TOTAL

QUALITY
KeyTerms and
Concepts
William 1, Duncan
hftig & Warren International

American Management Association

New York Atlanta Boston Chicago Kansas City San Francisco Washington, D. C.
Brussels Mexico City Tokyo Toronto
 This book is available at a special
discount when ordered in bulk quantities.
For information, contact Special Sales Department,
AMACOM, a division of American Management Association,

This publication is designed to provide accurate and authoritative

information in regard to the subject matter covered. It is sold with the
understanding that the publisher is not engaged in rendering legal,
accounting, or other professional service. If legal advice or other expert
assistance is required, the services of a competent professional person
should be sought.

Libra7 of Congress Cataloging-in-PublicationData

Duncan, William L.
Total quality :key lerms and concepts / (William L. Duncan].
p . cm.
Includes bibliographical references.
ISBN 0-8144-7876-X
I . Total quality management-Dictionaries. I . Title.
HD62.15.086 1995
658.5'62'03-4~20 95-33989
CIP
0 I995 Lufig & Warren International.
All rights reserved.
Printed in the United States of America.
This publication may not be reproduced,
stored in a retrieval system,
or transmitted in whole or in part,
in any form or by any means, electronic,
mechanical, photocopying, recording, or otherwise,
without the F o r written permission of AMACOM,
a division of American Management Association,
135 West 50th Street, New York, NY 10020.
Printing number
I 0 9 8 7 6 5 4 3 2
 Introduction
Total quality management (TQM) implementation is a lifelong, company-
wide commitment. It involves broad, sweeping change across the organiza-
tion and deep, fundamental process improvement. In most organizations,
the very culture of the company is restructured.
This book is a desk reference for T Q M practitioners who have under-
taken the TQM journey and for anyone interested in quality improvement,
a core element for organizational success. The definitions and discussions of
key quality terms and concepts help provide a better understanding of the
various aspects of the quality field. To help facilitate this learning process,
boldface type is used to highlight terms or phrases that are defined within
the glossary portion of this book. Because of the fundamental, culture-based
nature of TQM-related change, it is important to provide the following ele-
ments in a reference book like this one:

1. A general framework for ready reference that outlines the phases of TQM
implementation. An overview of the phases of successful T Q M implementa-
tion is provided in this Introduction, and each phase is detailed in individ-
ual listings within the publication’s definitions. The outline should allow
practitioners to quickly reference this book for (1) an objective benchmark
of the types of activities that should be completed or under way at any point
in the TQM implementation process and (2) a general understanding of
what lies ahead. This information will also prove invaluable in developing
introductory and overview materials about how T Q M implementations
work and what they involve. (A detailed implementation plan is not in-
cluded here, since many already exist in popular TQM books, and the spe-
cifics of each implementation plan vary greatly from organization to organi-
zation, reflecting the needs of the company involved.)
2. A glossary of TQM terminology. Cultures are held together not only by
a system of values and beliefs, but also by a common language. The lan-
guage of TQM is unique, and most of the terminology involved is listed and
defined within this text. Related terms are cross-references as appropriate.
Glossary entries span the entire spectrum of the quality field-from techni-
cal terms such as Poisson distribution to less technical but more frequently
used terms such as empowerment. (Note that a term may be more commonly
used without necessarily being more commonly understood!)

The Phases of TOM Implementation-An Overview

Phase I: Preparation Phase
Phase I involves an examination of the business-its objectives, problems,
and customer needs. During this phase, management decides whether to

iii
iv Introduction

use TQM or another approach to meet the needs of all of the company’s
stakeholders. For more detailed information, see the entry preparation
phase (phase I) in the glossary portion of this text.

Phase 11: Planning Phase

The second phase consists of the following steps: (1) securing of commit-
ments; (2) establishment of the business lead team, company steering com-
mittee, and process improvement team structures; (3) strategic planning
and policy deployment; (4) development of the implementation plan; and
( 5 ) awareness training. For more detailed information, see the entry plan-
ning phase (phase 11) in the glossary portion of this text.

Phase 111: Assessment Phase

This phase consists of identifying the structure, strengths, and weaknesses
of the organization, competition, products and services, work force, and cus-
tomers. At the conclusion of this phase, there should be a clear understand-
ing of customer needs, how well the organization fulfills those needs, and
where the organization’s processes need to be strengthened to improve cus-
tomer satisfaction and gain market share. For more detailed information,
see the entry assessment phase (phase 111) in the glossary portion of this
text.

Phase IV: Implementation Phase

Also called the deployment phase, this phase of T Q M implementation includes
the following steps: (1) alignment of the program with strategic plans, (2)
strengthening the program infrastructure, (3) initiating continuous im-
provement activities, (4) monitoring performance, and ( 5 ) communicating
lessons learned. At the conclusion of this phase, there should be a strong,
ongoing TQM-based continuous improvement process in place with tangi-
ble benefits manifested. For more detailed information, see the entry imple-
mentation phase (phase IV) in the glossary portion of this text.

Phase V: Networking Phase

In the final phase of TQM implementation, initial TQM successes are ex-
ploited, leveraging them to stimulate broader improvements throughout
the organization, and reaching out into the supplier community. The orga-
nization ( 1) communicates strategic improvement plans with outside organi-
zations, (2) shares lessons learned with these key organizations, (3) initiates
regular visits with suppliers and customers to cement the bonds of under-
standing and mutual growth, (4) institutionalizes the improvement process
throughout the customer/producer/supplier chain, and (5) develops the
process into a continuous “loop” that assures sustainability. For more de-
tailed information, see the entry networking phase (phase V) in the glossary
portion of this text.
 Table of Contents

Introduction iii

ABC - average outgoing quality (AOQ) 1

backflushing - business processes 14

c-chart - cycle time 21

D-chrt - duty cycle 48

economic order quantity (EOQ) - extrapolation planning 60

facilitation - FURSAP measures 69

GAAP - growth rate analysis 80

Hawthorne effect - hypothesis 83

imperfection - isolated peak distribution 86

JIT - just-in-time (JIT) manufacturing 92

kaizen-kutosis 95

latent defects - LSL 97

machine shop- multivariate charts 100

N - np-chart 107

objective - ownership 111

p-chart - push systems 118

QA - quick-change devices 142

R chart - run chart 147

s-chart - synchronous production 155

tactical metrics - tweaking 171

u-chart - USL 178

value - voice of the customer 180

WANs - world-class manufacturing 183

X, R chart - X chart 185

yield - yield 186

Z score - zero defects 187

ABC See activity-based costing.

absolutes of quality ma?ugement A set of principles that include:

1. Quality means conformance, not elegance.

2. There is no such thing as a quality problem.
3. There is no such thing as the economics of quality; it is always
cheaper to do the job right the first time.
4. The only performance measurement is the cost of quality.
5. The only performance standard is zero defects.

accqtable quality leuel (AQL) The percentage of defects that are accept-
able without rejecting an entire lot sampled. This concept, developed in
the 194Os, contradicts the philosophy of TQM in many important ways: It
allows d e f m to pass through to the final product, it implicitly encourages
suppliers to ship defective materials to their customers, and it reduces
quality to a statistical game in which the customer is the ultimate loser.
Also called acceptance number.

acceptance number See acceptable quality level.

accountability The level of ownership perceived by TQM participants

and shown in their actions. Accountability requires defined responsibility,
acceptancdownership of that responsibility, understanding of the objec-
tive and how one’s actions influence achievement of that objective, and
some level of enthusiasm for attaining the stated objective. Enthusiasm
may result fiom positive incentives (e.g., financial rewards). See also re-
ward and recognition system.

accreditation The credential-establishg action of official registrars. In

relation to achieving International Standards Organbation registration,

1
2 accuracy

the I S 0 dictionary defines accreditation as “a hvorable assessment by the

Registrar Accreditation Board in the United States or a counterpart in
another nation attesting to a registrar’s requirements with applicable re-
quirements.”

accuracy The degree to which a measurement process conforms to a

predetermined standard or accepted reference value. It is important to
distinguish between accuracy and precision; they are not synonymous.
Being precise might imply that the variation from measure to measure
on the same specimen is extremely small, or for that matter, practically
nonexistent. However, unless the measured value for the characteristic
fBlls at or near the true value of the master specimen or standard, the
sample measures will all be precisely wrong. The degree to which the
mean of repeated measures of a given specimen or standard varies from
the true value is a measure of the bias of the measurement process. Unbi-
ased measurement processes are said to be accurate; biased processes are
said to be in need of calibration. See also conformance.

activity-bcLsed costing
)C‘( A cost accounting process that begins with
the premise that products consume activities and that the activities in turn
consume resources (rather than beginning with the premise that products
consume resources). ABC differs sigmficantly from traditional methods of
accounting and is much more supportive of TQM-related improvement
activities. By i d e n w g the cost driver activities, a much more accurate
picture of costs associated with any given product can be obtained. The
two major elements of ABC systems are the process value analysis and
the cost driver analysis. The culmination of these efforts provides tight
linkage between financial performance and quality improvement activi-
ties.

adding value Taking action that enhances the satishction of customer

needs,and does so at a minimum cost in a safe environment. The cus-
tomer may be a downstream internal customer or an end consumer of
the product. See also customer, internal and customer satisfaction.

advanced pt.ocurement technology (APT) A field of just-in-heman&-

turing that integrates a supplier‘s operations as closely as possible with
those of its customer. In essence, partnerships are formed between suppli-
ers and customers. Specific elements of A€Tinclude (1) data sharing, (2)
improved ordering devices and methods, (3)commodity-based contract-
advanced procurement technology (APT) 3

ing, (4) comprehensive supplier performance measures, (5) supplier certi-

fication, (6) e f f d v e buyer evaluation criteria, (7)joint suppliercustomer
cost reduction initiatives, (8) supplier JIT training,and (9) enhancements
in the procurement organization structure.

1. Data sharing means the sharing of data between a manufac-

turer and a supplier to improve response time, increase efficiency,
and foster trust between the two organizations. Sharing data in this
fashion, especially with the manufacturer’s most important suppliers,
can provide vital early warning signs of impending changes in vol-
umes, delivery dates, and capacity constraints that could have sub-
stantive impact on the ability to deliver purchased parts on time.
However, effective data sharing can usually only be accomplished
when the supplier base has been narrowed to avoid risks associated
with industrial espionage and general leaking of competition-sensi-
tive data. Among the specific types of data that are commonly shared
with suppliers in advanced TQM environments are production
schedules, sales forecasts, design data, packaging requirements, dis-
tribution methods and volumes, and field service reports pertaining
to the supplier-provided components and materials.
2. Improved ordering devices and methods typically include elec-
tronic data interchange equipment (hardware and software) and
“pull”-type ordering methods similar to the Japanese kanban tech-
niques. (See pull system.) ED1 can make the sharing of data, place-
ment of orders, revision of designs, and other important data ex-
changes virtually instantaneous. The pulling of material via a kanban
approach turns a part of the suppliers’ manufacturing capacity into
an extension of their customers’ operations. It is a remarkably effec-
tive means to eliminate work-in-process queues of parts and mate-
rial.
3. Commodity-bused contracting is a purchase agreement written
to cover an entire commodity of parts or materials. Combining the
requirements for all the parts within a commodity frequently pro-
vides enough volume-based leverage to create efficiency and flexibil-
ity advantages for both suppliers and their customers.
4. Comprehensive supplier performance measures are metrics used to
track a supplier’s performance, such as price, delivery, product qual-
ity, financial stability, geographic proximity, special advantages (such
as providing an opportunity to enter foreign markets), capacity for
4 advanced quality planning (AQP)

growth, and the willingness to share data, undertake consignment

arrangements, and do commodity-based contracting, specialized
containering, and JIT and pull-based manufacturing.
5. Supplier certijication is a state achieved when the buyer gains
an adequate level of confidence in the suppliers’ process controls in
order to eliminate the need for incoming inspection activity. At this
point, infrequent random sample spot checks and the process con-
trol data of suppliers’ processes are adequate to ensure supplier
product quality.
6. Effective buyer evaluation criteria in a TQM environment in-
clude quality and delivery performance of the buyer’s supplier base,
cost containment and reduction, supplier lead time reduction, per-
cent of time spent at supplier facilities, and effectiveness in joint cost
reduction/profitability efforts with suppliers. Additional criteria
being used include ongoing education efforts and material inventory
turn rates.
7 . Joint supplier-customer cost-reduction initiatives involve the collab-
oration of professionals from multiple disciplines (typically including
engineering, production, and quality assurance) from both the cus-
tomer and supplier organizations with the purpose of reducing the
cost of the purchased goods. Any success in this area is then shared
equally between the supplier and the customer.
8. Supplier JZT training involves developing a training team from
the customer facility and/or from an outside consulting agency to
train a supplier in the application of TQM (and specificallyJIT) prin-
ciples in its operations. See also education and training (for total
quality).
9. Enhancements in the procurement organization structure may in-
clude the adoption of buyer-planner job responsibilities (combined
responsibilities of the former requirements planner and buyer func-
tions) and the organization of procurement professionals’ responsi-
bilities around commodities.

advanced quality planning (AQP) The process of designing final

quality into the product at the planning and engineering stage
rather than after manufacturing has been initiated. This concept is
essential in developing robust product designs and is fundamental to
concurrent engineering.
alpha (a) 5

afjnity diagram A tool that assists in problem definitionhesolution

by organizing ideas according to recognized relationships (levels of
affinity) between them. In TQM activities, the technique is most often
applied as a form of brainstorming where ideas about potential
causes of problems and symptoms are put on Post-it notes as quickly
as they can be identified by a group, then more deliberately orga-
nized into clusters that appear by group consensus to be related to
each other. This tool is frequently used together with a relationship
diagram to map various symptoms to causal factors in problem solv-
ing strategy (see Figure 1).

AZ See artificial intelligence.

alpha (a) The probability of making a Type I error; that is, reject-
ing a hypothesis when it is true. For control charts, the probability
that we think a process is out of control although it is still in control.

Figure 1. Affinity diagram for problem-solving tools.

Problem-Solving
Tools and
Techniques

(5 Categories)

SIPOC Brainstorming CPMIGantt

Check Sheets Histograms ChaftS

Respomibility
Flow Cham Arrinity Diagram Pareto Diagrams Run Charts Matrix

Graphing
T w Diagrams Techniques ControlCharts

Cause-and-Effect Scatter Diagrams

Diagrams

Matrix Diagrams

Ranking
Techruques
6 American National Standards Institute (ANSI)

American National Standards Institute (ANSI) An organization

founded in 1918 for the purpose of eliminating redundancies and
conflicts among the standards developed within industry. ANSI is the
U.S. member of the International Standards Organization, which is
based in Geneva, Switzerland. The IS0 is, in turn, the originating
body for IS0 9000 quality standards.

