1997 ul
Exhibit 17: Ponni Rice Sales and Purchase for Madras Region
irc
Month Sales Sales Selling Purchase Purchase Net Cost Margin Absolute
Value Price Value Margin
C
(kgs) (Rs) (Rs/kg) (kgs) (Rs) (Rs/kg) (%) (Rs)
Jan 97 5572 109783 19.70 5100 74573 15.72 20.20 22191.16

d
Feb 97 4689 92703 19.77 4900 71894 15.78 20.20 18710.58

Mar 97 6147 118685 19.31 6000 88022 15.77 18.30 21746.81
ite
Apr 97 5925 114353 19.30 5800 85313 15.82 18.05 20619.00

May 97 6974 129166 18.52 7100 100159 15.17 18.10 23370.42
June 97 10884 195011 17.92 11000 150300 14.69 18.00 35125.04
im
July 97 14082 265823 18.88 14000 201290 15.46 18.10 48115.28

Aug 97 13368 253456 18.96 13500 195194 15.55 18.00 45583.60
Sept 97 10192 192817 18.92 9900 141959 15.42 18.50 35656.36
rL
Nov 97 12022 201200 16.74 12000 153154 13.72 18.00 36258.16
Jan 98 11444 204414 17.86 11500 154357 14.43 19.20 39277.08

Feb 98 9848 176193 17.89 9000 121297 14.49 19.00 33495.48
Fo
Mar 98 9850 174899 17.76 10500 142180 14.56 18.00 31483.00

Apr 98 9314 165888 17.81 9600 131138 14.69 17.53 29065.34
May 98 13225 242520 18.34 13500 185085 14.74 19.61 47583.50
June 98 22959 388368 16.92 24000 319037 14.29 15.50 60283.89
July 98 69840 1098688 15.73 82500 1043625 13.73 12.72 139784.80
Aug 98 95760 1437846 15.02 150000 1891500 13.43 10.54 151789.20
Sept 98 106330 1603653 15.08 290000 3787400 13.55 10.15 162881.50
Oct 98 120000 2035054 16.96 320000 4848000 14.86 12.40 251854.00
Exhibit 18: Expected Profile of “Value Added” Food Consumption in India by 2005
(Rs Crores)
Sl No Items Amount
y
1 Oil 50,000
2 Packaged Milk 36,000
nl
3 Fresh Poultry 27,000
4 Sugar 24,000
O
5 Packaged Atta 15,000
6 Soft Drinks 10,500
7 Bakery 10,000
n
8 Cereals 10,000
9 Processed Meat and Poultry 9,000
io
10 Tea and Coffee 7,400
11 Indian Dairy Products 7,300
at
12 Confectionery 6,500
13 Value Added Western Diary Products 4,700
14
15
Fruit Drinks
Fresh Vegetables
ul 2,000
1,200
irc
16 Spices 1,200
17 Puree, Jams and Sauces 1,000
18 Frozen Vegetables 350
C
Total value added food consumption in 2005 (Approximately) 225,000

Total value added food consumption in 1996 (Approximately) 77,000
Total food consumption in 1996 250,000
d
Source: Compiled from FAIDA Report, CII & McKinsey and Co, 1997.
ite
im
rL
Fo
Exhibit 19: Economics of a Typical Store
Size : 3500 Sq Ft
Investment : Rs 40 Lakhs/Fitout
Rs 10 Lakhs/Rent Deposit
y
Rs 50 Lakhs
nl
Costs Rs
Rent 90,000
O
Salaries 80,000
Power/Fuel 50,000
Selling Exp 15,000
n
Bank Charges 10,000
Security/Others 10,000
io
Repair/Maintenance 20,000
Advt/Promotion 15,000
at
TOTAL 3,00,000
Depreciation 50,000
Interest on W/Cap ul 5,000
TOTAL OPEX 3,55,000
irc
C
d
The key issues to be dealt with are the following:

ite
1. The merchandising decision i.e. what to stock in each retail location?

2. The vendor development decision i.e. whom and where to source from?
im
3. The best distribution strategy, i.e. how to reduce total delivery time from vendor to
store, and reduction in the cost of procurement?
rL
Fo
ABC analysis (pareto analysis) and category management can help to manage variety better
and to make a better merchandising decision. In the organised segment, the approach to
vendor development can be through supplier relationship management in a variety of ways,
including collaborative marketing efforts and lobbying for a single point of contact at the
supplier (especially with large suppliers). In the unorganised sector, traditional vendor
development can be done.
The current ordering process and inventory norms can be assessed for cost effectiveness,
from the data given. The role of the warehouse in effectively servicing the retail outlets can
be examined.
Different supply chain structures are already identified for the different types of
products. Each of these needs to be evaluated for improvement possibilities.
y
nl
O
n
io
at
ul
irc
C
d
ite
im
rL
Fo
CASE 10
y
nl
O
Seth Dhaniram is a carrying and forwarding agent (C&FA). The case deals with his
company, Raj Distribution Services (RDS), which provides carrying and forwarding services
to JOSH Denims, a leading manufacturer and marketer of denim garments. A year after the
n
initial arrangement, an entire tier in the JOSH’s supply chain was eliminated and the
C&FA’s direct customers exploded from 10 distributors to 180 retailers. This change had
io
direct implications on the cost and complexity of the C&FA operation. Dhaniram was left
with the option of either continuing the agency operation with some renegotiation, or
at
giving it up.
This case highlights the significance of each channel partner in the supply chain, and also
ul
emphasizes the implications of eliminating a link in the supply chain to the other channel
partners. It also introduces some subtle aspects like family owned businesses, the role of
irc
networking, and the value-added by transporters.
C
SETH DHANIRAM C&FA

d
ite
July 1999
Seth Dhaniram was a worried man. As he looked out at the gathering monsoon clouds from
im
his plush office in the commercial district of Ahmedabad, his thoughts went back to the latest
problem that one of the businesses in his group faced. Since the past few weeks, he had been
pondering over whether to continue with the carrying and forwarding agency (C&FA)
rL
operations of “JOSH Denims,” or give it up.

Fo
Prepared by Chudasama and G Raghuram. Teaching Material of the Indian Institute of Management,
Institute of Management Ahmedabad.
Background
Dhaniram was at the helm of a family owned business, which was over a 100 years old. It was
jointly owned and operated by his brothers, sons and nephews. The organisation, “Logistics
Solutions” provided logistics services to both domestic and multinational corporations.
These services included carrying and forwarding (Glossary), distribution and stocking. The
firm offered its services to organizations in industries as diverse as pharmaceuticals, processed
foods, fast moving consumer goods (FMCG), consumer durables and readymade garments
Fo
(RMG). The C&FA operation of each customer was registered as a separate agency and each
agency was treated as a distinct business entity, mainly for accounting purposes. Dhaniram
had grouped the various agencies based on their synergies (Exhibit 1).
rL
These groups of agencies had been allocated among various family members based on
im their skill sets and experience. For example, his younger brother, Sakharam solely looked
after the edible oil agencies, while his sons Raj (BSc (Chemistry)) and Shyam (MCom
(Accountancy)) jointly handled the pharmaceutical agencies. However, the various agencies
fell under the umbrella of the parent company Logistics Solutions headed by Dhaniram.
ite
Over the years, the firm had developed an excellent portfolio of agencies by roping in
d
market leaders in pharmaceuticals, processed foods and FMCG. In fact, one of the market
leaders in processed foods, a Swiss multinational, had benefited from the firm’s services for
over 60 years. The company had a focused approach and over the years had (unlike other
C
firms in the C&FA/distribution business) restrained from expending its limited resources
into allied businesses like warehousing and transportation.
In spite of its long-standing reputation in the market and financial soundness, the
irc
working capital requirements of Logistics Solutions were financed by personal sources such
as borrowings from relatives and business associates. This practice was not unusual in the
Ahmedabad market where the average businessman was known for his dexterity at circulating
ul
capital effectively in the marketplace. The number of working capital turns/year for trading
firms could be as high as 36 to 40. This was essentially achieved because of high inventory
at
turnover resulting from low order cycle times and near zero ordering costs. In effect, as
io
Dhaniram put it, “judicious circulation of working capital could yield high rates of return.”
n
Since the mid-nineties, Dhaniram had withdrawn from day to day operations and had
restricted his role to that of providing strategic direction and grooming the next generation
to take over the reins of the business. His nephews, Amit and Akhilesh had returned from the
O
United States with MBA degrees in finance and were being trained regarding the financial
aspects of the Logistics Solutions operation.
nl
JOSH Denims and the C&FA Appointment
y
In January 1998, Dhaniram gathered from the grapevine that JOSH Denims (JOSH), a
market leader in denim garments was in search of a C&FA to administer the distribution of
its products in the state of Gujarat (Exhibit 2). Dhaniram was aware that JOSH conducted
its distribution operation from a company owned and operated facility based in Ahmedabad.
Case 10: Seth Dhaniram C&FA 495
The appointment of a C&FA was a part of the company’s ongoing policy of outsourcing
several key functions.
Dhaniram soon saw an advertisement in the local newspapers calling for applications from
experienced parties who would be interested in providing the C&FA services to JOSH. He
also received a phone call from an executive at JOSH, asking him to contact the General
Manager (Distribution) since JOSH wanted to consider Logistics Solutions as a possible
candidate. In the meantime, Dhaniram gathered that JOSH had contacted other similar
Fo
agencies and their major distributors who were well connected in the market. He knew that
this was a general practice followed by principals like JOSH.
Dhaniram decided to send in his organisation’s profile, along with an expression of
rL
interest in JOSH’s agency. A month later some of the executives from JOSH visited the
premises of Logistics Solutions and discussed issues related to their existing profile, services
im
and infrastructure. They said that such visits were being carried out to a short listed set of
applicants, primarily based on the following factors:
ü Market reputation
ite
ü Financial soundness
d
ü Infrastructure/facilities
ü Professionalism C
ü Techno friendliness
Subsequent to this visit, Dhaniram was invited to the corporate office of JOSH to discuss
the expected service levels and his capability to fulfill them. This meeting was attended by
irc
several key executives from the Distribution, Finance and Marketing operations of JOSH.
He came to know that there were four other candidates in the final list. Knowing that this
process was characteristic of how the market typically operated regarding appointment of a
ul
new C&FA, and given his reputation and networking skills, Dhaniram was sure that he was
one of the serious contenders. He decided to give his best, since he also valued this account.
at
There were a few more rounds of negotiations. Dhaniram was elated about the prospect of
io
handling a completely new, highly “sophisticated” product. The product range included
n
designer denim trousers and jackets with variations in colors, sizes, and style including
position of zippers and number/type of pockets. The role of the C&FA essentially involved
making supplies to 10 distributors, who in turn would cater to 180 company approved retail
O
outlets (Exhibit 3). Each major district had one distributor (Exhibit 4) and several retail
outlets.
The job profile of the C&FA involved the following:
nl
1. Warehousing
y
2. Order processing, invoicing, dispatching, transportation and tracking of goods
3. Collection of payments and acknowledging receipt of payments
4. Reordering inventory
5. Handling goods returned (reverse logistics)