American Productivity and Quality Center (APQC) An organization

based in Houston, Texas, that does extensive work in quality and
TQM consulting and has an especially strong practice in the area of
benchmarking. The APQC was founded in 1977 and currently has
approximately 300 members.

American Society for Quality Control (ASQC) A professional organi-

zation for quality assurance and quality management professionals.
Based in Milwaukee, Wisconsin, this organization was founded in
1946 and today has more than 120,000 members.

AMH See automated material handling.

analysis of means (ANOM) An analytical technique that uses the

graphical format of a control chart in the (statistical) assessment of
differences between group summary statistics for groups suspected
of being different by their preexisting nature, or because of specific
(experimental) manipulations. The ANOM technique employs upper
and lower decision limits as opposed to upper and lower control lim-
its used on control charts. Unlike the control chart approach, which
typically uses 2 3 standard error units around a grand average as
control limits for the relevant group statistics, the ANOM uses 2 “H”
standard error units around the grand average for assessing differ-
ences. The value for “H” is a function of the specific number of
groups (k), the alpha risk level (Type I error rate) the practitioner is
willing to accept, and the degrees of freedom (df)associated with the
variability within the groups being studied.

analysis of variance (ANOVA) A statistical method for analyzing

data, usually resulting from an experiment. The total variation of a
data set is subdivided into parts associated with each source of varia-
tion so that variance components can be estimated, or a hypothes-
is(es) can be tested related to the parameters of the model. The three
architecture, information system 7

basic calculational models are fixed, random, and mixed, often re-
ferred to as Model I, Model 11, and Model I11 respectively.

ANOM See analysis of means.

ANOVA See analysis of variance.

ANSI See American National Standards Institute.

AOQ See average outgoing quality.

APP See automated process planning.

appaisal costs The costs resulting from formal evaluations of qual-

ity levels and for maintaining the documentation associated with con-
formance to company quality standards. These costs include labor
(typically inspection time), measurement equipment (from scales
and micrometers to x-ray equipment), and associated costs such as
source inspectors’ travel expenses. Part of a cost of quality account-
ing system.

APQC See American Productivity and Quality Center.

APT See advanced procurement technology.

AQL See acceptable quality level.

AQP See advanced quality planning.

architecture, business See business architecture.

architecture, information systems The fundamental design of the in-

formation system. This design dictates the standards for all devices,
software, and user interfaces. It defines the capability and capacity of
the systems and is the principal guideline for all subsequent systems
development.
The information systems architecture is at the core of progres-
sive manufacturing and service companies and will prove to be criti-
cal to providing the kind of responsiveness and process control re-
quired for success in the decades ahead. Industry will move toward
8 arithmetic mean

seamless systems consisting of integrated workstations, database

management systems, servers, and some application-specific systems
for departmental use.
Typical information systems and communication systems archi-
tectures are depicted in Figures 2a and 2b.

arithmetic mean For any set of values, the average or the sum of the
values divided by the number of values summed. It is typically de-
noted in quality applications by K or X-bar. See also mean.

Figure 2a. Typical integrated information systems architecture

(1 995-2005).
1
I I

Data Processing Production Workstations

Terminals Mainframe
J
.

File
Public Processor
, Voice/Data
Telephone PABX
System

Other
Mainframes A A

Primary (Fiber Optic)

Local Area Network@) I
Department
Systems w Print Server@)

I II
m
Workstations! Workstations!

Workstations
artijcial intelligence (AI) 9

Figure 26. Typical communication systems architecture ( 1 995-

2005).

Wide Area Network

(Voice, Data, Images, etc.)

Main Facility Remote Facillty

Primary Local Primary Local Area
Area Network(s) Network(s)

Other Communications
Local Area Controllers Departmental
Networks Systems

Various
Sewers Various
Sewers

Secondary
End User End User Local Area
Devices Devices Networks

arrow diagram A popular term describing the output of the PERT

process and often called a PERT chart. A chart defining the prece-
dence and time spans of the sequence of events required to produce
a desired outcome. Often used as the top-level tool for monitoring
large projects with many interrelated steps because of its ability to
succinctly demonstrate complex relationships and interdependencies
(see Figure 3). See also program evaluation and review technique.

artificial intelligence (’I) A branch of computer science that deals

with the development of computers that can “think.” The two pri-
mary thrusts of A1 explored most extensively to date are expert sys-
tems and natural language processors. Expert systems are especially
relevant in the quality field because they enable us to capture some
level of human expertise in decision tree form within a computer.
10 ASQC

Figure 3. Arrow diagram/PERT chart.

3 days

This results in a decision support system that provides guidance for

people in problem solving strategy, by allowing them to apply the
expertise of more experienced professionals in the field. Natural lan-
guage processors are of value to the quality field as well because they
make it far easier for people with little or no computer experience or
training to use the machines effectively. Basically, natural language
processing technology teaches computers to communicate in human
language (through voice synthesizers, optical character recognition,
etc.) so that the computer can listen, talk, and correspond using
human language.

ASQC See American Society for Quality Control.

assemble-to-forecast A production approach in which assembly or-

ders are launched based on forecasted sales, typically offset by a stan-
dard assembly time.

assemble-to-order A production approach in which products are

only assembled when an actual sales order is received. Final assembly
of the end product is never based on a forecast or as the result of a
desire for work load stability. This approach is in direct contrast to
the assemble-to-stock or assemble-to-forecast approaches.

assemble-to-stock A production approach in which final assembly is

initiated based on the need to maintain a minimum level of finished
goods (or stock) in inventory. As finished stock is depleted beyond a
designated reorder point, an assembly order for additional products
is launched.
assessment phase (phase ZZZ) 11

assembly, design for See design for assembly.

assessment phase (phase ZZZ) The third major phase of a TQM im-
plementation plan. (See the Introduction for an overview of all five
phases.) This phase involves identifying the structure, strengths, and
weaknesses of:

0 Internal organization and processes

Competition
0 External products and services
Customer
Work force skill levels

It is the primary information-gathering phase of the T Q M plan, re-

sulting in an in-depth understanding of the organization’s current
state and potential opportunities. Many companies use the Malcolm
Baldrige National Quality Award criteria as a guide for their internal
assessment activities, and others use one or more of the I S 0 9000
quality standards as a road map. While it overlaps most of the other
phases of the T Q M implementation, the assessment phase typically
begins in about the second month and runs for about two years.
(Note that it is not necessary to wait until the end of the assessment
phase before you can expect positive results from a T Q M implemen-
tation. Because of the overlapping activities with other phases, posi-
tive results often begin to materialize after about eighteen months.)
The objective of the assessment phase is to achieve a clear un-
derstanding of the organization, the products and services of the or-
ganization, the customers, and the competitors. The assessment
phase is the first phase of a T Q M program that really gets a signifi-
cant part of the organization involved and attempts to garner sup-
port and commitment from the employee base.
The assessment phase begins with a review of the company’s vi-
sion and/or mission statement(s). As the mission is driven down
through the organization, all employees must understand how their
individual roles tie into that mission, who their customers are, and
how their customersjudge whether they are receiving good value.
The internal organization should be thoroughly assessed to un-
derstand its own company culture, processes, products and services,
and internal supplier/customer relationships. A gap analysis should
then be performed, based on the company’s mission and the current
12 asset utilization

culture, to identify the desired characteristics of the future company

culture; the technologies, skills, and process characteristics desired in
the future state; and what work will be required to achieve that future
culture. Many analytical tools described throughout this text come
into play in the assessment phase, including Pareto charts, brain-
storming, and cause-and-effectanalysis.
Developing an understanding of the products and services of
the organization requires the identification of the company’s critical
process characteristics that result in the quality, reliability, price,
and maintainability of the products or services provided. This is often
accomplished through both traditional and innovative applications
of quality function deployment.
Developing an understanding of the customers’ expectations
(see expectations, customer) and level of satisfaction (see customer
satisfaction) involves the application of several tools and techniques,
including regular formal customer surveys, customer involvement in
design processes, QFD applications to external customers’ stated de-
sires, and special teams of important current and potential customers
drawn into regular design and product assessment activities.
To understand the competitor base, an organization needs to
benchmark its products and processes against those considered “best
in class” by the organization’s customers.

asset utilization The product of the availability, duty cycle, effi-

ciency, and recovery of an operation.
assignable causes See special causes.

attribute data Discrete data (nominal data and ordinal data).Typi-

cal attribute control charts are p-charts (proportion defective), np-
charts (number defective), c-charts (number of defects), and u-
charts (average number of defects per unit).
attributes tracking A summary of go/no go conditions, such as
withidnot within tolerance. Uses the c-chart and np-chart if measur-
ing a sample whose size remains constant, or p-charts and u-charts
where the sample size varies.

audit The formal and systematic inspection or review of processes

to ensure compliance to predefined requirements, or the action of
performing such an inspection.
average outgoing quality (AOQJ 13

automated material handling (AMH) The handling of material

(packaging, movement, storage, or retrieval) by automated systems
and devices. This aspect of materials management is becoming in-
creasingly popular in quality improvement programs for eliminating
data entry errors through automated identification of material (bar
coding, optical character recognition, etc.) and the automated track-
ing of storage and retrieval transactions.

automated process planning (APP) The automatic generation of

production process routings and high-level work instructions from
computer-aided process planning systems.

automation The performance of work by machines. This process

involves designing both the work tasks and the equipment involved
to support automated operations, including designed-in process ca-
pability and reliability. These attributes of automation are demon-
strated by the consistency of the process in producing goods that
conform to all specifications. See also conformance.

availability The (total time) minus (scheduled, unscheduled, and

idle downtime) divided by (the total time for the operation).

average (X-bar) See arithmetic mean.

average and range chart (X-bar and R chart) A variables control

chart that depicts sample means (averages) and ranges for assessing
and maintaining statistical control of continuous data processes. The
range chart is used to assess the variability associated with the popu-
lation. The mean (Xor X-bar) is employed to assess the central ten-
dency of the process. If the process is in a state of statistical control,
data from the two charts together may be used to assess the capability
of the process.

average (mean) chart A control chart that uses the sample or sub-
group average, X-bar, to determine the stability of the process mean.

average outgoing quality (AOQ The expected level of quality in

products from a designated process. (This concept implicitly states
that some level of quality less than 100%is expected.)
bacitjushing A systematic approach to inventory accounting that relieves
quantities from inventory based on the assumption that when the next
assembly is used (or consumed), the components required to build it have
also necessarily been consumed. Therefore, when a product is sold fi-om
the last assembly operation, all of the parts that go into that assembly
may be automatically removed from inventory by the material accounting
system. Some companies involved in just-in-time manuhcturing opera-
tions use this method to trigger payment to their suppliers.

Baldrige A z m d See Malcolm Baldrige National Quality A d .

Band-Aid appmach A popular expression used to describe responses to

problems involving temporary solutions (that usually address symptoms)
rather than permanent fixes (that address the mot cause). Examples in-
clude increasing fhbrication order quantities to overcome high scrap levels
in production, and adding lead time to the ordering cycle for purchased
parts in response to part shortages.

barbell chart See histogram.

bar coding A means of identification that uses information encoded in a

series of vertical bars printed close to one another on the surface of an
object. Data fi-om the bar code are interpreted by an electronic scan, usu-
ally attached to a computer system. This is an extremely popular means
of part, product, and package identification in many industries, and man-
ufacturing companies are finding widespread usages, primarily in inven-
tory control and labor accounting areas.

basic tools of eXperimatal &sign A group of methods commonly associ-

ated with the conduct of experiments based on modern methods, gener-
ally including planned grouping, randomization, and replication. See
also experimental design and design of experiments.
14
benchmarking 15

batch manufacturing An approach to lot sizing for production opera-

tions that produces parts in large groups to maximize the production
yielded from a single occurrence of setup cost. Batch size (or lot size) is
typically decided in these environments using a formula called ecorurmic
wub quunkgy (EOQ). EOQ divides the holding cost of inventory by the
setup cost associated with multiple production runs and establishes an
optimum lot size.

batch size See lot size.

b a t W curue S e e reliability modeling.

bell-shu@d curue/distribution A graphic display of measurement occur-

rences that, when plotted, gives the appearance of a bell shape because
the occurrences stack up on each side of a central point, tapering off in
frequency as the occurrences are farther removed from the center. The
normal distribution is an example of a bell-shaped curve.

benchmarking A technique used to identlfy performance levels in prede-

fined areas from company to company, or, less often, between sister hctor-
ies within the same company. The technique provides an opportunity for
companies to compare their own performance in certain areas against
other companies considered the best at the operation being examined.
When benchmarked companies compete for the same markets, the pro-
cess is known as competitive benchmarking. (See also competitor influ-
ences.)Benchmarking has a track record of demonstrated success, partic-
ularly among some American companies (such as the Xerox Corporation)
regarded within their industries as quality leaders. The benchmarking
process compares methods and results of processes and practices and may
be applied in both manuhcturing and service environments.
Often, companies looking for others considered “best in class” against
whom to benchmark begin with a list of recent winners of the Malcolm
B am
l National Quality Award. Some companies use the base of IS0
9OOO registered companies as a starting point. Other potential sources
include the Thomas Rigister, trade journals, and trade shows.
The ten-step benchmarking approach most commonly cited grew
out of the Xerox activity in this area during the 1980s. The steps are:

1. Identlfy those processes or practices to be benchmarked, typically

based on relevance to customer satisfaction.
16 best practice

2. Identlfir the base of companies against whom a comparison will

be made.
3. Identlfir data collection methods and collect the data.
4. Identlfir current performance levels (clear both internal and at
compared-to companies) and note differences. For every sigmfi-
cant gap in performance (positive or negative), identlfir the
causes.
5. Establish targets for hture performance levels. They should be
based on not only existing performance levels at compared-to
companies, but also the targets that companies have set for their
own future performance.
6. Communicate findings to management, process owners, and
stakeholders to garner acceptance of reality and enthusiasm for
improvement.
7. Establish intermediate hnctional goals that will support the
achievement of targets established in step 5. Solicit stakeholder
and process owner participation in improvement strategies as
they are developed.
8. Develop detailed implementation plans and assign responsibili-
ties for each sigmficant action.
9. Implement improvement activity. Monitor progress and take
corrective action as required to maintain momentum.
10. Recalibrate benchmarks and begin the process all over again.
Make benchmarking an aspect of continuous improvement.

best p.actice See operator-controllable error.

beta @) In statistical quality control the probability of making a 'I)pI1

error; that is, of accepting a hypothesis as true when it is false. For control
charts, the probability that we think the process is still in control even
though it is out of control.

bias A systematic error that contributes to the difference between an ac-

cepted reference value and an observed sample mean.

bzg Q little q An expression used to indicate the difference between man-

aging for a comprehensive quality approach in all aspects of business (big
Q and managing for quality in a limited area, such as hctory processes
or products (little 4).
blanket orders 17

biimodal distribution See double-peaked distribution.

b i d distribution The distribution of probability that a specified

number of successes will occur within a given number of independent
trials (assuming that the probability of success in each trial is the same).
This distribution relates to discrete random variables (nominal data). It
takes the form

r! (n-r)! PTr

See Figure 4.

blanket orders A purchase order for a part or a part family that

contains part numbers, part descriptions, and prices (when possible),
but no specific order quantities. Blanket orders typically indicate only
an estimated usage over the period of time covered by the contract.
Actual shipments are authorized by the buyer as the parts are

Figure 4. Binomial distribution.

Calculation of Probabilities for Binomial Distributions

Probability of Occurrence (p): 0.8 np = 6,400

Number of Trials (n): 8
Number of Occurrences (r): 0

Prob Equal &

-
r at r Above
0 0.000 1.000 0.000
1 0.000 1.000 0.000
2 0.001 1.000 0.001
3 0.009 0.999 0.010 -
4 0.046 0.990 0.056 -
5 0.147 0.944 0.203
6 0.294 0.797 0.497
7 0.336 0.503 0.832
8 0.168 0.168 1.000
18 block diagram

needed. At the end of the period covered, the order may or may not
be renewed at the discretion of either party. This type of order is
frequently used in just-in-time manufacturing environments where
operations can be streamlined and individual releases may be made
via telephone authorization or an electronic data interchange link.

block diagram A diagram made up of simple blocks (squares and

rectangles) depicting the operations and interrelationships within a
specified system. The blocks represent the system elements (major
activities or processes), and the arrows between them represent flows
of data or materials. Block diagrams are an excellent way to quickly
and effectively communicate the basic operations of a system. See also
flow chart.

blueprint diagrandtolerance See specification limits.

boundaries In a frequency histogram, lines that distinguish be-

tween each interval shown in a sequence. In operations documenta-
tion, vertical lines that distinguish between the end of one process
and the beginning of the next process or horizontal lines that distin-
guish management spans of control.

brainstorming Usually a team-based activity used to develop group

consensus on probable causes for a specified problem. The objective
is to capture as many ideas as possible. The technique involves:

1. Defining the problem clearly for the group.

2. Soliciting and recording ideas from the group about potential
causes. (Typically, a facilitator records the suggestions as
quickly as the group can call them out. No criticism is allowed
during this process.)
3. Grouping the causes into similar categories, throwing out the
ridiculous, and retaining the grouped plausible causes for
further evaluation and analysis.

breakthrough improvements Dramatic positive changes in product

or service quality or process (manufacturing, assembly, administra-
tive, etc.) performance. These changes are typically accomplished by
specifically commissioned teams using one of several disciplined
strategies working on a regular released time basis to accomplish a spe-
business plan 19

cific mission or objective. Breakthrough improvements are large,

step-function improvements as contrasted with incremental small
continuous improvements described and characterized by kaizen.

buffer inventorylstock Inventory ordered or manufactured and

stored in excess of real demands to compensate for uncertainty in
downstream manufacturing or sales. Buffer inventory may be accu-
mulated at various stages, such as at the point between subassembly
and final assembly, to soften the peaks and valleys of production de-
mand and maintain more direct labor work load stability. Buffer in-
ventory (or buffer stock) is generally regarded by supporters of
TQM, and especially by advocates of just-in-time manufacturing, to
be an example of waste that should be targeted for elimination.

business architecture The operating and organizational structure of

a business, including all major processes and systems.

business architecture model A model developed by Svenson, Wal-

lace, Wallace, and Wexler in The Qmlity Roadmp (New York: AMA-
COM, 1993) that depicts business as a cube. On the face of the cube
there are three horizontal layers consisting of business drivers, busi-
ness processes and metrics, and resource infrastructure. On the side
of the cube are four vertical slices containing assess, design, deploy,
and integrate elements. On the underside of the cube are three addi-
tional front-to-back vertical slices comprised of executive, middle
management, and working level. The developers believe that this
model of business offers an effective alternative to functional silo per-
spectives on business operation.

business drivers Factors vital to the business performance of a com-

pany and its ability to compete in the marketplace. Examples include
assets, competencies, marketplace factors, and stakeholder require-
ments. Business drivers are best defined by the customer and are
extremely useful in identifying perceived strengths and weaknesses
for strategic planning and candidates for TQM improvement initia-
tives.

business lead team See discussion under planning phase (phase 11).

business plan A company’s operating plan and a fundamental ele-

ment of policy deployment. It should contain a restatement of the
20 business processes

company’s mission statement, long-term milestones, and the metrics

associated with each milestone. These data should include progress-
to-date, remaining gaps, and strategies to be used in attaining the
milestones listed.
The business plan is generally developed and deployed annu-
ally. Once the long-range strategic plan is developed, the business
plan becomes the primary vehicle for evaluating and documenting
progress toward long-range milestones.

business processes Functional activities within an organization that

typically fall into the following three categories:

1. Group of processes that are found along the value chain. In a man-
ufacturing company, value-adding processes include prod-
uctlprocess definition, production, distribution, and prod-
uctlcustomer support. This is the most important category.
See value.
2. Support processes. Procurement, industrial engineering, tool-
ing production, machine maintenance, and receiving and
shipping are processes that are required to directly support
the value-added processes.
3. Processes that bear little direct relationship to the value-added pro-
cesses, but are still a necessary part of any business (e.g., most ac-
counting and finance processes, business development pro-
cesses, and strategic planning).

It is important to note that most processes cross the boundary

lines of multiple functional organizations within a company, espe-
cially in those companies with traditional functional organization
structures. Increasingly, businesses recognize the need to manage by
process rather than by function, and this recognition is giving rise
to more horizontal spans of control and far greater effectiveness in
management.
c-chart A type of control chart, based on the normal approximation of
the Poisson distribution, that shows the number of defects for sample
subgroups that are of equal size (see Figure 5).

01D See computer-aided design.

calibration The comparison of a measurement instrument or system

with unknown or questionable accuracy to one for which the accuracy is
known. Thii process is performed to determine the accuracy of the first
system, identlfylng the type and degree of any variation from the required
performance spedcation. In a limited sense, calibration is also used to
describe the elimination of bias from a measurement process by moving
the mean of the measurement e m r distribution to the true value of a
standard or master specimen.

cupubilii~,process See process capability.

CAPP See computer-aidedprocess planning.

cause and effect A popular expression used to describe the relationship

between a condition or event and the results from that condition or event.
There may be both direct and indirect results from any given cause. A
combination of fixtors may be required for a causal situation to occur, and
multiple effects may result from a given cause or set of causes.

cuuse-und-eflect diagram A tool used to idenufy relationships between

specific effects and their mot causes. This tool is often called the Ishikawa
diagram, the fishbone diagram, and the Ishikawa fishbone diagram. Orig-
inally developed by Kaoru Ishikawa, the diagram has the structure of a
fish skeleton when it has been completed. Properly used, the diagram
helps to determine the main causes and subcauses that have resulted in

21
22 cause, common

Figure 5. c-chart.

24

21.364
20

11.2857

1.207&
0

0 5 10 15 20 25 30
Month (1/92-4/94)
NOTE. The point below the lower h i t of expected variatron (21st month) indicates
that there was somethrng drfferent about that month that led to fewer injuries
(e g , equrpment down)

an effect (symptom)or quality problem (see Figure 6). See also dispersion
analysis.

cause, commort See common causes.

cause, root S e e mot cause.

CDE See compound document exchange.

cellulur manufacturing A production approach that uses groupings of

manubcturing equipment, tools, and people organized to perform an en-
central line or center line (CL) 23

Figure 6. Cause-and-effect diagram: maintenance costs.

Machines Communication

Indqmta U n W PoaFdback
Tranmg Schedule to Deaign

Parts Out of Inadequate

Specification Storclee
PM Not poa
Wrone Standardized Vontihtion

Materials Methods Environment

tire sequence of manuhcturing operations in one contiguous physical lo-

cation (cell).Cells are typically designed to perform all of the manuhctur-
ing processe!~ required for production of a W y (or commodity)of parts.
Cellular manuhcturing typically involves one (or both) of two basic
manuhcturing floor configurations: the U-shaped cell or the serpentine
cell. The U-shaped cell uses a minimum number of operators (often only
one) inside an area where the work moves around the worker(s) on three
sides, and the operator is standing inside the “U.”This allows the operator
to move &om position to position with only a few steps and provides the
flexibility to perform any combination of three operations without leaving
the area. The serpentine cell is more than one U-shaped cell hooked to-
gether. This configuration allows for maximum flexibility to increase and
decrease operators to account for fluctuation in work volume without re-
configuring the work or physical material flow. See Figure 7.

centralline or center line (CL) The horizontal solid line in the center of a
process control chart that represents the central tendency of the sample
values, which (if the process is in a state of control) may be directly or
indirectly employed to assess a population parameter.
24 central tendency

Figure 7. Cellular manufacturing.

U-Shaped Cells Serpentine Cells

central tendency Measures that describe the scores, values, or observa-

tions that occur most often in the data set and/or the location of the center
of the data set respective to the data set involved. Measures of central
tendency include the mean, median, and mode.

CEO/genmal manager’s role in TQM In TQM implementations, the pri-

mary activities of the chief executive officer or general manager. These
include the (1) leadership of the ongoing TQM meetings of executive
management, (2) provision of strategic direction for the overall T Q M ef-
fort, (3)provision of all required resources in support of TQM implemen-
tation, and (4) the highest-level accountability for success.

1. Leadership should include development of a vision of fbture op-

erations and motivation of the work force to attain that vision.
2. Strutegic direction involves the creation of policies, standard prac-
tices, and process-oriented management organization structures
for deployment of those policies and practices in support of attain-
certification, I S 0 9000 25

ing the vision of hture operations. It is based on a strategic plan

and analysis.
3. Resuurces include people, equipment, training, and budget for m i s -
cellaneous tools and improvements. They may even involve capi-
tal expenditures for facility alterations or consulting fees.
4. A c c u u M q involves the establishment of appropriate and mea-
surable goals and objectives and performance.

certi_fication, I S 0 9000 Third-party registration agencies’ acknowledg-

ment that an organization or specific company site has met one of three
specific standards of the International Standards organization. Certifica-
tion is performed in order to create an objective level of assurance that
the specified organization can perform in accordance with specific quality
standards, especially when the standards are cited as contractual require-
ments for suppliers. Generally, the assurance of having met these require-
ments is called registration in North America and certijication in Europe.
International standards, and particularly international quality standards
such as those represented by IS0 9O00, are an important contribution to
the TQM environment because they provide an objective measurement
standard by which potential suppliers and partners may be measured.
(IS0 9000 differs sigrdicantly from other quality measures because IS0
9000 focuses on the ability to perform to standards that may be outlined
in a contract, whereas credentials such as the Malcolm Baldrige National
Quality Award and the Deming Prize focus on continuous improvement
and worldclass quality standing.)
The two basic phases of IS0 9000 registration are:

1. Creating and documenting a quality system (evidenced by a qual-

ity manual or master manual) that meets the criteria outlined in a
specific IS0 9000 standard.
2. Certification of compliance to the selected standard by a third-
party registrar. The registrar audits the applicant’squality system
to determine whether the system is adequate and is being com-
plied with throughout the organization. Registration, once
awarded, is renewed periodically (typically every two or three
years).

The central point of contact for IS0 9000 registration is the

American Society for Quality Control. The Registrar Accreditation
26 change

Board is a nonprofit subsidiary of ASQC, and this group is responsi-

ble for accrediting registrars and publishing a directory of them.

change A modification or transformation. In a T Q M environment,

two kinds of change are required: cultural change and process
change. Cultural change is reflected in the attitudes and behavior of
employees, but is based on an underlying change in the values used
to run the company. (See company culture and organizational Val-
ues.) These values are effected by everyone, but for the most part,
they are generated by top management and flow down through the
organization. They represent the character of the company, just as
individual value sets reflect the character of that individual. Exam-
ples of commonly stated values include “The customer is always
right” and “Our people are our most important asset.” Many compa-
nies involved in T Q M embrace values that align with ideals like the
ones expressed in Deming’s 14 points.
Process changes involve altering the inputs, activities, and/or
outputs of a process to gain efficiencies. These efficiencies are gen-
erally experienced as cycle time (or throughput time) reductions,
labor hour reductions, quality improvements, or general cost reduc-
tions (such as floor space reductions or inventory reductions). Other
reasons for initiating process changes include safety enhancements,
accountability improvements, and the attainment of uniformity of
processes between sister companies and organizations within a com-
pany. Process changes are generally achieved by documenting exist-
ing process flows, brainstorming to identify potential solutions, se-
lecting and implementing the most appropriate solution, and
documenting the changes. Note that there is often an interaction be-
tween cultural and process change, and process change (improve-
ment) can sometimes be inhibited by cultural aspects. In general, cul-
tural change is far more difficult to achieve than process change.

change management The management of change in a T Q M organi-

zation, which generally takes two forms. One form is the manage-
ment of major change initiatives associated with the T Q M implemen-
tation itself. This kind of change activity consists of cultural and
process changes, as described in change. The second type of change
management is the continuous, deliberate management of company-
wide change to keep abreast of technology, environmental factors,
customer needdexpectations, and competitors’ processes. Described
CNC 27

by Duncan in Manufacturing 2000 (New York: AMACOM, 1994), this

process involves continually monitoring both internal and external
sources to identify improvement opportunities, analyzing and assimi-
lating the appropriate opportunities, and deploying the results
throughout the organization. See also competitor influences.

channels of distribution See distribution channels.

characteristic Any property of an item that distinguishes it from

other items. Items usually have many characteristics and can cover
products, services, and so forth. In TQM implementations, the char-
acteristics most commonly dealt with are physical characteristics.
These include properties such as weight (which yields the character-
istic heaviness), surface finish (which yields the characteristic smooth-
ness), and dimension (which yields the characteristic size). See also
critical process characteristics, critical product characteristics, and
external critical characteristic approval and review system.

chartering a team A formal process that defines the purpose of the

project; the measures of success that will be used to assess the project;
identification of team member’s role, sponsors, and stakeholders;
identification of project deliverables; and termdimits of empower-
ment. Also called commissioning a team.

check sheet A clear, easy-to-use form used in data collection. Custo-

mized to fit the specific application involved, the check sheet (or tally
sheet) is usually made up of potential causes identified in the cause-
and-effect diagram and is structured in a manner that allows the
frequency of occurrences to be recorded for each cause. The check
sheet is one of the seven basic tools of quality. See Figure 8.

chronic problem A problem that occurs repeatedly.