6. Providing sales reports to the field force
7. Dealing with taxation and regulations (Octroi, Sales Tax & FDA)
8. While doing all of the above, adherence to the guidelines set by the principal
The remuneration, on a fixed package basis, was attractive at INR 60,000/month. This
amounted to a margin of nearly 2% on sales approximating INR 3 million/month.
Fo
The C&FA contract was ultimately awarded to Dhaniram. He was to begin operations in
April 1998. The contract stated the terms and conditions for the C&FA operation and would
be periodically reviewed every two years. It was orally understood between him and the
rL
JOSH executives that the periodic review of the contract would more or less be a
renegotiation of the remuneration, as most of the other JOSH’s terms and conditions would
im
remain stable.
The C&FA Experience

ite
The new C&FA was registered as Raj Distribution Services (RDS) and the operations began
d
at a leased warehouse in April 1998. The title of the warehouse was owned by RDS.
The transportation function was outsourced by RDS to several private carriers. A
C
preferred carrier was nominated for every destination depending upon i) distributor’s
recommendation and ii) carrier performance/service level. The service level of a carrier for a
particular destination was a function of the frequency of vehicles departing to that
destination. The distributor’s recommendation was normally accepted unless the
irc
recommended carrier quoted high freight rates or failed to meet Logistics Solutions’
minimum standards in terms of the service level, reputation and age of the fleet.
ul
RDS had a dedicated vehicle to ship the customer orders from the warehouse to the
carrier’s dock. The carrier had a unique role of consolidating shipments at both ends. Firstly,
it consolidated the shipments of several customers shipping to a particular destination.
at
Secondly, at the destination, it consolidated the shipments for a distributor from his/her
io
several suppliers. For example, if distributor ABC located in Surat had shipments from 15
n
suppliers arriving at the carrier’s dock on a given day, the carrier would consolidate all of
ABC’s shipments that arrived from several C&FAs/stockists before making a single door
delivery to ABC. This system was facilitated by the fact that RDS and other C&FAs in the
O
market usually appointed a distributor recommended carrier. However, inspite of
consolidation, almost all the shipments were less than truck load.
The average number of cases sent to a distributor was about 250 per month. Exhibit 4
nl
provides a list of the destinations serviced by carriers from Ahmedabad, along with the
y
freight rate and share. The transporters followed a system under which the greater the freight
volume for a destination, lower the transportation rate/case. The rates were also dependent
on the distance. The rates were negotiated annually for each destination based upon the
freight volume in the previous year.
JOSH extended a 21 day credit to the distributors. However it was the C&FA’s
responsibility to receive payment on behalf of JOSH and deposit the checks in JOSH’s bank
account. There were minimum order quantity norms and distributors could not place orders
for less than case lot quantities. (One case lot contained 12 trousers or jackets, and generally
of the same size and style).
Exhibit 5 summarizes the distribution of costs incurred at the C&FA between the C&FA
(RDS) and the principal (JOSH). There were marginal requirements of repackaging some
Fo
cases and sending some cases using a courier service.
Dhaniram found the task challenging and interesting. This agency was proving to be a
good training ground for his nephews. He expected that their business could leverage the
rL
experience with this account for further growth.
im
Policy Change
Within 15 months of operations, in a major policy change, JOSH decided to eliminate an
entire tier in its distribution structure by doing away with all its distributors. In July 1999,
ite
RDS was instructed to make direct supplies to all the 180 retail outlets, with no additional
d
remuneration. This was to begin as soon as possible and definitely by the end of the month.
Dhaniram was not exactly pleased with this development. Apart from being a surprise, the
policy change would have manifold consequences for RDS.
C
The retail outlets were spread in 14 districts (four more than the ten distributor locations).
These outlets were located in the major town (and some of the outlying suburbs) of each of
the districts. Usually a retail outlet sourced its requirements from the distributor located in
irc
its district. If a retail outlet was located in a district that did not have a distributor, then it
would approach the nearest distributor in another district. For example, a retail outlet
located in Amreli would source its inventory from either Rajkot or Bhavnagar. Now the
ul
expectation was that the retail outlets would source directly from the C&FA.
at
Retail outlets were currently permitted to order in less than case lot quantities. Based on
some data shared in one of the recent distributor meets, Dhaniram had estimated that the
io
total outbound shipments from the distributors was about 3,600 cartons, two-thirds of
n
which were a result of repackaging. The repackaging was necessitated both due to order
quantities and mix requirements. Repackaging requirements were more for the smaller
retailers. The number of carton requirements per retailer ranged between 10 and 35 per
O
month, with 50% of the retailers requiring 15 or less cartons. Distributors had to maintain
inventory records both at the piece level (in the case of repackaging) and the case lot level.
Apart from the repackaging activity, Dhaniram also visualised increasing complexity in
nl
organising the transport and delivery, credit and collection management, and customer
y
servicing including query handling. Additional labour and clerical staff would be required.
Distributors also took care of transportation to the retailers, who were mostly within a
radius of 25 kms of the distributor location. In terms of credit, retailers generally paid for the
previous order when the next order was delivered. There were instances, however, when the
credit went up to a month based on the relationship. For the various services including
transportation and ownership of the inventory rendered by the distributor, they were given a
trade margin/discount ranging from 7% to 10% of purchase value. Given JOSH’s price list,
retailers enjoyed a margin of 15% to 25% of the sales value.
The Decision
Dhaniram was in a quandary whether to continue with the agency operation or give it up.
Fo
Though financially the project was no more attractive, JOSH was a leading brand in the
market and operating its C&F was by itself a very prestigious affair. Renegotiation was a
possibility, but seemed farfetched. Several agents with experience in logistics in Ahmedabad
rL
were eager to take up the agency operations. His other hope was to negotiate a higher
remuneration when the contract came up for periodic review after about eight months.
im
Dhaniram felt to himself, “JOSH was saving a tidy amount by bringing about the proposed
changes in its supply chain. Did he, as a channel partner, not deserve a share of the savings?”
Dhaniram was wondering what to do.
ite
Glossary of Terms
d
1 BSc Bachelor of Science
2 C&FA
C
Carrying and forwarding agent one who does not assume the title of goods, but distributes them
and collects payments on behalf of the manufacturer
3 FDA Food and Drugs Administration
4 INR Indian Rupee
irc
5 MCom Master of Commerce
6 Octroi A tax levied by the local government on all products entering the municipal limits of a city
7 Principal A term which is the equivalent of a manufacturer
ul
8 Sales Tax A tax levied by the state government on every sale that takes place within the physical boundaries
of the state
at
9 Seth An honorable prefix used for a businessman
io
Exhibit 1: The Logistics Solutions Organisation Structure
n
Logistics Solution
O
Pharmaceutical Edible Oils Readymade Consumer FMCG
Group Group Garments Group Durables Group Group
nl
Agency 1 Agency 1 Agency 1 Agency 1 Agency 1
y
Agency 2 Agency 2 Agency 2 Agency 2
Agency 3 Agency 3 Agency 3
Agency 4
Note: Agency: Carrying and Forwarding Agency

Fo Exhibit 2: The District Map of Gujarat
rL
im
ite
d
C
irc
ul
at
Case 10: Seth Dhaniram C&FA
io
n
O
499
nl
y
Exhibit 3: The JOSH Denims Supply Chain
Tire 2 Tire 1 Tire 1 Tire 2 Tire 3

Suppliers Suppliers Customers Customers Customers
Supplier Distributor Retail Outlet End User (Consumer)

Focal Company:
JOSH Denims,
Supplier Raj Distribution Distributor Retail Outlet End User (Consumer)
Ahmedabad, India
Fo
ServicesC&FA

rL
Retail Outlet End User (Consumer)
im
Distributor
Retail Outlet
ite
d
C
Exhibit 4: List of distribution Locations
Distination Distancw Preferred Carrier Rate/Case (INR) % of Freight Share

irc
Anand 65 Suraj Carrying 12 7
Bharuch 200 Surya Roadlines 18 3
Bhavnagar 200 Raj Carrying Corporation 18 3
ul
Gandhinagar 30 Vashram Transport 16 5
Jamnagar 300 Raj Carrying Corporation 20 5
at
Navsari 300 Suraj Carrying 20 8
io
Rajkot 220 Raj Carrying Corporation 16 12
Surat 260 Surya Roadlines 14 19
Vadodara 120 Surya Roadlines
n 12 18
The tenth destination was Ahmedabad. The retailers were serviced at an average of INR
8/case
O
Note: INR: Indian Rupees
All transit times were overnight
nl
y
Exhibit 5: Costs Incurred by the C&FA (RDS) and the Principal (JOSH Denims)
RDS:
Remuneration per month: 60,000 INR
Costs
Item INR
Warehousing 12,000
Fo
Manpower (1 manager, 1 store keeper, 1 computer operator, 2 packer) 18,000
Transportation from warehouse to the carriers dock (60 trips/month * 60 INR) 3,600
Electricity 1,000
rL
Computer and related hardware (one time fixed expense) 70,000
JOSH Denims:
im
Costs
Item INR
ite
Transportation from Ahmedabad (C&FA) to distributor destination 34,000
Repackaging 3,000
Security
d 2,000
Courier 3,000
C
Consultancy and all other charges related to sales tax and other levies 1,000
Service tax (5% of the C&FA remuneration) 3,000
irc
ul
1. What is the estimated cost incurred by RDS due to JOSH’s changed policy?
at
2. Can the company still get advantages of economies of scale in the new arrangement?
io
3. Is it viable for RDS to still continue with the relationship with JOSH?
n
O
The present cost for distributing the goods to the 10 distributors can be calculated. In the
nl
new arrangement, there would be additional costs due to re-packaging and administrative
costs of dealing with many more downstream entities. Different options can be explored
y
in the distribution network to see if any partial aggregation can help. Credit arrangements
and cash implications for RDS also need to be considered.
These cost implications have to be considered in conjunction with Seth Dhaniram’s
overall business portfolio and the importance of companies like J
CASE 11
y
nl
The vision of AFL was to be the acknowledged leader in providing world class “Integrated
O
Logistics Solutions” (ILS). The company had the background of a family business involved
in clearing and forwarding, dating from 1867. AFL came formally into existence in 1945. In
n
1979, a tie up with DHL Worldwide Express permitted AFL to introduce India to the
concept of “Express Delivery”, giving birth to the term “The Courier”. In 1988, AFL began
io
a new untried service: Express Distribution, which involved providing distribution services
with time guarantees, thus adding more value to its already existing niche services.
at
With its wealth of experience and expertise, AFL decided to enter a new growth phase,
towards becoming the ILS provider to corporate India within the next three years. In this
ul
context, the chairman and managing director of AFL was further concerned with building
the appropriate information technology and physical infrastructure to enable this to
irc
happen. The question that he had was ‘how will we do it?’
C
AIRFREIGHT LIMITED
d
ite
Mr. Cyrus Guzder, Chairman and Managing Director, Airfreight Limited (AFL), in an
interview with the case writers in April 1997, was verbalising the implications of his vision
for AFL, which was to be the acknowledged leader in providing world class “Integrated
im
Logistics Solutions” (ILS). The company had the background of a family business involved in
clearing and forwarding, dating from 1867. The current generation prided itself in being the
fifth in the line. Mr. Guzder was concerned with strengthening company’s leadership
rL
position. Based on the company’s primary strength of “mastery over movement” and the
changing customer needs, he was quite clear that the future direction ought to be in
Fo
Written by S. Manikutty and G. Raghuram. Assistance provided by Mr. Sunil Jaryal is acknowledged.
This case was made possible by funding from the Research and Publications committee. Teaching
material of the Indian Institute of Management, Ahmedabad, is prepared as a basis for class discussion.
Cases are not designed to present illustrations of either correct or incorrect handling of administrative
problems. Copyright © 1998 by the Indian Institute of Management, Ahmedabad.
Case 11: Airfreight Limited 503
providing ILS, but felt that the market needed to be appropriately developed and accessed.
In this context, he was further concerned with building the appropriate infrastructure to
enable this to happen, namely information technology infrastructure and physical
infrastructure. The repeated question that he had was ‘how will we do it ?’
He was also clear about the framework within which this ‘how’ would be set and had
y
already expounded a statement of values for AFL. Exhibit 1 gives the vision, mission and
nl
values of AFL.
The Organisation
O
AFL came formally into existence in 1945, when Tata Airlines appointed its founder Jamshed
N. Guzder, as its first sole cargo agent in India. This started a long association with “freight
n
forwarding”. Following closely thereafter, in 1948, its travel division was set up. In 1979, a
io
tie up with DHL Worldwide Express permitted AFL to introduce India to the concept of
“Express Delivery”, giving birth to the term “The Courier”. In 1988, AFL began a new
at
untried service - Express Distribution, which involved providing distribution services with
time guarantees, thus adding more value to its already existing niche services. A new activity
ul
that AFL got into in 1994 was international money transfer in an express mode, in
collaboration with Western Union, a US based international giant in this service. With its
wealth of experience and expertise, AFL decided to enter a new growth phase, towards
irc
becoming the “Integrated Logistics Solutions Provider” to corporate India by the year 2000
AD.
C
AFL had the following business divisions and subsidiaries:

Business Divisions
DHL (International Express)
d
ACE (Domestic Express Distribution)

ite
Air Cargo
Indtravels (Travel and Tours)
im
Subsidiaries
Astra Shipping
rL
Astra Infotech
Astra Worldwide
The original beginnings of the freight forwarding service was in the cargo division,
Fo
followed by the travel division. The international express division provided the courier
service. While international courier service was the mainstay of this division, domestic
courier service was also provided with the same infrastructure. The DHL brand was used for
international while Airfreight brand was used for domestic. The domestic express
distribution service was offered through ACE (Advanced Cargo Expertise), using surface
transport. Astra Shipping was involved in shipping focused primarily on the ocean freight
business as a non vessel owning common carrier (NVOCC). Astra Infotech was involved in
developing software products related to the transport sector. Astra Worldwide was in
merchant exports, primarily focused on leather shoes and engineering valves. This subsidiary
was set up in 1990, primarily for the then needed foreign exchange cover and tax benefits.
The subsidiaries had marginal, but complementary business activities.
The corporate support functions for the business divisions were finance, information
y
technology, human resources, audit and methods, corporate relations, planning, and
nl
integrated logistics solutions. The integrated logistics function put all the business divisions
together with the subsidiaries and support functions for marketing integrated logistics
solutions.
O
Exhibit 2 gives the organisation structure of AFL. Similar support functions existed at
regional levels for each of the business divisions, which were run as profit centres. Until
n
March 1995, at the regional level, these support functions were available only as a common
resource for the various business divisions until a restructuring exercise was carried in 1995.
io
As per some of the executives heading the business divisions/corporate functions, the areas of
operations tended to overlap, sometimes resulting in internal confusion and extra need for
at
coordination.
While AFL was organised into business divisions and functions as given above in terms of
ul
man power and resource allocation, it described its products as:
DHL Worldwide Express
irc
AFL Domestic Express
AFL Cargo
C
AFL Travels
AFL Distribution Services
d
AFL Logistics
AFL Shipping
ite
AFL Infotech
The above structuring reflected the “product” focus that AFL was continuously striving
im
for, in its evolution. Exhibit 3 gives a brief business profile of the above products.
Exhibit 4 gives the growth in turnover, net worth and number of employees of AFL from
1990-91 till 1996-97 and the growth in turnover and revenue for the four AFL divisions.
rL
Exhibit 5 gives the financial highlights of the company over the same period. Exhibit 6 gives
the profit and loss account for the years 1994-95 and 1995-96. Exhibit 7 gives the balance
sheet for the years 1994-95 and 1995-96.
Fo
Some of the important achievements until 1995-96 were that the turnover had grown at
the rate of 25% annually during the last five years, networth had grown at a compounded
rate of 40% in the last five years, long term debt equity ratio had been maintained under 0.5,
and ROI exceeded 30% and PBT was between 10 - 12% of revenue.
As seen in exhibit 4, AFL had an estimated turnover of Rs 4850 million during 1996-97
with a PAT of nearly Rs 1400 millions. Currently about 50% of the turnover and 70% of the
contribution was from the Express divisions. Roughly Rs 200 million was from the AFL
Distribution services during 1996-97. This had grown from Rs 26 million in 1993-94 to Rs
76 million and Rs 142 million in the intermediate years. AFL Cargo had a turnover of Rs
1380 million during 1996-97. The travel division had a turnover of Rs 770 million. The
turnover and revenue figures differ significantly for the Air Cargo and Indtravels divisions,
and the international money transfer activity with Western Union (included in the DHL
y
figures). As per the industry practice, turnover reflects the actual freight and ticket billings,
while the commission earned is shown as revenue. The profit and loss account in Exhibit 6
nl
reflects the revenues for the group.
O
History of the Organisation
It all started in 1867, when Framji Guzder was entrusted with the entire transport and
n
shipping arrangements of the British Commissariat campaign supplies, by Nusserwanji Tata,
io
the founder of the Tata empire. Soon, Framji and his son, Sorabji, set up a clearing and
forwarding agency called Sorabji Pestonjee & Co., which was renamed N.S. Guzder & Co. in
1892. It came to be closely associated with the House of Tatas. In 1930 when J.R.D. Tata
at
thought of setting up an airline in India, it was N.S. Guzder & Co. that cleared the first
aircraft imported by him - a de Havilland Puss Moth. By 1934, the company had extended
ul
its operations to the Wadias (Bombay Dyeing) and the Century Mills. During the second
world war, the company secured the exclusive clearing agency contract for the Government
irc
of India Supply Department, resulting in clearing of lakhs of tonnes of freight, imported by
the Government, for nearly two decades after the war. It also included handling, clearing and
transporting critical machinery for major projects, such as the Bhilai Steel Plant, the Oil
C
Refineries of Baroda and many others. N.S. Guzder & Co. continued to be a parallel
transportation business operation, while AFL, which was set up in 1945, had taken on new
d
business dimensions, leading to the present business structure in the transportation industry.
Exhibit 8 gives a more lucid account of the history of the organisation in the words of
ite
Jamshed N. Guzder, the father of Cyrus N. Guzder and President of AFL.
Structure of Current Business

im
The structure of the current business of AFL in terms of various products were:
rL
Courier
International/Domestic door to door express delivery of documents, parcels and commercial
Fo
packages
Cargo
Air/Sea import consolidation
Air/Sea import/exports
Domestic air cargo
Customs clearance
Warehousing
Documentation
Express Distribution
Domestic surface door to door cargo
y
Warehousing
nl
Distribution
Invoicing
O
Collection
Statutory documentation
n
Travel
io
International/domestic travel
Inbound tours
at
Conferences
Trade fairs ul
Money Transfer
irc
International money transfers
ILS
C
Tailormade single window ‘Integrated Logistics Solutions’ to key accounts

As of now, relatively, the ACE division of AFL had built up greater experience towards
d
integrated logistics, by continously increasing its scope of services.

ite
ACE – Express Distribution

im
To strengthen Airfreight’s presence in the domestic market, ACE was launched in April 1993.
A new division was formed to offer door-to-door express cargo service using surface transport
to the Indian market.
rL
The service package originally comprised of:

Door pickup and door delivery
Assured transit times
Fo
Safety
All India network
Track and trace information
This service was essentially targeted at high-value commodities (value to weight ratio
upward of 500 rupees per kg) like:
Computers
Telecom
Consumer durables
Pharmaceuticals
Engineering spares, etc.
y
ACE had developed infrastructure in terms of:
Offices
nl
Information technology
O
Communications systems
Manpower
Hubs for transshipment
n
The procedures were set and well documented. As business progressed, additional routes
io
were introduced and existing ones modified related to capacity utilisation. The network had
expanded from servicing 25 to 250 locations and continuous efforts were being made to
at
expand it further. ACE had 9 hubs through which almost 90% of the cargo was transhipped
for the purpose of deconsolidation, sorting, and consolidation for onward movement
ul
After 2 years of launch, ACE had diversified into total distribution service, where, in
addition to door-to-door express movement, the services included:
irc
Warehousing
Order processing
Invoicing
C
Stock keeping and related documentation

Collection of payment on delivery
d
MIS
ite
The idea was that this would help the customer organisations focus on their core
competencies in marketing and manufacturing and outsource part of the logistics function to
a specialized agent. This was seen as a distinct value-addition and as a cutting edge to stay
im
ahead of competition. Related to expansion in the distribution services, infrastructure had

been enhanced, developing warehousing capabilities in about 20 locations in the country.
ACE had recently entered into refrigerated movement by providing logistics support to
rL
McDonald’s. ACE was handling the total supply chain movement for its frozen and dry
products and acted as a key interface between the suppliers and the distribution centers.
Exhibit 9 gives an overview of the performance, reports provided to customers and customers
Fo
of ACE. Typically, the ACE division did not deal with customers who required any
international servicing.
Integrated Logistics Solutions

This function was set up formally on January 1, 1997, after various brainstorming sessions
within AFL over the earlier few months. This was intended to be a coordinating function to
enable this value-added service to be provided, leveraging on the strengths of the other
divisions. This function was headed by a General Manager (the person selected for the post
was a long time company executive, committed to this idea, and described as a soft spoken
consensus seeker), a Deputy General Manager to support him, two Managers looking after
the west and south respectively, along with a secretary. The geographical focus on the west
and south was because most of the targetted corporates had their headquarters in theses
y
regions.
In the early months of this function, customers were targetted for presentations on how
nl
they could gain by out sourcing integrated logistics. Customers were also pursued based on a
rebound from any of the other divisions, if scope for integrated logistics was percieved. Most
O
of the early customers were those who required atleast some international servicing like
import consolidation, customs formalities etc. The presentations usually involved homework
on various costs to the customers, often as suggested by the customers. Representatives from
n
the business divisions also went for the presentations along with the GM - ILS.
io
It was still early to talk about successes. However, reflecting on some of the efforts which
did not yield results, Mr. Guzder felt that while the customer often explicitly stated that the
at
price was not right, there could be more to it. For example, the customer may have expected
the MD to be present, or the presentation to be made by one person rather than asking the
ul
functionaries of the divisions to talk about their strengths.
Exhibit 10 gives excerpts from a concept note presented to a customer, describing the
irc
experiences of providing ILS to two other customers.
Market Profile
C
Given the success of AFL in the variety of activities related to transportation, along with
value added attributes like timeliness and door to door deliveries, AFL saw the future in
d
providing integrated logistics solutions. The market for this was visualised as having the
following customer attributes:
ite
Increasing sensitivity of product movement to speed