CZM See computer-integratedmanufacturing.

CZP See continuous improvement plan.

CL See central line or center line.

CNC See computer numerically controlled equipment.

28 coaching

Figure 8. Check sheet.

CHECK SHEET FOR PRINTING ERRORS

BOOK #: 120.0385 DATE: 5/7/85
INSPECTOR: &W P, LOCATION: %%&OW ?%!ht

Defect vpe Check Subtotal

poor Ink coverage 13
wrong Ink 2
wrong paper 16
incorrect cropplng 6
dupllcate pages 12
mluing pages 1
incorrect folds 0
damaged materlals 2

Grand Total 52
REMARKS:

couching A style of management, under which the behavior of an

employee is changed through education, training, data-based feed-
back, and advice to improve that employee’s work performance. This
management style is common in T Q M environments, as peers be-
come more involved in each other’s performance appraisals and as
the rewards and recognition for performance are increasingly team-
based. This approach contrasts sharply with the traditional autocratic
approaches commonly used in more authoritarian, pre-TQM set-
tings. See also employee involvement, reward and recognition sys-
tem, and facilitation.

comb distribution A pattern of measured values that, when charted

in vertical bars, looks like the edges of a comb. There is a roughly
commodity-based contracting 29

even distribution between values with higher and lower values alter-
nating regularly. This phenomenon usually indicates errors in the
way data are grouped during histogram construction. Systematic
bias in the rounding of values may yield a similar result. See Figure
9.

commissioning a team See chartering a team.

commodity Any group of parts that requires the same raw materials
and the same production resources/equipment to produce them.

commodity analysis The grouping of parts and/or materials into

families based on similarities in raw materials, equipment used in
producing the parts, or in rare cases, commonality of supplier and
application (i.e., electrical connectors and wire). This technique is
most often utilized in just-in-time procurement initiatives to improve
negotiating leverage, level work loads across groups of production
equipment at supplier facilities, and minimize administrative effort
required to manage a large, diverse group of purchased items. An
important objective of performing the commodity analysis is to re-
duce existing supplier bases to one or two suppliers per commodity.

commodity-based contracting Involves the development of contracts

with suppliers for all (or nearly all) of the parts encompassed within
defined commodities. A commodity-based contract allows for im-
proved negotiating positions for buyers and results in a mutually
beneficial contract when the object of the negotiation moves away
from specific parts and delivery quantities toward negotiating for a
percentage of the supplier’s manufacturing capacity along general
machine routings. This approach allows the supplier to be secure
over the period of the contract and provides enormous advantages

Figure 9. Comb distribution.

.
30 common causes

to the buyer in terms of delivery flexibility due to reduced supplier

lead times.

common causes Sources of variation that consistently result from a

process being measured because they are intrinsic to it. These causes
may be indigenous to the combination of manpower, methods, mate-
rials, or machines used (see four M’s), or to any one of these factors.
These causes will affect individual results as well as more general
process measurements.

communication An exchange of information. The most commonly

overlooked, underestimated, and deadly enemy of T Q M implemen-
tations is poor communication. Communication must be two-way and
a high managerial priority. The daily management board is a work-
able tool to promote and improve communication for all organiza-
tional levels. Effective communication requires a common language,
clarity of expression, and active listening. In a T Q M environment, a
common language is essential. As people learn new concepts, they
often give different meanings to the same terms. A common language
is particularly important in defining customer needs, process re-
quirements, and expected outcomes to precisely define what things
mean without resorting to nebulous conceptual jargon. For example,
terms like world class often give people different impressions and
therefore communicate very little useful information. Beyond termi-
nology, it is also important to attain common definitions of the funda-
mental elements addressed in the T Q M implementation. When a
particular process is being evaluated, it is extremely important for all
parties to have a common understanding of where the process begins
and where it ends. Communication should be a primary concern of
all T Q M implementation leaders as problems are identified, pro-
cesses are defined, expectations are stated, and improvement teams
are chartered.

company culture A company’s underlying assumptions, values, and

beliefs. These are learned responses to the organization’s external
survival and internal integration problems, they are shared by em-
ployees, and they often operate on an unconscious level-they are
taken for granted. Company culture is displayed through individual,
team, and organizational behaviors. See change for a discussion of
cultural change. See also organizational values.
compound document exchange (CDE) 31

comparative and relative importance A reference to the nature of

the outcome associated with customedconsumer research. The terms
are primarily based on the method of data acquisition employed by
the researcher.
Comparative importance assessments result from analysis of
preference or importance on a one-attribute-at-a-time basis. Likert
scales and semantic differential scales are tools commonly used in
these assessments. See Figure 10.
Relative importance assessments, on the other hand, result from
advanced consumer/customer research methods such as conjoint
analysis. These techniques provide an assessment of each attribute’s
importance (or utility) relative to all other features or attributes as-
sessed.

competitive benchmarking See benchmarking.

competitor influences Those influences exerted by competitors that

are deemed likely to greatly affect the future of a company. These
typically include both deliberate and accidental phenomena, which
means that they are only somewhat predictable. Examples of deliber-
ate actions include the merger of competing companies and the in-
troduction of new product lines by competitors. Examples of acciden-
tal effects include serious defects in competitors’ products that
significantly affect their market share and natural disasters that se-
verely impact the competitors’ production capability or capacity.
Competitor influences are an important element for consideration by
strategic planners and in competitive benchmarking activities. See
also market share, importance of.

compound document exchange (CDE) The electronic exchange of

documents containing more than one of the following elements:
graphics-, text-, audio-, and video-based data. CDE is an important

Figure 10. Comparative and relative importance assessment scale.

Not very
important important
1 -
(Attributel) 1
Color
1 2 3 4 5
32 computer-aided design (CAD)

element in the development of effective ways to use the information

superhighway. The most common standard used for CDE in future
business environments is likely to be computer-aided acquisition and
logistics support (CALS).

computer-aided design (CAD) The use of computers to design prod-

ucts, typically in place of the traditional drawing board, pencil, and
vellum or paper. CAD systems became a widespread design tool in
the 1970s and have evolved through the 1980s and early 1990s into
tools that can produce not only two-dimensional drawings, but also
three-dimensional wire frame views and shaded three-dimensional
models that can be rotated in simulation on screen in real time. Along
with enhanced capabilities has come the need for increased comput-
ing power, and the developments in CAD software have closely fol-
lowed the power and speed evolution of computer hardware.

computer-aided process planning (CAPP) A system typically consist-

ing of part descriptions, manufacturing process information, tooling
information, and in some cases, labor standards information. The
purpose of these systems is to gain efficiencies in the development
of manufacturing processes. CAPP systems can be useful on a part-
by-part basis, but are especially useful when applied to part families
or commodities.

computer-integrated manufacturing (CZM) An approach to manu-

facturing management that focuses on shared information via com-
mon databases. It may be viewed simply as the use of computers and
information technology to integrate and automate manufacturing
processes and the processes that support manufacturing. Integra-
tion in this context means that all of the information needed to per-
form any process in the enterprise that is generated by or held any-
where within the enterprise is immediately and constantly available
through automated linkages. In the early years of computer-inte-
grated manufacturing research, CIM was thought to require a single
common database for the entire enterprise. As technology has
evolved, however, networks, servers, and related information tech-
nology system components have offered effective alternatives to the
single database approach. However, the fundamental concept of a
fully integrated information technology infrastructure in any busi-
ness operation remains valid and will continue to be an extremely
configuration management 33

important tool for businesses in order to remain competitive

throughout the foreseeable future.

computer numerically controlled (CNC) equipment Equipment, usu-

ally manufacturing equipment such as metal punches and lathes,
controlled directly by computer through a symbolic language. Types
of CNC equipment include robots, automated guided vehicles, and
any other equipment that may be directed by computer. As factory
floor machinery becomes more sophisticated, the skills and technical
sophistication of operators must increase as well. Through the bal-
ance of the 199Os, the application of CNC equipment will be an im-
portant factor in staffing decisions, and will present new challenges
and opportunities in terms of process control. Also sometimes called
computerized numerical control equipment.

concurrent engineering The simultaneous design of products and

their associated life-cycle management processes, including manu-
facturing, distribution, and even obsolescence-related disposition ac-
tivities. Participants in concurrent engineering teams often include
design engineering, manufacturing engineering, quality assurance,
procurement, distribution, marketing, and even supplier personnel.
Concurrent engineering has been embraced as a means to reduce
cycle-to-market and the development-related financial risk. When
done effectively, it also ensures that all customer needs are consis-
tently and effectively met through the design process. See also con-
figuration management.

conjigurution management The effective control of product struc-

tures. Configuration management is an extremely important aspect
of concurrent engineering and manufacturing operations since it
substantially affects the efficiency of manufacturing operations, in-
ventory levels, and customer satisfaction. Properly configured prod-
ucts minimize redundancy and non-value-added labor in manufac-
turing operations. They also prevent ordering expensive product
components earlier than necessary to support manufacturing se-
quences. In addition, properly configured products ensure that cus-
tomers receive exactly what they ordered. As we move through the
last half of the 199Os, configuration management will be a fertile area
for the application of expert systems and knowledge-based systems.
Companies with excellent configuration management will certainly
34 conformance

be perceived by their customers as having quality levels that are su-

perior to their competitors.

conformance Concordance with predetermined requirements,

meaning that the product or service has met the relevant specifica-
tion and contract requirements.

conformance quality The consistency with which each unit of prod-

uct delivered, or each aspect of service provided, is faithful to the
design specification and relevant target value.

conjoint analysis An advanced and statistically based approach to

the assessment of customer or consumer preferences associated with
attribute, feature, or element analysis for a product or service. In this
type of research, fractional factorial diagrams (often of the classifica-
tion called orthogonal arrays) are used to create a series of product or
prototype descriptions. The descriptions are then assessed by a Sam-
ple of customers or consumers who rank the descriptions from most
to least desirable. The mean rank values are subsequently used in
an analysis of variance to generate utility indices that provide an
indication of the relative importance of the attributes or features as-
sessed. Virtually any product or service may be assessed in this fash-
ion. See also comparative and relative importance.

conscious error See operator-controllableerror.

consensus The understanding and commitment of an entire group

on a concept, approach, or strategic direction or intent. Achieving
consensus among members is essential to effective team operations.
Consensus, however, does not imply (as popularly portrayed) total
agreement, but total support. As Margaret Thatcher said, “Consen-
sus [referring to the goal of 100% agreement] is the abrogation of
leadership.’’ There are several techniques that may be used to
achieve consensus. Two of the most popular tools are brainstorming
and the nominal group technique.
Often, some individual group members remain privately uncon-
vinced that the approach adopted by the entire group is the optimum
solution, but agree to go along with the group decision. This situation
leads to an apparent consensus but not a true consensus and is often
dangerous because these group members have a way of undermining
continuous improvement 35

progress during downstream implementation activities. Team lead-

ers and facilitators should be wary of this condition and make every
effort to remedy it through education, persuasion, or replacement of
team members.

constraint management A process of identifying constraints or bot-

tlenecks in manufacturing operations, then making adjustments in
the process to prevent recurrence of each constraint. As knowledge-
based systems are more frequently applied, expert systems will han-
dle constraint management as a normal part of administrative over-
sight functions. The principles of constraint management apply in
almost any business.

consultant’s role in TQM The part that consultants play in effective

TQM implementation. The primary role of the (external) consultant
is to bring profound knowledge (Deming) to the organization at-
tempting to develop a TQ culture. As implied by Deming, a group of
people in a pit cannot dig themselves out of that pit. They may dig
harder, or faster, but the pit simply gets deeper. The role of the (ex-
ternal) consultant is, in essence, to bring the ladder and to train the
people in the organization (pit) to use it on their own.
Roles, in the context of the provision of profound knowledge,
where consultants can be extremely valuable in T Q M implementa-
tions, include planning of the overall TQM implementation activity;
general TQM training, especially for initial implementation teams
and executive management; specific technical training for implemen-
tation team members; and facilitating initial team meetings where
team building and objectivity are critical.

continuous data Data that are quantifiable, measurable, and can be

classified according to some continuum such as ounces, meters, or
feet. See also interval data, ratio data, and variable data.

continuous improvement The improvement of products, processes,

and/or services on an ongoing basis. The gains made through contin-
uous improvement activities are generally incremental, small-step
improvements, as contrasted with more dramatic and sweeping im-
provements typically associated with initiatives such as policy deploy-
ment. In Japan, the continuous improvement process is often called
kaizen.
36 continuous improvement plan (UP)

continuous improvement plan (CIP) A plan designed to incorporate

the philosophy of continuously improving every process and prod-
uct into the culture of a company. (See company culture.) The plan
should outline specific training and improvement guidelines. The
training included in this plan should prepare participants to seek out
root causes in problem solving and stress prevention as opposed
to detection of defects. The plans typically include initial awareness
training; more detailed problem solving and T Q M tool training; ini-
tiation of improvement activities; and the monitoring of regular, on-
going improvement activities, including reward and recognition sys-
tems to reinforce the application of continuous improvement
throughout the company. See also detection vs. prevention.