Increasing sensitivity of product movement to reliability
im
Increasing awareness of inventory costs

Increasing focus on core competence of manufacturing, marketing and purchasing, while
rL
outsourcing procurement and distribution logistics

According to an industry executive, “in the conventional distribution model, distribution
costs form about 2-3% of sales while inventory costs form about 4%. By adopting this
Fo
outsourcing model, the distribution cost increases to 5% of sales, but there is virtually no
inventory. As a result, there is a net saving of about 2%.” It was estimated that logistics and
supply chain management related costs are in the range of 12 to 15 per cent of the GDP. In a
developing country context, the same proportion was in the range of 18 to 20 per cent. The
difference resulted in better customer service parameters and in fact a better material quality
of life. Since customer sensitivity was moving along similar lines in India, the significance of
this activity would certainly grow.
Some of the changes taking place in the Indian corporate sector were:
(a) Increasing functional focus on this activity by appropriate restructuring and positioning
of this function in the organisation structure.
(b) Increasing coordination between marketing, manufacturing and procurement through
better planning systems and use of IT. However, this area was the one where a lot of
work was required since the corporate ethos in India was strongly “departmental” with
y
implications on educational background, protected promotional avenues, ability to pass
the buck etc.
nl
(c) A recognition that inventories needed to be reduced by reducing lead times and
uncertainties. However, organisations were often not clear as to how to go about it,
O
especially since measuring the impact on inventory related costs was difficult, while
assessing other opposing costs like transportation, which were essentially out of pocket,
n
was easier.
(d) A recognition that customer service needed to be measured and monitored if supply
io
chain performance had to be improved.
(e) A slow openness to review manufacturing processes including choice of technology,
at
production structuring, layouts and product design and packaging with a focus on
improving the supply chain. ul
(f ) Increased focus on vendor development and outsourcing to rely more on one’s own
competence. The outsourcing also extended to activities related to logistics like
irc
transportation, storage and even managing the entire procurement and/or distribution
processes. Many of the multinationals coming into India preferred using third party
logistics services.
C
The supply side to effective logistics and supply chain management was characterised by
poor infrastructure, especially in the area of transportation and storage. IT infrastructure was
d
improving at a faster pace. The service sector which was characterised by “not so” professional
and local agents involved as clearing and forwarding agents, distribution agents, stockists,
ite
wholesalers etc. who competed primarily on “contacts” was giving way to a branded national
service requirement being provided by third party logistics companies. These companies also
managed the transportation and storage infrastructure, an area crying for attention.
im
In this context, AFL was planning to offer integrated logistics support and supply chain
management. This would require providing services to customers for streamlining
rL
distribution in order to ensure that products reached their customers in the shortest time
possible. This not only meant transport of finished goods from factories to point of sale, but
also transport of raw materials or components from suppliers and vendors to points of
Fo
manufacture, and storage of finished goods and raw materials in their godowns. This
extension of activities to supply chain management was a logical extension of AFLs expertise
in transport, warehousing and - just as importantly - tracking the whereabouts of goods.
Much of the demand was expected from new entrants who did not have an existing
distribution network. For multinationals, outsourcing their logistics requirements provided
an easy route to the Indian market. Since their intention was to penetrate the market in as
little time as possible, they could not affort to waste time in setting up a distribution
network.
An integrated logistics service provider would also be expected to handle activities like
order processing, sales tax and excise duty documentation, invoicing, repacking and
secondary distribution, collection of bills, tracking consignment movements, claims
processing, warehouse management and inventory management.
According to a business magazine, the total market size was estimated as a Rs. 5,000
y
million business, with a potential of Rs.1,0000-1,2000 million within a couple of years. The
nl
growth was expected to be at a healthy 30-40%, persuading a number of traditional surface
transport companies, courier companies and freight forwarders to cash in on their existing
infrastructure for this business. Transport companies leveraged their fleet size and experience
O
in transporting cargo. Courier companies banked on experience in moving high value, small
volume items and could draw on their huge staff strength, large network of offices, tracking
n
expertise and infrastructure (including cargo planes) to help them set up logistics operations.
They expected the growth of the industry to come from the domestic market where the yields
io
were high, and the faster growth of parcels over documents. Freight forwarders leveraged
their consolidation expertise and knowledge of the transport system. The entry barriers were
at
the long gestation period, high capital investment in physical and IT infrastructure and low
margins at 15-18%.
ul
Some of the competitors were Dart Apex and Dart Surfaceline, divisions of the Mumbai
based courier company Blue Dart; Mumbai based courier company Elbee express,
irc
Hyderabad based GATI and XPS, Chennai based Sembawang Shriram Integrated Logistics,
Pune based Dynamic Logistics Delhi based courier company Safexpress, Clockwork Cargo, a
division of road transport major Prakash Roadlines and Bangalore based Pafex, a division of
C
Patel Roadways. Exhibit 11 gives a brief overview of the activities of some of the competitors.
The AFL Advantage

d
ite
The distinguishing features that AFL felt that it could offer based on its internal strengths
were:
1. The one stop shop - ILS services
im
Tailor made
Single window
rL
2. Inhouse express agencies

Time saving
Flexibility
Fo
3. Physical distribution network

Across the country
International
4. Information Technology
Own network
Inhouse software
Focused on customer
5. People
Dedicated and trained staff
Key account manager
6. Cost
Communication and co-ordination efficiencies and long term variable cost structure
y
Low entry cost due to established infrastructure
nl
The company executives elaborated on their expertise and network:
“In the process of moving people and things from one place to another over the
O
last fifty years, AFL has learnt, not just the basic art of movement, but observed and
studied, important aspects such as, ‘how to move’, ‘when to move’, ‘the shortest
route to be followed’, ‘the most economical mode of movement’, ‘the different types
n
of packaging required’, the legal and regulatory formalities involved in movement of
io
men and materials’, such as octroi, customs etc., and just about anything required to
ensure smoothness of the process. Time and again, the company has pioneered new
at
concepts in relation to movement and set trends and standards for the other players
in the industry to follow.”
ul
“Over 200 hundred offices, 150 warehouses, about 330 vehicles and 3500
employees is what keeps AFL always in a state of readiness. Since inception, AFL has
been very concious about the need for a strong, far flung, but intricately woven
irc
network. It has constantly kept adding to this network, both nationally as well as
internationally, to give itself a global reach. In November 1995, the country’s fully
C
automised Hub was commissioned at Marol in Mumbai. It is housed in the most

intelligent building in India, ‘Airfreight House’. Investing in similar Hubs, equipped
with hi-tech material handling facilities, in Delhi, Calcutta and Chennai, has
d
become a priority for the company. As a result, today, their is a bigger time window
available to customers to send documents and packages, allowing them maximum
ite
flexibility.”
“ The IT network called ENA (Electronic Network of AFL) has a 64 kbps
im
backbone connecting the metro cities of Mumbai, Bangalore, Chennai, Delhi and
Calcutta, with 9.6 kbps links to atleast 20 other locations. The ENA is also linked by
a 64 kbps line to the international DHL NET, a 64 kbps line to the Internet through
rL
VSNL, and by a 9.6 kbps link to the Vashi hub of AFL Distribution Services.
Customers can connect to the closest ENA node through dial up/leased line for
shipment and inventory data flow.”
Fo
Exhibit 12 provides a view of how the business activities of AFL would all come together
for providing ILS. Exhibit 13 details the same framework showing the ILS model with
information and material flows.
AFL felt that it had all the ingredients required to make a business out of integrated
logistics solutions based on some of the early successes with clients served by the DHL, ACE
and Air Cargo divisions. One of the questions facing AFL was whether it was necessary, and
if so, what was the appropriate time to make ILS as a profit centre.
Exhibit 1: Vision, Mission and Values
Vision
To Be The Acknowledged Leader In Providing World Class “Integrated Logistics” Solutions.
y
nl
Mission
O
To Provide Organisations And Individuals The Best Value In “Integrated Logistics” Through A
Global Network, Innovative Use Of Technology And Caring People
n
Values
io
We shall respect the dignity of the individual and so we shall:
Not discriminate on the basis of caste, religion or gender
at
Be caring to all employees
ul
Treat people as we would like to be treated ourselves
Create an environment that is conducive to learning and development
irc
We shall be transparent and fair in our behaviour, by:
Being honest in our dealings with customers and business partners
C
Being open to new and diverse ideas and approaches from our employees and customers
Ensuring a fair system of assessment of performance
d
We shall nurture personal initiative, commitment and loyalty, by rewarding those who:
ite
Take the lead and are pro-active in their approach to work

Take a continued interest in improving the performance of our business
Set and exhibit high standards of performance and excellence
im
We shall foster team spirit by supporting those who:

Build good working relationship with colleagues across all departments
rL
Are willing to take responsibility and offer support through their own initiative
Contribute actively to the improvement of services to customers
Fo
We shall always feel responsible to the communities in which we live and work through:
Good conduct in public and with our neighbours
Support of civic improvements
Reducing wastage of every kind and protecting the environment
Source: Company Documents,1997
Exhibit 2: Organisational Structure
y
nl
O
n
io
at
ul
irc
C
d
ite
im
rL
Fo
Source: Company Documents, 1997

Exhibit 3: Business Profile
DHL Worldwide Express

DHL Worldwide Express is the worlds largest international express distribution company,
y
delivering documents and parcels in over 226 countries across five continents. As of 31-12-
nl
1996, it had over 49,000 employees worldwide and was backed by state-of-the-art
equipment and technology. It handled over 88 million consignments, for over 900,000
O
regular customers worldwide. Its annual turnover exceeded US$ 4.2 billion.
Geographically speaking, it is the largest express company in the world delivering to
90,000 destinations in over 226 countries, where it is the leader in the documents and parcel
n
market, with an average market share of around 40%. It services 97% of the fortune 500
companies and every 58 seconds, a flight takes off with DHL material on it. Over the last five
io
years, it has grown at the rate of 20% per annum.
AFL, pioneered a new concept in 1979, when it tied up with DHL Worldwide Express,
at
formally introducing to India to “The Courier”. Today, DHL Worldwide Express has
become synonymous with AFL Limited. Through DHL India, the company controls a
ul
market share in excess of 55% in the international documents and packages business.
DHL India is now diversifying into the area of commercial as well as heavyweight, high
irc
value packages (JUMBO and JUMBO JUNIOR BOXES), heeding to the demands of the
local export fraternity, who were in dire need of an easier and economic system, to send their
product samples overseas.
C
DHL has worked towards forging alliances and forming partnerships with customers,
while establishing on-line, direct communication links between itself and them. The
d
advantage of having a national network to deliver things in a pre-determined time window,

has also opened up many more opportunities, one of them being, the money transfer facility,
ite
from overseas to any place in India, in association with Western Union, USA.
On the anvil is a ‘Repair and Return’ facility for foreign electronic goods, through the
im
creation of Express Logistics Centres, at strategic locations in the country. At the end of
1995, DHL’s most modern HUB in India (with hi-tech material handling facilities) was
commissioned at Marol in Mumbai, giving customers the flexibility of a bigger time window
rL
to send their documents and packages. Still allowing documents to be on-time, every time.
AFL Domestic Express