Contracting, commodity-based See commodity-based contracting.

control An approach to process quality that involved the identifica-

tion of changes in processes due to special (external) causes of vari-
ability so as to allow the process owner(s) to take appropriate action.
(See special causes.) In those cases where the special causes of varia-
tion are positive, the process owner(s) attempts to make those effects
common to the process. In the event that the special cause of varia-
tion is negative, the process owner(s) has the responsibility to elimi-
nate the effects and take those actions necessary to prevent their re-
currence. These actions are accomplished through the use of process
control charts (called Shewhart charts in select applications) and may
be applied to critical quality characteristics (satisfiers),defects, or de-
fective unit monitoring. Further, the (statistical) control process may
be utilized in conjunction with product quality, product perform-
ance, service, administrative, and delivery applications.

control chart A chart containing control limits and a plot of statisti-

cal values from a series of samples or subgroups. (See also upper
control limit and lower control limit.) The chart also contains a cen-
ter line to help detect trends away from the process center and
toward either control limit. The control chart is one of the seven
basic tools of quality. See Figure 1 1 .

control group A group of subjects matched with an experimental

group, but not subjected to the action being taken in the experiment.
The control group is used to more clearly demonstrate the effects of
convergence, sequential 37

1 Expected Value
Due to Chance
(Sampling Error)

Recorded Data

the treatment applied by contrasting its condition at the end of the

experiment with the condition of the experimental group. Since the
experimental group is exposed to some treatment and the control
group is not, a common assumption is that the differences in the
condition of these two groups at the end of the experiment result
from the application of the treatment to one group and not the other.
control limits The boundaries (represented as horizontal dashed
lines) on a control chart that define the limits within which statistical
values calculated on subgroups of data will naturally fluctuate when
the process is in a state of control. When in a state of control, a
process is influenced only by common causes of variability, and as
such, the process output is stable and predictable and should reveal
only a random pattern of variation through time. The observation of
points outside the control limits, or of specific (nonrandom) patterns
of variability within the control limits, is an indication of the presence
of special causes of variability, and in such cases, the process is iden-
tified as out of control. Control limits are calculated using (historical)
process data and are not associated with specification limits (or tol-
erances) or desired levels of performance. See lower control limit
and upper control limit. See also in-control process and out-of-con-
trol process.
convergence, sequential The combination of two or more technolo-
gies over a substantial period of time, resulting in a third type of
38 convergence, simultaneous

technology or a new field of technology application. For example,

composites technology, which had already been in fairly widespread
use for many years, converged with the relatively new field of robot-
ics. A resulting application technology is composite robot compo-
nents. See also convergence, simultaneous.

convergence, simultaneous The combination of two or more technol-

ogies within a relatively short period of time, resulting in a third type
of technology or a new field of technology application. For example,
the development of highly pure plastics converged with the develop-
ment of fiber optics technology, which came along about the same
time. A resulting materials technology application is plastic fiber op-
tics. See also convergence, sequential.

COQ See cost of quality.

corrective action The act of implementing a solution that removes a

root cause and lessens a known problem.

correlation coefficient A value that represents the relationship be-

tween two or more variables. These indices may indicate indepen-
dence between the variables measured, or may suggest a relationship,
positive or negative, between the elements measured, although the
presence of a relationship does not guarantee that a causal associa-
tion exists.

cost drivers Aspects of an activity that generate cost. These elements

are usually identified within an environment of activity-based cost-
ing, during a process called process value analysis (PVA). There are
two categories of cost drivers: primary (highly qualitative) and vol-
ume-based (highly quantitative). The costs associated with volume-
driven factors are usually readily identified within the cost account-
ing system, whereas those associated with the primary drivers are
not.

Primary Cost Drivers include:

communication redundancy
availability (of tools, material, data, and people) 0 priority
0 complexity 0 timeliness
costs, appraisal 39

Volume-Based Cost Drivers Include the Number of:

inspections transactions
0 material moves setups
0 orders receipts
0 shortages

costing, activity-based See activity-based costing.

cost/scheduZe/quality relationship A widely held (but incorrect)

axiom in pre-TQM environments that there is a fixed relationship
between these elements and that an increase in any one of the factors
would generate a corresponding increase in the other two areas. In
reality, increased quality levels lower cost and improve schedule per-
formance. A fundamental flaw in the logic that produced the axiom
was the definition used for the word quality. Managers believed that
quality could be “inspected in,” which merely increases cost and
lengthens schedules to improve process yields. Quality is produced
by adequate process controls, preventing defects, and reducing
downstream inspection, rework, and scrap costs.

cost of quality (COQ A term used to describe the total cost associ-
ated with the quality of products or services. Costs of quality include
prevention costs, appraisal costs, and internal and external failure
costs. (See internal failure costs and external failure costs.) These
costs are usually much greater than those immediately recognized
because they include systems costs and personnel costs usually con-
sidered fixed. Among the least recognized elements are engineering
changes resulting from poor-quality designs and customer relations
costs associated with poor-quality products and services.

cost-price model An equation used to describe the relationship be-

tween product cost and the price of that same product. In pre-TQM
environments, this equation is often thought to be: Cost +Desired
Profit = Price. In TQM environments, the equation is more correctly
defined as. Market Price - Cost = Profit. This approach shifts the
emphasis from a producer-based emphasis to a customer focus and
recognizes that success will hinge on reducing cost from market price
through process control, stability, and improvement.

costs, appraisal See appraisal costs.

40 counseling

counseling A management style and technique that is supportive of

the employee, assisting the employee in identifying and working
through personal and professional problems that are adversely af-
fecting job performance.

covariate See independent variable.

C, A statistical index value that reveals the capability potential of a

process. This index takes into account only the variability of a pro-
cess relative to the specifications or tolerances. The C, index is a
“bigger is better” index with a value of 1.0, indicating minimum ac-
ceptable capability, although some organizations require at least
1.33. C, values less than 1.O indicate that the process lacks the poten-
tial to be capable. This index should only be calculated for processes
running in a state of control. See in-control process. See also process
capability for a full discussion and computational information.

C, A statistical index value that reveals the capability of a process,

which takes into account not only process variability (as does the C,
index), but also considers the degree to which the process average is
centered between (bilateral) specification limits. In its original form,
c , k is specifically associated with the percentage of individual obser-
vations expected to be out of specification on the side of the distribu-
tion where the capability is the worst. C,k is at its maximum value
(which is equal to C,) when the process average is exactly halfway
between specification limits. As with C,, this index should not be cal-
culated unless the process has been documented to be running in a
state of control. See in-control process. See also process capability
for a full discussion and computational information.

C, A statistical index value which, like Cpk, reveals the capability of

a process. However, unlike c p k , which assesses capability as a func-
tion of both variability and the degree to which the process average
is centered between (bilateral) specification limits, C,, assesses capa-
bility as a function of variability and the degree to which the process
average deviates away from the designated ideal or target value. C,,
takes on its largest value when the process average is equal to the
target value or nominal value (No).As with C, and c p k , C,, should
not be calculated unless the process has been documented to be run-
critical path method (CPM) 41

ning in a state of control. See in-control process. See also process

capability for a full discussion and computational information.

CPM See critical path method.

CQA See customer quality assurance.

criterion measure A method or scale by which a dependent (or re-

sponse) variable is quantitatively or qualitatively assessed. For exam-
ple, the dependent variable in an experiment might be identified as
(6
metal can quality.” This dependent variable, however, might pos-
sess numerous characteristics to be measured, analyzed, and bal-
anced within a single experiment. Trim height, axial load, metal ex-
posure, and the presence of voids might constitute the criterion
measures associated with the dependent variable of “metal can qual-
ity” for a given experiment. If only one criterion measure exists, the
dependent variable and criterion measure are typically the same
quantity.

critical mass The number of people in any organization who must

be actively involved in a transformation to assure that the change
achieves long-term sustainability.

critical path method (CPM) A technique for identifying all of the

activities within a process, along with their sequences and interde-
pendencies. The longest (in terms of time) chain of activities that
must occur in sequence (cannot be done in parallel) within a process
is the critical path of that process. The identification of the critical
path is a subset of the activities performed as a part of the program
evaluation and review technique. Identifying the critical path in any
process is essential in reducing the throughput time for materials
produced through that process. Identifying the critical path in a
manufacturing process typically involves these steps:

1. Identify all of the activities in the process.

2. Beginning with the last activity performed in the process,
work your way backward through the process, identifying
which activities must be completed before the one you are
reviewing may be started.
3. Document the dependencies. This is usually done with circles
42 critical process characteristics

depicting the activities and arrows between the circles depict-

ing the interrelationships.
4. Identify and document the throughput times associated with
each activity, labeling the circles with their respective
throughput times.
5. Sum the total throughput times for each sequence spanning
the first through the last activity.
6. Identify the chain of activities that represents the longest total
throughput time, based on step 5. This chain of activities is
the critical path. See Figure 12.

critical process characteristics Process characteristics or variables

such as temperature, pressure, feed, and run speed that are shown
to have a significant and important effect on an in-process or end-of-
line critical product variable satisfier or dissatisfier. Importance is
assessed on the basis of the calculation of the amount of variability
or influence that the process characteristic explains as related to the
process variable. This degree of importance may only be assessed as
a result of a designed experiment.
Critical process variables may also be categorized as first-, sec-
ond-, or third-order characteristics. First-order variables are those
that directly impact a product characteristic. Second-order process
variables are those that directly affect a first-order process variable,
thereby indirectly affecting the product characteristic in question.
Third-order process characteristics are critical influences as they di-
rectly impact critical second-order variables.

Figure 12. Critical path method.

NOTE: The critical path is highlighted. requiring a total of 79 days.

curue, bell-shaped 43

critical product characteristic A quality characteristic (or satisfier)

that is significant and important to the customer as related to form,
fit, function, use, or safety as associated with the part or product sup-
plied. These characteristics will, at a minimum, possess a nominal
and/or target value, and at least one specification that will describe
the minimum acceptability level of the product, part, or component.
Typically, the nominal value and/or target value and specification(s)
are and should be based on engineering design, testing, and analysis.
See Figure 13.
cross-functional team A special-purpose group composed of per-
sons from different areas (cutting through the organization horizon-
tally rather than vertically) that has been commissioned to address a
specific issue or subject.
cultural change See change.
culture (company or corporate) See company culture.
curve, bell-shaped See bell-shaped curve and normal distribution.

Figure 13. Critical product characteristics.

uLI
Thickness A
I)
No Effect on
Requirements
Insignificant
Customer I) Product
Variable

B vs C Yields S' nificant

A I) E~~~~~ I) &Not
Yields No Important
Thickness A

8 vs C Yields
Effect: A vs B Significant
Meaningful Critical
Thickness A Difference

I A, 8,C = Parts A = Nominalflarget I

44 customer, external

customer, external An individual (or group of individuals) outside

the organization who are the recipients of products or services of that
organization. Examples include product consumers, clients, local
communities, and the general public. See also customer, internal.

customer, internal An individual (or group of individuals) inside an

organization who is the recipient of a product or service produced
within that same organization. See also customer, external.

customer delight The pleasure customers get when the delivered

product or service meets their needs or requirements and exceeds
their expectations.

customer expectations See expectations, customer.

customer feedback See feedback, customer.

customer focus Constant awareness of customer needs, and a con-

sistent effort to satisfy and exceed those needs. Adequate customer
focus requires that customers and the needs of those customers be
discretely identified and addressed and provides the opportunity to
measure the degree to which the needs are satisfied.

customer map A perceptual map; an illustrative device used to dis-

play the results of a consumer/market research study. The map is
generally organized in a configuration designed to display how an
individual customer assesses the critical product or process charac-
teristics associated with a supplier’s performance in two dimensions:

1. How the supplier is performing relative to competitors pro-

ducing the same product or providing the same service
2. The degree of importance associated with each critical char-
acteristic as perceived by the customer

The customer map may be used to display the results of such

research for a single customer or for a common group of customers.
Further, both products and services may be displayed. The primary
end use of this device is to provide an illustrative basis for planning
and policy deployment. See Figure 14. See also critical process
characteristic and critical product characteristic.
customer needs 45

Figure 14. Sample configuration for a customer (perceptual) map.

CRlT ;AL
TO THE C STOMER

Opportunity Area Firmly Entrenched

for Strategic Attributes
initiatives (candidates for
(candidates for daily management
policy deployment) and kaizen)

WORST BEST
IN CLASS IN CLASS

Secondary Potential
Opportunity Area Attributes for
for Future Focused
Strategic initiatives Differentiation

NICE TO HAVE
BUT NOT CRITICAL TO THE CUSTOMER

customer mapping process A basic customer research approach that

is also known as customer window, voice of the customer, and listening to
the customer. In this process, customer representatives are interviewed
or surveyed in order to assess their perceptions of the products and
services of a particular supplier. Additional information gleaned
from this process may also include customer assessments of what the
supplier might consider improving. The outcome of this research
approach, which is most functional for dissatisfier analysis, may be
the development of a perceptual or customer map. See also critical
process characteristic and critical product characteristic.

customer needs An important feature of a company’s strategy and

quality system. Customer needs help to define quality, since quality
products conform to customer needs, translated through the busi-
ness to mean conformance to design specifications. Customer needs
are identified in several ways, including market surveys, analysis of
buying patterns, and the Delphi technique. They are defined and
46 customer quality assurance (CQA)

translated into specific business actions through quality function de-

ployment. Note that while a customer needs a product within speci-
fications, the same customer wants a product at the optimum target.

customer quality assurance (CQA) A supporting element of total

quality assurance. The components of customer quality assurance
include advanced quality planning, customer plant integration,
tracking the voice of the customer, and new product introduction/
marketing. See Figure 15. See also customer mapping process.

customer satisfaction The result of meeting customer needs or re-

quirements when delivering a service or product. See also dissatis-
faction.

customer satisfaction measurement A process, typically owned by ex-

ecutive management, in which customer satisfaction data are contin-
uously gathered (through customer surveys, interviews, sales force
feedback, telemarketing feedback, etc.) and analyzed to identify both
levels and trends of customer satisfaction with the organization’s
products and services. See also survey-feedback process.

customer window See customer mapping process.

cycle-to-market The time required from concept to initial product

delivery, including all design, production, and distribution activities.

cycle time In cycle reduction activities, the elapsed time of a process

from beginning to end. This elapsed time includes both value-added
and non-value-added activity. In machine operations, cycle time is
more accurately defined as the time required for the machine to cycle
through one recurring sequence of activities. See also value-added
analysis, concurrent engineering, and lead time.
Figure 15. A breakdown of customer quality assurance.
Strategic Business
Plan Plan

Customer
Quality Stratv
Product- arket
Analysis
Competitive
Analysis
Assurance
Us Competition
Current
Customer
x x (Both) Time
Frame
Product Design Ongoing
Feedback
System
& Planning x x (Both)

Us and the
Product Competition
Performance
Product
price Dlssatisfiers A P Process
Owner
Delivery Service C D
Product
Quality
 D
D-chrt See demerit chart.

dudy m n u g m One of the three integrated management technolo-

gies supporting TQM. It is intended to prevent backsliding and enable
continuous improvement in the safety, quality, delivery capability, and
cost of processes. Daily management hnctions and activities encompass
communication, focal point item management, daily control (operation4
tactical), and crisis management. See also total quality management and
focal point chart.

dudy managentent board Visual display tools designed to convey infor-

mation or feedback of a specific nature to a specific target population.
TQM-based organizations are relatively “flat” and decision making is car-
ried out at lower levels by informed and empowered teams and employees
managing their own processes. Daily management boards provide the
vital information needed by these employees in their efforts to control and
improve processes within their span of control.
Daily management boards typically display a department mission
statement and focal point chart and show the current status of each focal
point item. They also have a section on employee involvement and a
technical section that monitors critical tactical metrics for the department.
Note that other groups, not only departments, use daily management
boards as well (e.g., a plant, a team). See Figure 16.

dumugingpuctice See operator-controllable error.

data,attribute See attribute data.

data,continuous See continuous data.

48
data collection 49

Figure 16. Examples of daily management boards.

I Board Name1 Purpose I Target Population

Department To convey feedback information Members of
Daily on the progress of the organization the department
Management toward achieving critical targets
Board and goals and continuously
improving performance
Project To explain, describe, and monitor Employees affected
Management the purpose and progress of a
Board critiial project those
by the imp ementing
and it
Scheduling or To describe the sequence, timing, or Employees affected
Production provide critical informationconcerning by or erforming the
Control Board the scheduling of changeovers, sche&led activties

r
assianments. deDbvments. etc.
Workstations To displa critical information needed Workers running
Y
to contm and monitor a process or
o eration (control charts, SOPS,
the process or
operation
cEeck sheets)
Kanbans To control the flow of material Workers and
through a series of operations supervisors
or processes running the area

data coaection The gathering of data to iden* methods of, and prog-
ress toward, process improvement. A sound strategy for data collection
includes the answers to these questions:

1. What do we plan to learn from the data?

2. What specific data are needed?
3. What specific metrics will be needed to document the data?
4. At what points in which specific process must the data be col-
lected?
5. What sampling methods will be used?
6. Who will be responsible for collecting the data and how fre-
quently?
7. How will the data be reported, to whom, by whom, and how fre-
quently?

Gathering too much data and gathering the wrong data can be
counterproductive. Faulty analysis of even flawlessly gathered data
can be misleading. However, good, solid data-gathering strategies
and tactics are essential to monitoring and improving process and
system health.
50 data encryption standards (DES)

data encrypion standards (DES) Standards utilized to secure the

data communicated between two or more computers or telecommu-
nications devices. DES devices are typically comprised of a single
computer chip used for encoding and decoding that is installed at
each end of a communication line.

data interchange See electronic data interchange.

data sharing See advanced procurement technology.

decision matrix A matrix-based chart used to assign relative priori-

ties among corrective actions. Problems are listed on the Y axis of
the chart, and potential solutions are listed on the X axis, at the top
of the chart. Each corrective action (potential solution) is then rated
within vertical cells of the matrix corresponding to each problem.
Ratings for potential solutions are summed to assign relative priority.
See also comparative and relative importance.

decision support systems (DSS) Systems (typically software) designed

and deployed to support human decision making in some technical
area. Decision support systems are constantly growing in application
base, sophistication, and importance. They will eventually take over
much of the redundant analysis in low-level decision making in busi-
ness organizations. Initial applications, which are likely to become
flagships for future development work, include inventory planning
and management, sales and market analysis, and transportation
routing activity. See also artificial intelligence.

defect Nonconformity to specification, including any failure to sat-

isfy an intended use. (See also conformance.)A part may have any
number of defects. Defects may be categorized as:

Critical.Capable of resulting in serious injury or loss of life.

Sem'ous. Capable of resulting in catastrophic economic loss.
Major.Capable of resulting in product or service failure under
normal usage conditions.
0 Minor.Capable of resulting in minor problems during the nor-
mal, intended use of the product or service, and/or some eco-
nomic loss.
Deming Prixe 51

defective Term used to identify any product or service that fails to

conform to specifications and/or does not satisfy its intended pur-
pose or function. See also conformance.

degrees of freedom (df) Generally, the freedom of scores to vary.

This term also refers to the number of observations that can vary
independently of one another. Often, conditions exist that impose
restrictions such that the number of degrees of freedom is less than
the number of scores in the sample. In random samples, the degrees
of freedom are equal to the sample size minus one.

delight See customer delight.

Delphi technique A subjective forecasting method that involves the

use of a panel of experts brought to group consensus on the most
likely course of a specified subject. In the Delphi process, each indi-
vidual expert develops his or her own forecast on the subject and
the forecasts are submitted anonymously to a group facilitator. The
facilitator reviews the forecasts and prepares a summary report,
which is returned to all the Delphi experts. The experts then review
the summary report and prepare a revised forecast based on the
summary. The process is repeated until the forecasts converge.

demand elasticity Term used to describe an environment where

market demand for a product or service substantially increases when
the price of the product or service is reduced. When a price decrease
results in a sales increase (measured as a percentage) greater than
the price reduction (again measured as a percentage), the demand is
considered elastic.

demerit chart (D-chart) A type of control chart using attribute data

that show the number of defective units weighted by their impor-
tance. See Figure 17.

Deming cycle See PDCA cycle and ShewhadDeming cycle.

Deming Prize A quality prize awarded annually to selected organi-

zations that, according to award guidelines, successfully applied com-
panywide quality control based on statistical quality control and
have demonstrated a commitment to maintain quality systems. The
52 Deming Prize

Figure 17. Demerit chart (D-chart).

12 -l " " " " i ' " ' ~ ' ' " l ' " ' ~ ' " ' ~
-

..................................................... ............................... UCL

10 -
-

- -

- - -
D

4 - -

2 - -

...................................................................................... LCL
0 - -
l ~ ~ . . i . ~ ~ . l ~ ~ ~ ~ i ~ .

0 5 10 15 20 25 30

Month (1/92-4/94)

1 = fint aid injury

5 = serious injury (medical attention required)
7 = lost time injury

award is named after W. Edwards Deming, though its criteria are not
directly aligned with Deming's teachings. The three separate award
categories are the Deming Application Prize, the Deming Prize for
Individuals, and the Deming Prize for Overseas Companies. The De-
ming Prize Committee of the Union of Japanese Scientists and Engi-
neers (JUSE) in Tokyo oversees the award process.
design for disassembly 53

Deming, W. Edwards A renowned quality leader, consultant,

teacher, and author. Deming shared his knowledge of statistical
quality control to help the U.S. war effort during World War I1 and
was then sent to Japan in 1946 to help the nation recover from its
wartime losses. He authored more than 200 works, including ac-
claimed books such as Quality, Productivity, and Competitive Position and
Out ofthe Crisis.Deming’s 14 points still represent a widely respected
set of quality management guidelines.

dependent variable The variable affected in an experiment; also

called response variable. This variable represents the outcome or end
condition under study. It is usually the factor that the research is
aimed at understanding and/or controlling. Examples of responses
(or dependent variables) include trim height, cleanliness, and time to
failure. Virtually any measurable customer satisfier (quality charac-
teristic) may be studied as a dependent variable. See also criterion
measure.

deployment phase See implementation phase (phase IV).

DES See data encryption standards.

descriptive statistics Values that describe a sample or subgroup in

terms of central tendency, dispersion, and shape. See also param-
eter.

design, experimental See experimental design, design of experi-

ments, and basic tools of experimental design.

design for assembly (DFA) An approach to the design of manufac-

tured products characterized by a deliberate intent and consistent
actions to facilitate assembly of the product. This approach usually
focuses not only on ease of fit-up and clarity of assembly sequence,
but also on minimization of required tools for assembly tasks.

design for disassembly An approach to the design of manufactured

products characterized by a deliberate intent and consistent actions
to facilitate the disassembly of the product at the end of its useful life.
(This is usually done for disposal, remanufacturing, or recycling.)
54 design for lhgistics

design for logistics An approach to the design of manufactured

products characterized by a deliberate intent and consistent actions
to facilitate the physical handling and distribution of the final prod-
uct through delivery to the end customer.

design of experiments (DOE) A branch of applied statistics in which

controlled tests are planned, implemented, and interpreted in order
to identify and assess the importance of those factors affecting the
value of one or more parameters associated with a dependent (re-
sponse) variable or condition. Specifically, a designed experiment is
intended to allow the researcher to efficiently manipulate indepen-
dent variables in such a way that effects noted on the dependent
(response) variable are correctly attributed and quantified. Methods
of manipulation include incorporation, nesting, blocking, control,
and limitation. In industry, the designed experiment is constructed
to maximize the information gained at the minimum time and cost
possible, while obtaining internally and externally (i.e., generalizable)
valid results. See also experimental design and basic tools of experi-
mental design.

design quality Accurate identification of critical quality characteris-

tics (satisfiers)and their appropriate nominal or target values that, if
met, will result in optimal product or process performance. Optimal
performance is defined by stated performance and reliability re-
quirements. All stated requirements should directly or indirectly re-
late to the needs of the customer. See also customer needs.

design review A systematic search for design problems conducted

at designated intervals throughout the product design process. The
purpose of the review is to identify problems, identify alternatives to
resolve those problems, and assure performance to quality, cost, and
schedule milestones, and in so doing, to assure design quality.
Among the benefits of the design review are:

1. Provision of an organized forum for the discovery and exami-

nation of product problems and issues
2. Provision of consistent direction for all working levels and
identification of specific responsibilities and assignments
3. Deployment of design and development resources on the
most critical problems in order to assure timely resolution
detection us. prevention 55

4. Minimization of the potential for a problem to be overlooked

5. Application of the collective experience of all affected func-
tions and departments
6. Enhanced coordination of activities within the design process
and with other nondesign activities

design verification Term used to describe a series of tests required

to qualify new/modified designs. The tests are compiled in a docu-
ment called a design verification plan, which is the responsibility of
design engineering. The design verification plan is revised as needed
to account for changes in operational requirements, customer expec-
tations, and government regulations. (See also expectations, cus-
tomer.)The plan is developed annually, or as significant changes are
made in product design. It covers test specifics, acceptance criteria,
sample sizes, and completion dates. It also indicates which individu-
als are responsible for each designated activity.

designer materials A class of new materials that are formulated and

constructed to exhibit specific properties in use as part of a product.
Composites and other polymers, as well as specially developed metal
alloys, are typical representatives of this category of material.

designing in quality vs. inspecting in quality See detection vs. pre-

vention.

destructive testing Any form of product testing that renders the

product unusable. Most commonly, this type of testing is used to eval-
uate structural strength and durability. It may involve using the
product until its useful life has been exhausted, or even cutting the
product in half to evaluate internal characteristics. See also nonde-
structive testing.

detection Identifying a nonconformance after it exists.

detection vs. prevention A comparison of two types of quality activi-

ties: identifying a nonconformance after it exists (detection) versus
not permitting it to occur in the first place (prevention).One of the
most important aspects to understand about TQM is that prevention
is far less costly than detection. Process control and improvement
are the only ways to prevent defects. Improving processes improves
56 deviation

product quality levels, and improving processes means reducing

variation in those processes around their optimum targets. This
makes process control the least costly and most effective method of
ensuring quality. See also nonconformity,conformance, and preven-
tion costs.

deviation Any nonconformance in a product or service from the

specifications for that product or service. A product may deviate
from the stated specification, or nominal position, without being out-
side of stated tolerances. In this case, the product deviates without
necessarily being defective. See also nonconformity and nominal
value.

df See degrees of freedom.

DFA See design for assembly.

diagram, blueprint See specification limits.

diagram, fishbone See cause-and-effect diagram and dispersion

analysis.

disassembly See design for disassembly.

discrete data See attribute data.

dispersion The distribution of scores or values within a data set or

within a population of occurrences.

dispersion analysis A method of constructing cause-and-effect dia-

grams. The three steps involved in a dispersion analysis are:

1. Identify the quality characteristics to be improved. These should

be primary customer satisfiers/dissatisfiers. List them as main
branches on a fishbone or cause-and-effectdiagram. (If participants
have difficulty organizing their ideas in this manner, they may start
with the standard manpower, materials, methods, and machines elements
common in the Ishikawa fishbone diagram. See four M’s.)
distribution channels 57

2. Brainstorm ideas of possible causes of defects in the quality charac-

teristics designated in step 1. List them as secondary branches stemming
from the main branch categories listed.
3. Brainstorm all possible causes of the main cause categories listed from
step 2. Show them as minor “twigs” stemming from these secondary
branches.

The completed fishbone diagram can then be used in identifying

most common causes and deploying corrective action and defect
prevention resources in the most fertile areas of opportunity.

dissatisfaction The state one (usually a customer) experiences when

his or her expectations are not met. This term is usually used within
the context of customer satisfaction. When used in this context, cus-
tomer satisfaction is considered to be a continuum, with dissatisfac-
tion on one extreme of the continuum and customer delight on the
other extreme. Satisfaction should be viewed as a point of indiffer-
ence in the middle of the continuum. (Note: Many customer satisfac-
tion surveys seek responses to questions or statements on a scale
ranging from “very dissatisfied” to “neither satisfied nor dissatisfied”
to “very satisfied.” Items of this type on a survey do not allow one to
determine customer delight.)

dissatisfier Any feature or attribute of a product or service that is

not associated with a quality characteristic (satisfier), and if present,
represents a deficiency, potentially resulting in scrap, rework, or cus-
tomer complaints. Examples include cracks or voids in product sur-
faces, invoice errors, and “bugs” in computer programs.

distribution In statistical applications, the pattern that randomly

collected data displays when depicted graphically. Types of distribu-
tions include normal, binomial, skewed, and Poisson. Variations in
the processes involved generate differences in the distribution of
sample values that result in these different pattern types.
In manufacturing, this term refers to the methods and activities
used to move products from the producer into the marketplace.

distribution channels The primary methods of distribution activity

used to move products from manufacturers to end users. These in-
clude motor carriers, air carriers, water transportation, and rail.
58 distribution requirements planning (DRP)

distribution requirements planning (DRP) The system used to man-

age warehousing and distribution activity, usually considered a mod-
ule of manufacturing resource planning systems. It maintains ware-
house and distribution center inventory, monitors transportation
and shipping activity, and sometimes includes electronic communica-
tion between the factory, distribution centers, and commercial car-
riers.

distribution types See binomial distribution, comb distribution, ex-

ponential distribution, and frequency distribution.

distributive numerical control (DNC) Machine control managed by

one or more computers, typically at the machine site itself.

diversification phase See networking phase (phase V).

DNC See distributive numerical control.

document exchange See compound document exchange.