Fo
In 1982, AFL extended “The Courier” concept to the domestic level, starting a domestic
courier services called, AFL Domestic Express. AFL Domestic Express has been expanding
rapidly using the network of local AFL local stations and express centres all over the country.
In this process , the Domestic Express Service has not just covered, but practically managed
to link the whole of India.
Under this service, all consignments are handled on desk - to - desk basis, exactly like
DHL Worldwide Express. In short it carries documents and packages from one place to
another, anywhere in India.
Keeping up-to-date with latest technology, AFL Domestic Express monitors the
movement of shipments from origin right upto the destination, transmitting online
document and parcel information between AFL stations around the country. By doing this,
customers can have easy access to all the necessary information regarding their shipments, by
just calling the AFL station of origin or destination.
y
AFL Cargo
nl
AFL pioneered the freight forwarding industry in India, beginning with the air cargo
operation s in 1946. The same year it was appointed the first Cargo Agent of IATA
O
(International Air Transport Association). It has blazed new trails time and again in the cargo
industry, setting new trends and forcing other companies to follow suit.
n
With an extensive network IATA approved cargo offices throughout the country, AFL
Cargo is responsible for the expeditious handling of incoming and outgoing shipments by air
io
and by intermodal transportation facilities, between various countries around the world and
in India. Its gateway offices monitor the transit of cargo moving between inland stations and
at
foreign destinations.
It now offers a wide range of services, covering pick-up of cargo, warehousing pending
ul
export clearance, carting, processing of customs clearance and export formalities, booking of
space, consolidation of inward air cargo, constant monitoring of shipments, via state-of-the-
irc
art computer systems, break-bulk and delivery facilities at destinations.
Its global tie-ups with AFT Inc. of Canada, LEP International of UK, and Yusen Air of
C
Japan, make its reach truly global. Despite stiff competition AFL Air Cargo continues to
enjoy a strong leadership position in the domestic air cargo segment, while also having a
significant market share in the Air Exports Cargo Movement (non perishables).
d
Today, AFL Cargo is being scaled up to Multimodal Transport Operator business.

ite
AFL Travels
In 1948, AFL decided to move into the business of travel and Indtravels was born. It was
im
among the first travel agencies to be approved by IATA in India and was also among the first,
to be linked to Air India’s Computerised Reservation Systems, for offering instantaneous
bookings to customers. Besides international and domestic air bookings, AFL Travel’s
rL
services include, obtaining and renewing passports and visas, RBI sanctions, booking of hotel
accommodation, handling unaccompanied baggage, personal and baggage insurance, etc.
Fo
AFL Travels is one of the most experienced and leading government Tour Operators in
India, and member of reputed travel and tourism related bodies. In association with the
world’s leading tour promoters abroad, it offers foreign tourists to India, well planned
itineraries for their inbound tours.
In 1988, AFL Travels heralded the management of conventions, conferences and
exhibitions in India. Infact it is the member of the Indian Convention Promotion Bureau
and is on the approved list of Trade Fair Authority of India.
Today, it’s transforming itself, from a Travel Agency to a Corporate Travel Counsellor, and
thereby optimise the travel costs of the customers. It has tied up with Carlson Wagonlit - one
of the top Corporate Travel Agencies in the world (worldwide turnover of US$ 12 billion)
and has integrated its systems with their global system, resulting in world class service to its
corporate clientele.
y
Additionally, medical insurance cover is offered to foreign tourists, through a tie-up with
Europ Assistance - one of the worlds leading companies in travel insurance and emergency rescue.
nl
AFL Distribution Services
O
An offshoot experiment of AFL’s Cargo business in 1988, gave rise to its Advanced Cargo
Expertise, nicknamed as ‘ACE’ - the Express Distribution Division - specialising in total
n
distribution management.
io
Its initial concentration was on tailor made Door-to-Door cargo services across the length
and breadth of this country, taking care of everything from pick-up of cargo, booking
through economic and convenient mode of transportation, clearance of consignments
at
coming by air, at the destination and the final delivery to the consignee. Today, besides this
activity, AFL Distribution Services focuses on a specialised Logistics functions - Distribution
ul
Management.
A tie-up with Detroit based Logistics Management Company F.X.Coughlin has further
irc
honed its expertise in logistics, making it capable of offering specialised logistics solutions on
a turnkey basis, to a host of Indian and multinational clients, who are setting up operations
in the country. This, in turn, has led to the establishment of a ‘Cold Chain’, consisting of
C
Refrigerated Transport vehicles to take care of the corporates in the Food and Pharmaceutical
business.
d
By providing Logistic Services, to the full extent of managing the customer’s Supply
Chain, corporates like HP, HCL, Microland, Baccarose, Godrej Locks and McDonald’s now
ite
have more chance, to concentrate on their core business.

im
AFL Logistics
As the Indian Economy moves into the next phase of its growth, many of the leading
corporates are exploring the possibility of outsourcing, and need experienced partners, who
rL
can take care not only their Distribution, but Supply Chains as well.
AFL Limited, as expected has taken the lead in this area. Starting with its AFL
Distribution Services’ division, handling the transportation, warehousing and distribution of
Fo
high-value cosmetics for Baccarose Cosmetics. It was the company’s ability to its information
technology network, to pro-actively provide them information with regard to movement and
distribution of their stocks, inventory position and cash collection, with together helped
evolve Logistics.
‘Logistics’ simply mean planning the distribution process in a scientific and systematic
manner, right from transportation of raw material to the place of manufacture, to the
reaching of finished product to the customer, using proper routing, best mode of transport
available in terms of speed and economy, type of packaging required, etc. This concept has
been conceived with the very idea of offering customers all movement related services under
one roof. Providing them complete solutions. In a nutshell, this simply means ‘Supply and
Distribution Chain Management’.
Today this concept has been institutionalised and AFL has a full fledged Integrated
y
Logistics Solutions Department, which focuses on ‘Customization’ and ‘Account
Management’. This entails study of customer’s business, supply chain and distribution
nl
network, in order to formulate a comprehensive integrated logistics strategy, which will help
render all supply related services from a single window point.
O
The ability to provide integrated logistics solutions under one roof, is the result of the
company’s expertise, built over the last fifty years and the investments in the vast
n
infrastructure, networks and international tie-ups.
io
AFL Shipping
AFL Shipping was formed in early 1996, with the idea of focusing more closely on
at
consolidation of Ocean Freight business. AFL Shipping gives a distinct identity to the sea
freight business of the company. ul
AFL Shipping is a Non-Vessel Owning Common Carrier (NVOCC), which enjoys all the
previliges of a normal shipping line. It is engaged in booking container space on any shipping
irc
line and handles export consignments in any part of the country with ease, due to its national
presence.
C
Its main business focus is Less Than Container Load Consolidation (LCL Consolidation),
with its prime customer targets being Custom House Agents and Exporters, for whom this
service is of vital importance.
d
Utilising the AFL network, AFL Shipping services odd destinations, guarantees sail time
ite
and continuously tracks its cargo.
AFL Infotech
im
AFL Infotech was formed in 1994, as a niche software development house, for the transport
sector. Backward integration helps it to use the existing industry knowledge for developing
software for other players, in similar businesses. It specialises in providing complete world
rL
class software solutions on a global level, primarily to the Airline and Transportation
industries as well as to Distribution and Freight Forwarding companies and Travel Agencies.
Developing specialised and creating tailor-made software packages for the Airline
Fo
industry is AFL Infotech’s primary focus.

Its most popular software is the innovative passenger reservation system “Reser Vision”
which is an ideal software solution for small and medium size airlines across the world. This
software incorporates ticketless and credit card authorisation modules, while providing
comprehensive reporting from its Informix database.
Exhibit 4: AFL – Growth
Year Turnover (Rs million) Networth (Rs million) No of Employees
1990-91 975.2 46.2 1800
y
1991-92 1305.5 77.9 2000
1992-93 1740.7 122.7 2500
nl
1993-94 2239.9 206.8 2600
1994-95 3130.0 377.4 2700
O
1995-96 4090.0 541.9 3100
1996-97 (Est) 4850.0 650.0 3500
n
AFL – Divisionwise Turnover and Growth
io
(Rs million)
1993-94 1994-95 1995-96 1996-97
at
Turnover Revenue Turnover Revenue Turnover Revenue Turnover Revenue
DHL 1250 1250 1530 1530 2020 1920 2460 2300
Air Cargo
Indtravels
780
490
150
80
980
540
ul240
100
1250
610
270
110
1380
770
350
140
irc
ACE 30 30 80 80 140 140 200 200
Others 40 40 30 30 70 70 40 40
TOTAL 2590 1550 3160 1970 4090 2500 4850 3020
C

d
Exhibit 5: AFL – Financial Highlights

ite
1990-91 1991-92 1992-93 1993-94 1994-95 1995-96

im
Operating Margin (%) 12.60 13.00 12.60 12.60 18.40 13.90

PAT/REVENUE (%) 4.75 4.23 4.59 5.82 9.17 7.83
Return on Net Worth (%) 1 59.90 42.50 40.50 42.60 47.20 40.30
rL
Long Term Debt/Equity 0.36 0.20 0.25 0.42 0.39 0.31

Current Ratio 1.17 1.23 1.38 1.52 2.06 1.65
DSCR (Times) 2 4.67 7.50 7.29 10.38 7.47 4.82
Fo
Interest Cover 3 10.39 16.29 16.97 11.96 9.96 5.71

DSO (Days) 4 54 52 53 53 64 60
1{PAT/(Equity + Reserves)}
2Debt Service Coverage Ratio { PBDIT/(Interest + Principal repayable in one year)}
3{PBIT/Interest + Principal payable in one year)}
4Days Sales Outstanding {Outstandings/Daily Turnover}
Exhibit 6: Profit and Loss Account for the Year Ended 31st March, 1996
Rs in millions
1995-96 1994-95
y
Revenue
nl
Servicing, Transportation, Clearing
and Handling Receipts 2,356.86 1,844.49
O
Commission 67.41 54.80
Export Division 0.00 12.137
Shipping & Heavy Transport Division 2.00 3.60
n
Financial Services Division 36.59 23.32
Other Income 7.73 29.88
io
2,470.59 1,968.21
at
Expenditure
Operational Expenses 1,370.83 1,041.70