DOE/DOX See design of experiments.

double Pareto diagram A special “back-to-back” Pareto chart (one

of the seven basic tools of quality) used to depict and contrast two
sets of data, or before-and-after data pertaining to a process or sys-
tem. See Figure 18.

double-peaked distribution A graph, typically a bar graph, that de-

picts a distinct “valley” in a range of data, with “peaks” on either
side. Most often, this distribution indicates two processes are at work
in the population shown and/or two different sets of data have been
tabulated together. Also commonly called a bimodal distribution. See
Figure 19.

drivers, business See business drivers.

drivers, cost See cost drivers.

DRP See distribution requirements planning.

duty cycle 59

Figure 18. Double Pareto diagram.

Number of Expense Account Errors

After Training Before Training

10
I5-c
I m
5

Figure 19. Bimodal/double-peaked distribution.

DSS See decision support systems.

duty cycle (Run time) divided by (run time plus setup time for the
operation).
economic order 4uantity ( E W See batch manubcturing.

ECR See error cause removal.

EDZ See electronic data interchange.

education and training fbr total quality) The required knowledge for
total quality implementation and management. (See also training.) The
education and training requirements for TQM operations vary widely de-
pending on the type of environment (manubcturing vs. service, no prior
TQM exposure vs. mature quality environment, etc.).

effect size The minimum difference or relationship that is desired for

detection at selected Type I (alpha) and Type I1 (beta)error levels. For
continuous data,this minimum value must consider the inherent variabil-
ity of the population(s)involved in the study.

eficiency Minimum cost to produce the desired product or service. Effi-

ciency is typically calculated by comparing the quantity of resources (labor,
materials, equipment time, floor space, etc.) planned for consumption in
producing the product or service to those actually consumed. A one-to-
one comparison would be considered perfect (or 10W0)efficiency. An-
other way to calculate efficiency would be to divide the run speed of the
equipment by the design speed of the equipment.

elasticity See demand elasticity.

electronic data interchange (EDI) The exchange of information by elec-

tronic means. The combination of computer and communications tech-
nology is yielding extremely efficient, effective, and relatively low-cost elec-
tronic data exchange methods. These developments support the

60
employee selection and training 61

movement of TQM companies toward sharing a great deal of their operat-

ing data with suppliers and customers. (See the data sharing discussion
in the advanced procurement technology listing.) In nearly every major
organizational process, electronic data interchange is becoming a central
tool for daily operations. In the future, as individual departments are
linked to other departments through local area networks (LAN), and
those networks are linked to wide area networks (WANs), and the WANs
are linked in turn to national and international networks such as Internet,
ED1 will become absolutely necessary to remain competitive. (See the illus-
trations shown in the architecture, information systems listing.) Among
the most popular hardware and sohare used for ED1 in our current
environmentsare computers, modems, h machines, and telephone lines.
However, we are also moving into more common use of microwaves, satel-
lites, infrared light, and a host of other technologies as the capabilities of
new materials and sohare become evident.

empluyee involvement A formal, ongoing process (internal to an organiza-

tion) in which employees regularly and substantively participate in deci-
sion making pertaining to the operations within their span of control. This
involvement may take place through participation on company, plant, or
local TQM steering committees; on commissioned teams (e.g., quality im-
provement or problem-solving teams); on sanctioned teams or natural
work groups; or through an employee suggestion system.
In order for employee involvement initiatives to be successful, man-
agement must respect the ideas and efforts of the work force. Manage-
ment must also support the concept of employee involvement by provid-
ing a formal structure for the process and deploying resources as required
in a responsible manner to incorporate the suggested improvements.
Management personnel must also be role models of employee involve-
ment themselves.

empkyee selection and training The choosing and preparation of workers

to undertake a TQM initiative, which is critical to achieving early success.
Employees selected to lead the way in any TQM initiative must be able to
deal with the stress that change brings, including the criticism of their
ideas, skepticism and cynicism from their peers (and sometimes from
management), and the need to compromise in order to achieve team ob-
jectives. Beyond thispersonality-related element, some fundamental skills
will be required of these employees. Team building, basic TQM training,
and specific targeted training as described under education and training
62 employee suggestion system

(for total quality) are good prerequisites to initiating improvement activi-

ties using any employee.

employe suggestion system See suggestion systems.

empowerment Giving people the power to improvdmanage a system or

correct an error. However, this brief description can be misleading, and
when it is, very destructive. Empowerment means the endowment of
power, but the power must be clearly defined, focused, and limited. When
a team or employees are empowered, they should be chartered to achieve
specific goals and objectives, informed about the amount and types of
resources available to them, and told about the time h e involved in
their empowerment and the limitations of that power. (See chartering a
team.)In thisway, no one develops unrealistic expectations or feels unnec-
essarily constrained in achieving the purpose of the empowerment; at the
same time, no one should exceed the limitations of his or her formal
power. A good rule to remember for this aspect of TQM is the phrase
“no unrealistic expectations, no unnecessary constraints, and no excuses.”
Empowered employees are provided with the knowledge, tools, responsi-
bility, accountability,and authority within their designated span of control
to be effective decision makers and take appropriate actions within a desig-
nated span of control without prior approval. For example, an employee
can be empowered to stop production if he or she detects a problem
without an immediate resolution.

end-of-linecharacteristics F’roduct variables that have been defined by

the customer as:

Quality characteristics or satisfiers that are critical to pro-

cesses or production systems, and
0 Measurable at the end of the supplier’s production process,
immediately before shipment or warehousing for future ship-
ment

See also critical process characteristics and critical product

characteristics.

engineering, concurrent See concurrent engineering.

environment and variation A look at the environment’s influence on

variation. The environment is a factor contributing to the variation
European Quality Award, The (TEQA) 63

in virtually every process, although some processes are less sensitive

to environmental factors than others. For example, chemical pro-
cesses are often very sensitive to temperature and humidity whereas
photoelectric processes are more sensitive to light. Generally, when
the environment is closely controlled, the effects of these factors are
minimal and may be expected to generate only common causes of
variability, producing a normal distribution (bell-shaped curve)
pattern when process output values are plotted.

EOQ Economic order quantity. See batch manufacturing.

error Act or instance of deviation from unexpected procedures or
methods of work that results in defects, scrap, rework, waste, injur-
ies, or any non-value-adding natural consequence. See also value-
added analysis.

error, conscious See operator-controllableerror.

error, experimental See experimental error.

error, inadvertent See operator-controllableerror.

error, operator-controllable See operator-controllableerror.

error, technique See operator-controllableerror.

error cause removal (ECR) A process designed to provide a formal

mechanism enhancing employees’ ability to communicate problems
effectively to management. (See also communication.) In the ECR
process, employees are provided with simple, one-page forms that
are usually made available in wall-mounted boxes at various locations
throughout the workplace. An employee removes and completes the
form, then returns it to the box. Completed forms are picked up and
then acted upon by the appropriate management on a daily basis.

European Foundationfor Quality Management A foundation located

in the Netherlands that awards an “all-Europe” quality award known
as the European Quality Award.

European Quality Award, The (TEQA) An award administered by

the European Foundation for Quality Management, which is similar
64 evergreen system

in concept to the Malcolm Baldrige National Quality Award, al-

though it differs somewhat in point values by category. To receive
this award, a company is required to demonstrate that its commit-
ment to T Q M has contributed significantly to the satisfaction of all
company stakeholders: customers, employees, and other holders of
an interest in the company.

evergreen system A term coined by Mahoney and Thor in their book

The TQM Trilogy (New York: AMACOM, 1994) to describe a method-
ology for sustainable quality management systems implementation.
The people dimensions stressed in this system are leadership, human
resources development, and management. The three fundamental
elements of their program include:

1. Total involvement, stressing organizationwide participation

2. Continuous improvement, emphasizing the ongoing nature of
the program
3 . Customer orientation, focusing on the end user’s definition of
quality

excursion analysis A strategy used to investigate failures that can

identify a multitude of root causes. Excursion analysis should be con-
ducted by the appropriate problem solvers in the organization on all
critical failures that impact the organization or the customers. Excur-
sion analysis can lead to such root causes as:

0 Processes that are out of control (see out-of-control pro-

cesses)
Processes that are not capable
Equipment failures
Incoming material defects
Poor design quality
Environmental effects
Operator-controllable errors
0 Management-controllable errors
Management system failures

executive’s role in TQM A role of strategy development, progress

monitoring, and macro-level project management. Executive man-
agement constantly assesses the gaps between current and desired
expectations, customer 65

future states, and steers the T Q M efforts in directions best suited to

filling the largest gaps. Outputs from the executive level activities
include:

1. A business architecture. Includes identified business drivers,

business processes, process performance metrics, and a re-
source infrastructure.
2. A strategic plan. Includes all areas defined in the business
architecture, a future state vision of all major business pro-
cesses, and specific strategic targets associated with each area
and process.
3. A high-level improvement strategy. Includes specific targets,
team structures to address each improvement area, improve-
ment activity time lines, and a formal T Q M program manage-
ment structure.

expectations, customer The expectations of a customer that are ei-

ther satisfied or dissatisfied over the life cycle of the product or ser-
vice. Typically, a customer must believe that the product or service
will satisfy his or her need before buying it and satisfies the same need
while using it. Finally, the customer must be satisfied with service
responsiveness and effectiveness when the product fails. These ex-
pectations are actually groupings of more specific expectations in
each phase of product or service life, with some of the specific expec-
tations listed below:

1. Pre-sale customer expectations:

A. Clear specifications relating to the perceived need
B. Clear compliance with all applicable safety and usage reg-
ulations
C. Clearly defined performance and reliability characteris-
tics
2. Post-sale customer expectations:
A. On-schedule receipt of product or service
B. Full operability and inclusion of all components at time of
delivery
C. Freedom from defects
D. User-friendliness
E. Reliability and ease of maintenance and repair
66 experimental design

3. Customer expectations at the end of the product or service’s

useful life:
A. Economical repair or replacement
B. Ease and low cost of disposal
C. In some cases, upgradability to new technology

experimental design The process by which factors are selected for

study and deliberately varied in a controlled manner, in order to
observe the effects of these actions. The designation of a particular
plan for assigning subjects or experimental units to experimental
conditions, which may also include statistical analysis of the resulting
data. See also basic tools of experimental design and design of ex-
periments.

experimental error A measure of all uncontrolled and/or unex-

plained sources of variation that affects a particular score. Also called
unexplained variation.

expert systems See artificial intelligence.

exponential distribution A continuous random variable with the fol-

lowing form:

where p is the population mean or mean time between failures.

While the normal distribution reflects an area of 50% above and
below p, the exponentially distributed population has 36.8% above
and 63.2% below the average (p). Examples of characteristics de-
scribed by the exponential distribution include loading patterns for
selected structural members and failure times for complex equip-
ment. See Figure 20.

external critical characteristic approval and review system One com-

ponent of the supplier quality assurance system model that relates
to the ongoing maintenance of a critical characteristic profile, in-
cluding nominalharget and specification control and review systems
as associated with all purchased materials. See Figure 2 1.
external m'tical characteristic approval and review system 67

Figure 20. Exponential distribution.

P
Time

Figure 2 1 . Breakdown of elements associated with the external crit-

ical characteristic approval and review system.

Coordination
of Engineering
Ana sisfor
?
Initial arget and
Specification
Specification Approval
A proval and
h!odification
System
Critical
Characteristic Coordination
Approval and of Specification
Rewew System Modification and
Im act Analysis
Coordination kocedures
of Internal QFD
Table for Critical
Characteristics
68 external customer

external customer See customer, external.

external failure costs Components of the cost of quality that would

disappear if there were no defects. Unlike internal failure costs,
these defects are found after shipment to the customer. Examples
include complaint adjustment, returned material, warranty charges,
and allowances.

extrapolation planning The practice of setting goals based on ex-

trapolating or projecting current business performance improve-
ments into the future. This approach contrasts sharply with a bench-
marking process in that the extrapolation method is oblivious to best
practices or any comparison to outside organizations. It is, nonethe-
less, one of the most popular methods of goal setting in a pre-TQM
environment.
facilitation The leadership of a team through coaching. Facilitation is
designed to overcome the difficulties inherent in a team operating envi-
ronment. Among the challenges hcilitation should resolve are artificial
consensus, “group think,”intermember conflict, and the tendency to be-
come so enamored with the team process that the end objectives not only
aren’t reached, but are completely forgotten.

facilitator A team member or outside resource who possesses good

group process skills and guides the team through coaching. Facilitators
should have excellent interpersonal skills, be able to synthesize groups of
ideas into understandable concepts and effectively communicate them, be
prepared to deal effectively with strong and emotional team members
while remaining objective, and keep focused on the team’s processes for
achieving its mission. See also communication.

f a c i h t m qudijicatiotls The qualifications of a facilitator include:

1. The ability to help others reach conclusions and achieve con-

!3enSus.
2. The ability to withstand strong emotional pressures such as bully-
ing, shouting, and crying.
3. The ability to separate himself or herself from the team and its
objectives on an emotional level-that is, detachment and objec-
tivity.
4. The ability to remain focused on the team problem-solving pro-
cess and not get caught up in the problem itself. The hcilitator
must be able to lead the team through the problem-solving and
consensus-buildingprocess whether he or she believes in the solu-
tion the group develops or not.

f d h t m ’ s respons&il&s See hcilitator‘srole.