Employees Remuneration and Benefits
Administrative Expenses
ul 2,85.33
471.74
188.56
387.43
Financial Expenses 50.57 30.64
irc
Depreciation 53.76 45.47
2,232.23 1,693.81
Profit Before Tax 238.36 274.40
C
Less: Provision for Taxes 45.00 92.50

Net Profit After Tax 193.36 181.90
d
Add: Earlier Years Adjustment (Net) -10.58 -16.68

Earlier Years Adjustment (Tax) 223.18 21.26 -20.86 -3.75
ite
Net Profit 214.62 178.15

Add: Balance Brought Forward 12.00 6.35
226.62 184.50
im
Less: APPROPRIATIONS
Proposed Dividend 100.00 75.00
Transfer to Reserve Under Section 205 (2A) of the
rL
Companies Act, 1956 250.00 200.00

Reserve for Earning in Convertible Foreign Exchange 25.00 50.00
Transfer to General Reserve 1625.00 1400.00
Fo
200.00 172.50
Balance Carried to Balance Sheet 26.62 12.00
Source: Annual Report, AFL, 1995-96

Exhibit 7: Balance Sheet
Rs in millions
As at 31st March, 1996 As at 31st March, 1995
y
Sources of Funds
nl
Shareholders Funds
Capital 50.00 50.00
O
Reserve & Surplus 532.05 327.43
582.05 377.43
Loan Funds
n
Secured Loans 431.16 307.19
io
Unsecured Loans 102.97 172.97
534.13 480.16
at
1,116.18 857.59
Application of Funds ul
Fixed Assets
Gross Block 792.02 517.74
irc
Less : Depreciation 162.22 114.69
Net Block 629.80 403.06
Investments 32.64 23.49
C
Current Assets, Loans & Advances

d
Spare Parts 6.00 6.50

Sundry Debtors 674.48 481.05
ite
Cash & Bank Balance 81.00 67.88

Loans & Advances 283.79 283.74
1,045.27 839.16
im
Less : Current Liabilities & Provisions

Liabilities 621.65 400.62
Provisions 10.00 7.50
rL
Net Current Assets 631.65 413.62 431.44

Miscellaneous expenditure (to the
extent not written off or adjusted) 40.12 0.00
Fo
1,116.18 857.59
Source: Annual Report, AFL, 1995-96

Exhibit 8: History of the Organisation*
In 1867, my great-grandfather, Framji Guzder, came to Bombay from Gujarat to seek his
fortunes.
That same year, King Theodore of Abyssinia hanged the British Ambassador and took 60
y
Europeans hostage. Britain decided to retaliate, choosing Bombay as the base from which to
nl
launch the offensive.
General Sir Charles Napier, the Commander-in-Chief of the British forces in India, was in
O
charge of the operations. He awarded the entire Commissariat Contract for the campaign
supplies to Nusserwanji Tata, the father of Jamshedji Tata - founder of the house of Tata’s.
Nusserwanji Tata entrusted Framji Guzder with the transport and shipping arrangements.
n
And this was the beginning of a long, solid relationship with the house of Tata’s, which
io
continues till today!
A meeting of destinies... and the seeds of AIRFREIGHT were sown.
at
Pioneering with Pioneers
ul
After the Abyssinian expedition, Framji Guzder, and his son, Sorabji, set up a clearing and
forwarding agency called Sorabji Pestonjee & Co. In 1892, this agency was renamed N S
irc
Guzder & Co.
The first major industrial enterprise set up by Jamshedji Tata was the Empress Textile
Mills at Nagpur in 1877. The clearing and forwarding of all the important machinery, from
C
the docks to Nagpur, was entrusted to N S Guzder & Co. Subsequently, the shipping,
clearing, forwarding and transporting operations of each industrial enterprise that Jamshedji
d
Tata set up, and the enterprises that were set up after him, were entrusted to N S Guzder &
Co. As a result, they handled all imports and transport-related operations not only for the
ite
Tata Textile Group, but also for Tata Chemicals, the Taj Mahal Hotel, Tata Electric
Companies, Tata Airlines and others, right from their inception.
im
In fact, when Jamshedji Tata first conceived the idea of generating hydro-electric power
for Bombay utilising water resources of the Western Ghats, my father provided the launches
and accompanied Jamshedji and his project team to explore the coastal rivers. These
rL
expeditions led to the identification of several sites, where Tata’s hydro-electric projects were
eventually set up.
As the country progressed rapidly along the path of industrialisation, N S Guzder & Co
Fo
grew with it. When J R D Tata thought of setting up an airline in India in the 1930s, the first
aircraft imported by him - a de Havilland Puss Moth (which made the historic solo flight
from Karachi to Bombay in 1932) - was cleared at the Bombay Docks and transported to
Juhu by us.
*Source: The words of Mr. Jamshed N Gulzar, President AFL from A Portrait of Many Portraits, AFL,
1993.
When this flight carried the first Air Mail bags into Bombay, my father was prominent
among those who received J R D Tata at the Juhu airstrip. In fact, when J R D Tata repeated
his solo flight 25 later, my father was the only surviving member of the first welcoming
entourage to greet him when he landed.
AFL Takes Root
y
nl
On January 18, 1934, I joined my father actively in the management of N S Guzder & Co.
Hitherto the company had restricted its activities to two major groups - the Tatas, the Wadias
(Bombay Dyeing) and the Century Mills.
O
During the Second World War, I expanded its activities and secured the exclusive clearing
agency contract for the Government of India Supply Department. As a result, we handled
n
the clearing of lakhs of tonnes of freight imported by the Government of India for almost
two decades after the War.
io
We exclusively handled, cleared and transported import machinery for major projects
at
around the country including the Bhilai Steel Plant, the Oil Refineries of Baroda, and many
others.
ul
In January 1945, it was evident that the Second World War was coming to an end. I
thought that a new era in air transportation would soon begin and therefore, that same year,
irc
I requested Tata Airlines to appoint me as their first air cargo agent in India, which they did.
I therefore set up a company called Airfreight Limited, and transferred the agency license
from my name to this new venture which became Tata Airlines’ first, and at that time the
C
only, air cargo agent in India.

Thus, AFL was born.
d
AFL Spreads it Branches

ite
Tata Airlines became Air India in 1946, and grew rapidly, domestically and internationally.
AFL kept pace with this expansion. The next year, AFL was appointed as official Cargo
Handling Agent for TWA, the first international airline to touch Bombay in 1947.
im
From then on, AFL has moved from one landmark to another:
1946 Housed India’s fir Air Customs unit to clear import air cargo, For the first six
rL
months, this unit operated from our premises. The only instance where a customs
house operated from the premises of a private company.
1947 Awarded the handling contract for both, incoming and outgoing cargo by TWA,
Fo
thus becoming the first Air Cargo Agent to be appointed at Bombay Airport.
1948 Among the first of the IATA-approved Air Cargo Agents in India; and set p our
Travel Division—INDTRAVELS - among the first to be approved by IATA in
India.
1954 Appointed as the first accredited Passenger and Cargo Sales Agent for Indian
Airlines Corporation and Air-India upon the nationalisation of the airlines in India.
1960 Opened the Tours Division to promote tourism to India and earn foreign exchange.
1965 Became Indian Airlines’ exclusive Delivery Agent in Bombay for incoming domestic
cargo.
1979 Pioneered international courier services in India in association with DHL
International.
y
1981 Together with N S Guzder & Co Pvt Ltd, handled one of the single largest packages
nl
—the Stator-Generator for the Tata Electric Power Plant (weighing over 250 tonnes)
—to have landed in the port of Bombay.
O
1982 Introduced domestic courier services in India.
1983 Set up a Conference Division in New Delhi.
n
1985 Opened the Project Transportation Division to transport and delivery heavy project
io
equipment to remote areas in India.
1987 Together with N S Guzder & C Pvt Ltd, successfully completed the clearance and
at
transport of Indo-Gulf Fertilizer Corporation’s total plant to site (Jagdishpur, Dist.
Amethi, U.P.). We used our own barges and tugs to move the heavy packages (250
ul
tonnes to 400 tonnes each) by river from Calcutta to Mirzapur.
1988 Launched ACE - the first “door-to-door” domestic cargo services within India.
irc
1988 Opened the Trade Fairs & Exhibitions Division, accredited by the Trade Fair
Authority of India.
C
1992 As part of a planned diversification programme, set up an Export Division.
Over Four Fruitful Decades

d
AFL began modestly in 1945 with a staff of five persons in a small office. It has grown
ite
rapidly into an enterprise with nearly 100 offices all over India employing 2,400 persons. We
have expanded into such diverse fields as Air Cargo, Import Consolidation, Sea-Air, Airport
Cargo Handling for Airlines, Travel & Tours, Document and Parcel Courier Services, Heavy
im
Lift Transportation, Trade Fairs & Exhibitions and now, Exports.

Through the years of rapid growth and expansion, we have developed a reputation for
service and reliability. This tradition is being carried on by the next generation - my three
rL
sons and a highly competent team at the helm. We aim to enhance this image, always
keeping in mind the best interest of our clients.
Fo
Exhibit 9: ACE Express Distribution
Performance
1993-94 1994-95 1995-96 1996-97 (Est)
y
Turnover (Rs million) 260.0 760.0 142.0 200.0
nl
Tonnage (metric tonnes) 2400 6900 12900 16500
Shipments (Nos) 34,000 86,400 1,51,900 1,50,800
No of branches/express centers 24 40 55 70
O
Network (No of locations) 120 160 200 250
Manpower (Nos) 170 270 400 475
Vehicles: own 29 36 36 36
n
:contracted 15 20 30 30
io
Investment in IT (Rs millions) 5.6 8.0 25.0 28.0
Warehouse space (Sq ft) 15,000 25,000 45,000 80,000
at
Reports to Customers
Stock Statement and Productwise Stock Ledgers
ul
Stock Valuation
Goods Received Report
irc
Goods Issued Report
Sales/Stock Transfer Statement
C
Adjustment Report
Order Executed/Pending Order Statement
Dealer/Stockist Reports
d
Sales
Collections
ite
Outstanding
Debtor Ledgers
im
Customers
Hewlett-Packard
HCL Frontline
rL
Larsen & Toubro

Microland (Compaq)
Godrej & Boyce (Locks division)
Fo
Hughes Escorts Communication Ltd.

Unicorp
Orbit Peripherals
Baccarose Cosmetics
Tupperware
Over 3,000 large/medium organisations for door-to-door express transportation.
Exhibit 10: Concept Note on ILS
... We have foreign tie-ups with leading players in the international market which enables not
only to service domestic customers abroad but also have access to international practices and
y
standards. It is pertinent to note that all the above mentioned services will be required by any
manufacturing organisation. With the necessary resources in terms of physical infrastructure
nl
and systems already in place, we are ideally suited to handle the total logistics requirements of
our customers. We term this comprehensive service package when designed to meet the
O
logistics needs of a customer as an Integrated Logistics Service.
The activity similar to yours is being handled for organisations such as Hewlett Packard,
Godrej, Larsen & Toubro, HCL etc. The typical activities performed for two of our
n
customers are as follows:
io
CUSTOMER 1
at
The company is a multinational organisation involved in the manufacture of Information
Technology equipments. The manufacturing facility is situated overseas and the product is
ul
marketed in India through distributors. The logistics services involved were:
irc
a) Pickup from the supplier overseas
b) Import into India
c) Customs clearance
C
d) Movement from port to warehouse

e) Warehouse management
d
f) Order processing
g) Despatch
ite
h) Provide information
im
Warehousing Activities
Receipt in pallets, break bulk, storage, order processing, data management, documentation,
shipment tracking.
rL
The India office of the company which manages the operations, has been dealing with
different agencies for different activities in the initial stages when the product ranges and
their respective volumes were relatively small. The contract has been with us for the last three
Fo
years though the scope has reached its full potential in the last one year based on the
confidence created through achieving the committed service levels. The service scope
evolved as follows:
a) We were involved only in transportation part of the business, i.e. moving the
consignments from one location to another.
b) Subsequently the customs clearance part of the business was included in the scope.
c) Finally the warehousing and information management of the whole operation was
handed over to us.
Their plan was to gradually hand over the responsibility to an outside party only after
ascertaining the partner’s capability and intentions to go along on a long term basis. The
situation at present has reached a stage of fruition when we have instilled the necessary
y
confidence in them to select us as their logistics partner in India.
nl
Benefits
O
Since the entire service is handled by a single agency, the total cycle time required for
servicing the Indian market has been reduced by almost 50% from 38 days to around 18
days. (Though some of this benefits has accrued due to better inventory management on the
n
customer’s part). From our side, the contribution has been in the area of customs clearance,
documentation clarity and prompt despatch of the consignments.
io
Before using Through us
at
Average time from supp 8 days 4 days
Customs clearance time ul 15 days 4 days
Order processing time 3 days 1 day
Average transit time 7 days 4 days
irc
There is a reduction of 20 days in the cash-to-cash cycle resulting in tangible savings for
the organisation.
C
CUSTOMER 2
d
The product range handled by us are safety locks which are sold through dealers. The dealers
ite
then route it through a chain of retailers to the final customers. The business involves
sourcing from the manufacturing sites through a series of warehouses and final delivery to
dealers. The logistics services involved were:
im
a) Pickup from the production sites

b) Documentation
c) Warehouse management
rL
d) Movement
e) Order processing
Fo
f) Recovery and banking

g) Provide information
We have worked in tandem to develop a customised software capable of merging the data
with the total information base of the customer.
The activity flow is relatively simple, i.e. production to warehouse to dealers. The
complications arises due to the number and location of production centres, warehouse and
dealers.
No. of production centres - 10

No. of warehouses - 18
No. of dealers - 50
The documentation involved in the movement depending on the nature of movement
(stock transfer or sales) and the location also adds to the complication with rules being
y
different for different states and changing periodically. Data management of this many-to-
nl
many movements also poses a stiff challenge to our capabilities.
Benefits: Since the operations was being managed under the group’s distribution system,
O
the resources had to be allotted in terms of vehicle space, order processing priority and
responses to queries/complaints. This being one of the many products sold by the company,
the response time for any indent or order to be processed and despatched was almost 6 to 7
n
days. This has been brought down to a single day in our system resulting in not only reduced
io
pipeline inventory, but also relieving the resources previously used to other products of the
group.
at
There are other benefits in terms of improved information flow and the possibility of
using the resources for other products of the group.
Basis of Costing
ul
irc
With a view to simplifying the method of arriving at the total cost payable for the service
rendered, we have divided the entire process into basically four activities.
C
Activities Basis
a) Customs clearance No. of TEU/Shipments in case of air

d
b) Movement from port to the factory No. of ULDs/Distance

ite
c) Warehousing Fixed monthly charge

d) Movements between factory, Weight/Mode/Distance
Warehouses, Dealers, Direct Customers
im
Data Requirements
rL
As mentioned earlier in this note, we have listed the data that would be required by us to
make a commercial proposal. They are as follows:
a) Places where service will be required
Fo
b) Number of line items in the warehouse

c) Method of measuring service levels with benchmark
d) Volumes and frequency of movement between suppliers, factories, warehouses, dealers/
customers
e) Reporting frequency, formats and mode
General
There are areas which has not been covered in this note. This has cost and service
implications and is possible to comment on them after further discussions. These are:
a) Software required to be used for managing the warehouse and movement
b) Hardware required in terms of computers, printers, scanners, modems etc
y
c) Packing material to be used for repackaging
nl
d) Any other services not covered
O
Kindly note that this is a concept and will be firmed up after exchanging our views on the
above mentioned points...
n
io
Exhibit 11: Brief on Competitors*
at
Blue Dart
ul
Blue Dart Express was a public limited company based at Mumbai. It was started in 1983
and had a revenue of Rs 1145 million and a profit after tax of Rs 98.6 million during 1995-
irc
96. During 1996-97, the figures were Rs 1486 million and Rs 4.9 million respectively. As
stated by the company, the reduced profits were attributable to an increased interest and
depreciation due to project related expenses. Blue Dart had a tie up with Federal Express of
C
USA. But its operations were still strongly domestic. Blue Dart had acquired two airplanes
(Boeing 737s) in 1996, because it was thought that this could enable much later cut off times
for pick ups, and reduced costs. It had pumped in lot of money in technology. The Chief
d
Executive of Blue Dart remarked: “Capacity was another important factor which we have
ite
achieved with the aviation business. Our timings have become convenient and our cut-off
time is the best. Our competitors would have a cut-off time of 8:00 p.m. while our plane
lands in Mumbai at 1:30 a.m. and we can take the consignment till one hour before. Our
im
major growth has come from that segment.”

One of the airplanes had got grounded and consequently stripped and sold. A third
aircraft was purchased as a replacement. The total cost of the three aircrafts was Rs 300
rL
millions.
Blue Dart was now trying to target logistics and distribution services for the industrial belt
Fo
for which a multimodal system and 19 warehouses had been set up across the country. They
had 237 branches and were planning an extensive franchise programme. They were also
looking at alliances with other smaller operators to increase reach and had one such alliance
in place in the south. Blue Dart was planning to get more aggressive about the logistics
business including provision of warehousing, order processing, express distribution and
*Source: Business Magazines and Newspapers, and Company Brochures, 1996 and 1997
inventory management to its clients. It was hoping reach a revenue of Rs. 10,000 million by
the year 2000.
One of the earliest and long-standing customers of Blue Dart was HCL, for whom the
entire peripheral and spare part distribution was being managed.
A new agreement had been signed with Federal Express, starting May 1997, in which Blue
y
Dart was to be a pick up and delivery agency, apart from their own operations. Thus Federal
nl
Express would have a more significant presence in India.
Elbee
O
Elbee had a strong domestic network and acquired two planes around the same time as Blue
Dart. It had a tie-up with UPS of the US. In the domestic courier and integrated logistics
n
business, Elbee held the number two position. One of the aircraft, unfortunately, crashed in
the first week of July 1997, taking with it both the pilots.
io
An important customer that Elbee had was Walt Disney, which had set up shop in India to
at
manufacture a range of products that bear distinctive Disney characters. It had decided to
farm out all its distribution requirements to Elbee Express. Says a Walt Disney executive
looking after retailing : “At our current volumes, it was just not economical to establish our
ul
distribution network. And if we did, it would have taken us four-five years to get our
network in place. Disney’s decision to turn to Elbee is also expected to help the company
irc
keep inventory costs low. At present, Disney has leased 1,000 sq ft of warehouse space from
Elbee only in Bangalore. However, once it expands operations to other cities, it will hire six
more 1,000 sq ft warehouses for its chain. ”
C
GATI
d
GATI (from the Hindi word gati, or speed) was originally a subsidiary of a large
ite
Secundrabad-based trucking major called TCI. They were the pioneers in surface-based
express distribution in the country. Subsequently, they have become an independant
company and also have a tie up with Indian Airlines. During 1996-97, GATI had invested
im
Rs.100 million in fleet addition and in information systems such as the satellite-based vehicle
tracking system for its logistics division, Business Logistics Services (BLS).
Sony was one of GATI’s clients. When Sony entered India, it faced a classic dilemma:
rL
should it set up its own distribution network or hire a logistics service provider instead? After
much deliberation, the Japanese electronics giant decided to sign on GATI, one of the dozen
or so companies specialising in transportation and warehousing that had sprung up in recent
Fo
years. “Since it was necessary for us to quickly establish a strong distribution network to
achieve a high level of penetration in the market, we chose to outsource our distribution
requirements,” says a Sony India manager looking after logistics.
Sembawang Shriram Integrated Logistics

Sembawang Shriram Integrated Logistics, a new joint venture company between Sembawang
Corp of Singapore and the Shriram group of Chennai, had invested close to Rs. 1,100
million in land for warehouses and on its fleet. The company had yet to begin operations in
a significant way. In the meantime, the company was building a professional base through its
manpower sourced from Indian Railways and other transport related organisations.
Dynamic Logistics
y
Dynamic Logistics, which was incorporated in the year 1983, was one of the flagship
nl
companies of the ‘Talera Group’. It had expanded from just leasing out warehouse space to
handling integrated logistics solutions, supply chain management systems, re-engineering
processes and distribution systems of various companies especially in the automotive sector.
O
Their brochure claimed the following specialisation:
JIT Systems
n
EDI
io
Quick Response Systems
Reducing Inventory Levels
at
Reducing Transportation Costs
Routing and Scheduling ul
Cutting Warehouse Costs
Increased Flexibility
irc
Cross Docking/Freight Consolidation
Custom Bonded Warehouse
C
Container Freight Station

Contract Warehousing
PDI (Pre-Inspection Delivery) and more
d
ite
The brochure mentioned the following customers:
TELCO
im
Mahindra Ford India

Mercedes Benz India
Motorola
rL
Greaves
Larsen & Toubro
Fo
Asian Paints
Castrol
ABB Alfa-Staal
Glaxo
Steel Tubes of India
Gulf Oil
Kirloskar Oil Engines
One of Telco’s executives elaborated on the service being provided to Telco, “the Telco’s
spare parts warehouse was located at Thane earlier. In December 1992 it was shifted to Dighi
in Pune which is very close to our Pimpri works. Telco appointed Dynamic Logistics as our
logistics contractors as is the trend worldwide. Spare parts required for the Telco’s 207 family
of vehicles as well as shop made items for Pune manufactured commercial vehicles are
y
stocked here to fulfill the requirements of our valued customers. At Dighi we have a covered
area of 1,00,000 sq.ft.
nl
Advantages of setting up warehouse at Dighi:
O
1. The passenger car customers get the attention they deserve.
2. Service levels are near 100% and as Dighi is in the octroi free zone, Telco saves on octroi
n
duty.
3. Telco saves on overheads as Dynamic Logistics looks after our warehousing activity.
io
4. Packing and unpacking costs are also eliminated as the material from Pimpri Works to
at
Dighi, can be sent loose in open bins rather than in packed cases.
The monetary savings due to the above practice is expected to be around Rs. 40 millions
ul
per annum based on an estimated turnover of around Rs. 800 million in 1995-96.”
Another example is Mahindra Ford, which had a warehousing arrangement with Dynamic
irc
Logistics. The logistics company stored spare parts and components for the automobile
major. Whenever a Mahindra Ford dealer needed a particular spare part, Dynamic Logistics
delivered in two to eight days, depending on the location. While Mahindra Ford looked after
C
the placement of orders, inventory planning, scheduling, and providing the required
software, Dynamic handled stocking and delivery. Dynamic Logistics chief executive Praful
Talera said “Nowadays companies do not want to get mired in operative logistics. They prefer
d
to outsource.”
ite
Questions
im
1. What would be the profile of the market for ILS?

2. What are the strengths needed to provide ILS?
3. Given the strengths and weaknesses of AFL, what internal strucutres and processes are
rL
required to make ILS successful?

Fo
1. What would be the market profile for Integrated Logistics Services (ILS)? Try to detail
this profile in terms of type of product, importance of logistics services in the customer
firm, customer capability in managing its own logistics, and scale of operations.
2. For key market segments, what would be the major requirements that an ILS provider
would have to satisfy?
3. What is the major advantage that an ILS provider can offer to a company, as compared
to the company doing those functions itself?
4. Given the strengths and weaknesses of AFL, what internal structures and processes are
required to make ILS successful?
y
5. Assess the major infrastructural elements required to make the ILS concept workable,
keeping in mind the existing infrastructure of the various arms of AFL?
nl
6. What would be a reasonable method of costing and pricing for ILS? Since many
aspects of ILS would be internally provided by different existing businesses of AFL,
O
what would be the rationale for turning ILS into a profit centre?
n
io
at
Logistics services become important to a firm at a certain stage of growth of volumes of
business, spread of markets and value perception of customers. AFL is trying to provide a
ul
quick and cost effective way of providing these logistical services just at the time when they
become necessary, without the company having to invest in infrastructure for the purpose.
AFL would rely on using their established infrastructure for a number of customers so that a
irc
competitively priced service could be provided. This is the economy of scope argument for
AFL.
C
From an analysis of existing and potential customer profiles and requirements, the
specific gaps in the existing service options (offered both by AFL and by competitors) can be
identified. The idea of a one-stop ILS provider has to be fleshed out by looking at the
d
various specialised services (both within AFL and though possible external sources), to
ite
decide whether co-ordination per se is a viable option or whether a major restructuring is

required to achieve the desired integration. Here, the economy of scope for AFL would be
that it is relatively easy to diversify to provide or make provisions for the service package that
im
may be required for offering a product such as ILS.

Apart from general infrastructural requirements, a customer focused activity as envisaged
by the ILS concept would require appropriate use of IT for managing the information flow
rL
related both to the services to the customers and the cost control of AFL. These needs can be
spelt out.
Fo
Index
y
A Contracts 164
nl
A-type flow 82 Cost accounting 9
Activity-based Costing (ABC) 112 CPFR 242
O
Air transport 199 Cross-docking 202
Allocation 180 Customer time-based demand profile 68
n
Analytics 215 Customer-centric era 6
io
Application layer 217 D
Architecture of an IT system 218 Data layer 217
at
Assemble-to-order (ATO) 67 Demand volatility 73
Assignment problem 251 Distribution management 176
ul
Auction 161 Distribution Network Design 258
irc
B Distribution Resource Planning (DRP) 180
Balanced scorecard 109 Double marginalisation 262
Duality 257
C
Batch production 7
Beer game 40 Dynamic prices 161
Benchmarking 72 E
d
Build-to-order (BTO) 67 Economic Order Quantity (EOQ) 243

ite
Build-to-stock (BTS) 67 Elasticity 263

Bullwhip effect 39 Electronic auctions 170
im
Buy back contracts 187 Electronic exchanges 161

C Engineer-to-order (ETO) 67
Enterprise Resource Planning (ERP) 219
rL
Cellular layout 142

Channels of distribution 176 Exponential smoothing 239
Clarke Wright Heuristic 261 F
Fo
Client-server architectures 216 5-S 125

Competences 117 Fleet management 196
Competitive benchmarking 72 Forecast error 240
Container operations 199 Forecasting 45, 215, 238
Continuous improvement 149 Functional benchmarking 72
534 Index
G News vendor model 245

Global sourcing 159 O
H Offshoring 52
Hub and spoke networks 201 One-piece flow 140
y
I Order qualifiers 66
nl
Information alerts 214 Order winners 66
Information exchange and reporting 214 Outsourcing 52
O
Inland water transport 206 P
Integer programming 258 Penalty costs 168
n
Inventory control 244 Performance measures 167
io
Inventory planning 243 Performance monitoring 108
J Perishable inventory 248
at
Just-in-time (JIT) 122 Physical auctions 170
Just-in-time Distribution (JITD) 39 Point-of-sale (POS) 42
ul
Portal technologies 223
K
Predictive maintenance 148
irc
Kanbans 145
Presentation layer 217
L Price discovery 170
C
Lead time 43 Pricing 261

Lead times 44, 238 Prisoner’s dilemma 94
d
Lean supply chains 63 Process flow chart 128

ite
Linear programming 250 Procurement 157

Little’s law 48 Producer-centric era 6
Pull replenishment 145
im
M
Material accounting 180 Purchase Price Variance (PPV) 159
Metrics 86 Push and pull systems 39
rL
Mixed Integer Linear Programme Q

(MIP) 258 Quantity flexibility contracts 187
Fo
Mixed model scheduling 138

R
Moving averages 238
Radio frequency based identification 162
Multi-modal transport 185
Radio Frequency Identification Technology
N (RFID) 232
Near-beer game 46 Rail transport 198
Network flow models 251 Raw as Possible (RAP) 75
Index 535
Regression 239 T
Return on invested capital 97 T-type flow 84
Return on investment 10 Takt time 131
Revenue sharing contracts 187 Third party logistics services 178
y
Risk pooling 51 Time based contracts 188
Routing 260 Total Cost of Ownership (TCO) 160
nl
R, Q policy 244 Total Productive Maintenance (TPM) 148
O
S Toyota production system 122
Scenario modelling 215 Transaction engine 219
n
SCOR model 111 Transfer prices 262
Seasonality 239 Transportation mode choice 185, 194
io
Sensitivity analysis 257 Transportation problem 251
at
Service measures 167 Travelling Salesman Problem (TSP) 260
Service Oriented Architecture (SOA) 233 Trend 239
Simulation 249
ul
Trucking 197
Two bin policy 247
Single source 157
irc
Spaghetti diagram 130 U
Spot purchases 158 User interface layer 217
C
S, s policy 244
V
Standard cost accounting system 14
V-type flow 81
d
Standard costing 112

Value stream 70
Standard work 144
ite
Value stream map 70

Supplier selection 158
Vehicle Routing Problem (VRP) 260
Supply chain design 62
im
Supply chain drivers 4 W

Supply chain dynamics 38 Warehouse management 186
rL
Supply chain metrics 105 Warehousing 204

Supply chain performance 92 Water transport 200
Systems perspective 64 Work-in-process Inventory (WIP) 8
Fo

Book SCM

Uploaded by

Book SCM

Uploaded by

S upply Chain M anagement

Narayan Rangaraj is Professor of Industrial Engineering and Operations Research at

visiting faculty at universities in USA, Canada, Yugoslavia, Singapore, UAE and

Mandyam M Srinivasan is the Ball Corporation Distinguished Professor of Business

1992. He has a Ph.D. from Northwestern University, a PGDM from IIM-

Srinivasan has many years of experience in the automobile industry. He has

University of Tennessee, USA

Tata McGraw-Hill Publishing Company Limited

Senior Production Manager: Manohar Lal

Asst. Product Manager: Vijay Sarathi Jagannathan

Asst. General ManagerProduction: B L Dogra

highlighted in Chapter 4. This chapter uses a case to demonstrate how organisations

chain efficiencies and the benefits gained by applying lean principles.

a routine application oriented view of highly specific techniques within artificial

appropriate places, the specific conditions of the Indian sub-continent in terms of

Research Associate, IIMA Faculty, IIMA

Preeti Monippally Rohit Pathak

PGP Student, IIMA Faculty, Railway Staff College,

PGP Student, IIMA Research Associate, IIMA

Bayer CropScience Limited

the case Chilli in Soup (A).

I thank the College of Business Administration at the University of Tennessee,

Chapter 2. Understanding Supply Chain Dynamics 36

2.2 The Bullwhip Effect 40

2.4 Inventory Management and Supply Chain Dynamics 49

Chapter 3. Designing the Supply Chain 60

Chapter 5. SCM Effectiveness and Lean Thinking 120

5.1 The Toyota Production System 122

Chapter 6. SCM Across Organisations: Upstream Interface 154

6.1 Procurement 157

6.6 Markets and Auctions 169

8.4 Rail Transport 198

8.6 Air Transport 199

8.7 Water Transport 200

8.9 Storage and Warehousing 204

9.6 The Promise of Information Technology 231

Case 7 Chilli in Soup (A) 412

Case 9 FoodWorld (B): Supply Chain Strategy 456

Case 10 Seth Dhaniram C&FA 493

JITD Just-in-time Distribution

MIP Mixed Integer Programming

OLAP On Line Analytical Programming

PPV Purchase Price Variance

RUNS Requests for Underwriting

BCS Bayer CropScience: Science for a Better Supply Chain

AFL Airfreight Limited

improve the image and quality of chilli

8. Bayer CropScience: Science for a Supply chain improvement possibilities

Better Supply Chain for a pesticide firm

11. Airfreight Limited Strategy for developing a third party

the strengths of a courier business

1.3 Supply Chains in the Producer-Centric Era

1.3.2 Management Structures in the Producer-Centric Era

1.4 Financial Management and Return on Investment

1.5.1 New Business Environments Need New Management Systems

effective manner. ures do not, for example, consider the

petitors. A survey conducted in 2001 [R. D’Avanzo, H. von Lewinski, and L.

tions was between 3 per cent and 6 per competitive advantage.

1.1 OWNERSHIP OF THE SUPPLY CHAIN

1.2 SUPPLY CHAIN DRIVERS

flexible manufacturing technology. Information and communication technology

1.2.3 Business and Social Environment

1.2.4 Policy and Regulation

used to manage supply chains in a customer-centric era are significantly different

1.3 SUPPLY CHAINS IN THE PRODUCER-CENTRIC ERA

a problematic period in a number of ways. Lack of global competition created, in

1.3.1 The Rise of Batch Production

reduce distribution costs by viewing this activity as an integrated, rather than a

industries fostered the growth of large warehouses to stock finished goods.

1.3.2 Management Structures in the Producer-Centric Era

industry during the producer-centric era. The partitioning of organisations into

quantitative data that were available and deemed important by accounting

as production, distribution, and procurement.

Thus, during the producer-centric era, organisations were managed typically

optimisation. Each department attempted to maximise its own objectives, objectives

essentially adversarial in nature. The conflicting departmental goals and adversarial

Asst. General ManagerProduction: B L Dogra

2.3.1 Littles Law