69
70 facilitator’s role

facilitator’s role The responsibilities of the facilitator:

1. Helps the team to c l a e the mission

2. Builds the foundation for effective team communiCation
3. Is the neutral guide of the team
4. Does not evaluate or contribute unqualified ideas (unless agreed
to in advance by the team)
5. Suggests alternative methods and procedures
6. Protects individuals and their ideas from attack
7. Encourages participation by all on the team
8. Helps the team to come up with win/win solutions
9. Helps to create an atmosphere of synergy within the team
10. Makes certain that outside influences are minimal
11. Works for consensus on those items requiring it
12. Focuses on the process of achieving the team’s mission
13. Works with the team leader to discuss meeting agendas and out-
comes

fd
u?udyi.s An analysis procedure in complex experimental de-
signs. In hctorial analysis, the influence of the experimental hctors is
evaluated in terms of what their main effkct or, in combination with other
hctors, what their interaction effects have on some criterion measure or
quality characteristic (satisfier).The designs in which hctorial analysis is
appropriate may be full hctorial designs, when each factor level of each
hctor appears in combination with each and every level of every other
factor in the design. Or, since this setup can lead to numerous combma-
tions in complex designs, a f i m W hctorial design may be employed.
In both cases, factorial analysis, typically employing analysis of variance
techniques, is proper and desirable.

fudure In reliability, the inability of the equipment to satisfj perform-

ance or meet design specifications after the equipment has experienced
successhl operation, or when there is an expectation that the equipment
will perform successfdly without adjustmentlrework. Failure is not con-
fined to mechanical breakage or electrical malfunction. Other forms of
Mure include the following:

1. Loss of speed
2. Performance degradation
3. Out-of-adjustment conditions
fault tree analysis (FTA) 71

4. Lackofquality
5. Leaks
6. Operational malfunction

See also reliability.

failure costs See internal failure costs and external failure costs.

failure mode analysis (FMA) An analysis procedure used for the

identification of potential ways in which a system, process, or piece
of equipment can fail. Once the modes of failure are identified, pre-
ventive plans, actions, and options may be developed and imple-
mented to lessen the likelihood that the potential failure modes will
become realities. FMA is a component of reliability and advanced
quality planning.

failure mode effects analysis (FMEA) An extension of failure mode

analysis in which effects of failure are identified (anticipated) as is
the mode or method by which failure might occur. Once each failure
mode and the effects of such a failure are identified, they are rated
according to severity, likelihood of occurrence, and likelihood of
being detected. They are then prioritized for preventive action and
control.

failure mode effects and criticality analysis (FMECA) A technique

performed following a failure mode effects analysis in which each
potential failure effect is classified according to its severity and prob-
ability of occurrence.

fault tree analysis (FTA) An analysis technique designed to help

avoid problems or eliminate historical problems through identifica-
tion and resolution of root causes. The fundamental steps involved
in a fault tree analysis are:

1. Identify and clearly define the fault to be addressed.

2. Identify all potential causes of the fault.
3. Define underlying causes for each of the potential causes
identified in step 2. Repeat the process at lower and lower
levels until root causes are identified.
72 feedback, customer

Standard symbols are used in conducting an FTA to identify the

various events and logical associations. See Figure 22.

feedback, customer Information gathered from customers as a re-

sult of listening to their responses (both positive and negative, both
direct and indirect) about the product or service. This feedback oc-
curs in all three stages of the productlservice to customer relation-
ship: presale, after sale, and after the product or service’s useful life
is over. The objective of constantly gathering customer feedback is
to become and remain consistently aware of customer needs and how
well they are being met. Formal customer feedback systems take
many forms, from simple customer reply cards packaged with the
product to full-blown surveys and complex computer systems. Note
that customer feedback should be two-way, including feedback to the
customer of the work the company is doing on critical customer satis-
fiers. See also survey-feedback process.

Feigenbaum, Armand A contemporary quality leader who is cred-

ited with having invented total quality control. Feigenbaum empha-
sizes quality as a business strategy and a primary source of competi-
tive strength.

Figure 22. Fault tree symbols.

Top Undesired Event: Primary or Basic Secondary Fault Event:

gate event or Fault Event: inconsequential event
command event: does not require or insufficient data
requires further development further development to develop

n
T
Primary or Basic And Gate: Or Gate (inclusive):
Fault Event ail inputs are required if any input events occur,
dependent on lower events; to produce output event; the output event will occur;
develop separate fault tree parallel redundancy series reliability

a
Normal Event:
event that must occur
Transfer Event:
information in or out
fixible manufacturing 73

fishbone diagram or model See cause-and-effect diagram and dis-

persion analysis.

fitnessfor use The degree to which a product or service satisfies the

purposes of the customer. It is judged by the customer, not by the
manufacturer or service provider, and is universally applicable to all
goods and services. Parameters for fitness for use include design
quality, conformance quality, and reliability. Fitness for use is the
objective of a three-part system developed by Juran, comprised of:

1. Breakthrough projects. Includes the creation of a climate for

change, identification of the vital few projects, analysis, over-
coming resistance to change, instituting change, and instituting
change controls. (See also resistance, planning for.)
2. Control sequence. Includes optimizing the cost of quality, es-
tablishing the needed professional staff, establishing feedback
loops, and establishing a repeated action cycle.
3. Annual quality program. Includes establishing quality objec-
tives and maintaining the initiative and momentum.

five S’s Five Japanese terms perceived by many to represent the

fundamental elements of a total quality management approach:

1. Seim’ (organization)
2. Seiton (neatness)
3. Seiso (cleaning)
4. Seiketsu (standardization)
5. Shitsuke (discipline)

Jex-time An approach to employment that includes allowing em-

ployees to vary their starting and quitting time each day to suit the
needs of their job (fluctuating work load) and their personal needs.
Usually, there are parameters involved in this type of arrangement
such as “Employees may begin their workday any time between 5 A.M.
and 9 A.M., and may leave any time between 1 P.M. and 6 P.M.”

JEexible manufacturing A manufacturing environment typified by

factory floor layouts organized by product family. The purpose of
flexible manufacturing is to optimize the flow of multiple products
through any of a number of series of the work centers involved. The
74 jlow chart

flexibility achieved by layouts and work load balancing resulting from

this approach often produces shorter manufacturing cycle times,
lower inventory levels, and shorter setup times than more traditional
approaches where equipment is grouped by type (e.g., all the lathes
in one place, all the presses in another place). Machine shops are the
most common setting for the deployment of flexible manufacturing
today. In machine shops, the parts to be run are assigned a specified
amount of capacity on each machine. The machine’s available capac-
ity and capability information is stored in a computer database, as is
the part routing information, and the machining equipment itself is
most often numerically controlled. As parts move through the pro-
cess, the computer continually adjusts priorities to ensure adequate
capacity on capable machines is available as needed.

flow chart A chart representing the steps of a process. These charts

typically use a standard set of symbols to depict different process
steps, including subprocesses, decisions, information storage points,
and termination points (see Figure 23). The flow chart is one of the
seven basic tools of quality. See also block diagram.

FMA See failure mode analysis.

FMEA See failure mode effects analysis.

FMECA See failure mode effects and criticality analysis (FMECA).

focal point chart A fundamental tool developed for the facilitation

of policy deployment. The focal point chart provides, on a single
document, a listing of measurable and timelined strategic objectives
selected as a result of a strategic analysis and planning process. The
milestones, or focal points, are associated with product performance,
quality, defects, delivery, service, and cost factors important to the
customer that are in need of a breakthrough rather than tactical
(kaizen)improvement. (See also breakthrough improvements.)They
are arranged on this focal point chart according to four major catego-
ries:

1. E x t e m l customer satisfiers. Requirements or quality character-

istics.
2. E x t e m l customer dissatisjers. Defects.
foolproojing 75

Figure 23. Flow chart symbols.

8 START

Procedure
or Step

Procedure
or Step

3. Internal customer needs. Internally driven business require-

ments.
4. Projects.

The focal point chart serves as the end of the policy deployment
process at each level of the organization. It also serves as the subse-
quent initiation of the daily management function. See Figure 24.
See also customer needs, satisfier, and dissatisfier.

focus factory approach See process-oriented flow.

fooZproo$ng Fail-safing; a process derived from the Japanese ap-

proach called poka-yoke. It was initially widely used in machine set-
ups to ensure accuracy and vastly reduce the time required to do
the setups. Later, the concept migrated to assembly design, and was
76 force field analysis

Figure 24. A focal point chart for the organization of deployed criti-
cal characteristics.
External Internal
Customer Needs Customer Needs
Priority
Quali Defects 8, Safety, Quality, Projects
Charactenstics Defectives Operational Costs,
(Sabsfiers) (Dissatisfiers) Housekeeping, etc.
Details
1. Flatness Scratches Accident Reduction New Press Start-up
2. Hardness Edge Bums Recovery New Alloy Develop.
3. Tensile Strength Voids Eliminating Cobbles
4. Surface Quality Smears
5. Profile
6.

Quality Problem Any/All Any/All

How Improvement SOIW Strategy Appropriate Appropriate
(StrategY) Strategy (QW fss) Strategy(ies) Strategy(ies)
Who Assigned Cross-
Assigned Quality Problem Natural Work Teams; Funcbonal Project
improvement sOlviMJ Employee Suggestions TMmS
bersonnel) Team Team (QWPDCNSDCA)
spc/sQc Took Deming meel; Any/All Any/All
what
(Major Tools) yf',","'7 Basic Tools Ap opriate
Sols
Appropriate
Tools

applied in the manner that parts can only be assembled in the in-
tended sequence, with the intended orientation.

forcefield analysis A technique used by T Q M improvement teams

to identify the forces at work in an environment. It graphically repre-
sents the forces that will support and/or enable the group to reach
its desired objectives, as well as the forces that impede or restrain
progress.
Conducting and using the force field analysis includes the fol-
lowing steps:

1 . List the enabling forces and prioritize them in descending

order of importance.
14 points 77

2. List the inhibiting or restraining forces on the opposite side

of the page in the same manner.
3. Take whatever action is required to eliminate or mitigate the
inhibiting forces. (Utilize the plan-do-check-actapproach.)

fostering innovation See innovation, fostering.

four M’s Terms popularized in conjunction with the Ishikawa fish-

bone diagram, they represent manpower, methods, materials, and ma-
chines. These terms are used to define broad categories of possible
problem causes for problem solving through identification and elim-
ination of root causes. See also cause-and-effectdiagram.

14 points Dr. W. Edwards Deming’s 14 management obligations

that are intended to help companies increase their quality and pro-
ductivity:

1. Create constancy of purpose toward improvement of prod-

uct and service, with the aim to improve competitive posi-
tion and to stay in business and provide jobs.
2. Adopt the new philosophy.
3. Cease dependence on inspection to achieve quality.
4. End the practice of awarding business on price alone. In-
stead, depend on meaningful measures of quality along with
price and move toward a single supplier for any one item,
on a long-term relationship of loyalty and trust.
5. Improve constantly and forever the system of production
and service, to improve quality and productivity, and thus
continually decrease cost.
6. Institute training on the job.
7. Adopt and institute leadership.
8. Drive out fear, so that everyone may work effectively for the
company.
9. Break down barriers between staff areas and departments.
10. Eliminate slogans, exhortations, and targets for the work
force; don’t ask for new levels without providing methods to
achieve them.
11. Eliminate numerical quotas for the work force and numeri-
cal goals for management.
12. Remove barriers that rob people of their right to pride of
78 frequency distribution

workmanship and eliminate the annual rating or merit

system.
13. Institute a vigorous program of education and self-improve-
ment for everyone.
14. Put everybody in the company to work to accomplish the
transformation; create a system to push every day on the
above 13 points.

frequency distribution A graph that represents the frequency of oc-

currences for possible outcomes for a data set. Frequency distribu-
tions may be ungrouped (each class interval is a single observation)
or grouped (outcomes are grouped into class intervals of more than
one observation). See also histogram.

frequency polygon A line graph that depicts a frequency distribu-

tion.

FTA See fault tree analysis.

functional layout A factory floor layout that groups similar equip-
ment into distinct and separate areas (i.e., all machining equipment
in one area, all sheet metal equipment in another area, etc.) This
contrasts with the process-oriented layout, which is organized into
cells that perform all the production operations for families of parts.
Functional layouts typically require more extensive material han-
dling and travel distances and result in more work-in-process inven-
tory than process-oriented layouts. See also process-oriented flow,
just-in-time manufacturing, and parts flow.

functional responsibilities Those duties that are performed as a nec-

essary and value-added part of the daily routine for a specific job
function.

funnel experiment An experiment that illustrates the results of tam-

pering or tweaking. Marbles are placed in a funnel held above a flat-
surfaced target. The person conducting the experiment will tend to
overcompensate both for missed targets and direct hits, demonstra-
ting that an adjustment made to a stable process will often produce
worse products than those generated from the process when it is left
alone (here, when the funnel is held in one place). See also tweaking.
FURSAP measures 79

FURSAP measures A generic set of product attribute categories pro-

duced by A. Richard Shores in 1988, which may be used in conjunc-
tion with quality function deployment to eliminate the confusion
common in generating attribute information for QFD house of qual-
ity activities.
The categories comprising the FURSAP acronym are:

1. Functionability
2. Usability
3. Reliability and Performance
4. Supportability/Serviceability
5. Availability
6. Price
 G
GAAP See generally accepted accounting principles (GAAP).

guimharing An incentive plan that distributes bonuses on a regular basis

to all employees based on the performance of the company. In many
companies, the distribution differs in size between levels, but must remain
proportionate so that greater company performance equals proportion-
ately greater yield to every employee. In other companies, the distribution
is identical for all employees and varies in size only as a direct result of
differences in total company performance.

Gantt chart A chart that uses horizontal bars to depict planned and com-
pleted work activities on a scale of time. See Figure 25.

gap andysis A technique for identdjmg needed improvements by com-

paring the current environment and current processes against an envi-
sioned hture state. After i d e n w g the differences between the two, the

Figure 25. Gantt chart.

0 5 10
-=1
15 20 25
Time
80
30 35 40
golno go 81

magnitude of the differences between them is analyzed and becomes the

basis for defining improvement strategies.

gauge and rep.oducibilily (studies) Refers to the analysis of

the precision of a measurement process through the assessment of its:

1. Rqbeatabzl22y (urn).The degree to which repeated measures of the

same object or specimen yield the same result.
2. JZqbmdWCy (uRPD). The degree to which the means of two or
more measurement error distributions agree, or codormance to
the same value.

Note that measurement error (uE)is a combination of variability due

to repeatability and reproducibility.
Some repeatability and reproducibility studies are called R&R
studies and are generally conducted at one point in time. These stud-
ies may also be conducted over the short and long term; however, in
these cases, the assessment of statistical control is required. All preci-
sion analyses assessing repeatability and reproducibility are used with
continuous data. Measurement processes associated with discrete
data for analogous applications are assessed for their reliability, based
on internal consistency and concordance.

generally accepted accounting principles (GAAP) Formal written

principles that govern the behavior of certified public accountants,
who prepare financial statements, provide tax services, and offer ac-
counting advice and legal opinions about the accuracy of a com-
pany’s financial statements.

generic routings Manufacturing process routings that pertain to a

group or family of parts. These routings often depict the physical
movement of part families through cellular manufacturing opera-
tions.

George M . Low Award See NASA Award.

go/no go A term used to refer to the acceptability of a unit or prod-

uct. This term reduces all possible conditions to a binary decision
about product acceptability: go means acceptable (conforms to speci-
fication), and no go means unacceptable.
82 goals

goals Defined conditions to be achieved by specific effort applied

over the long term. In common terms, a goal is what one sets out to
achieve.

goals, strategic Defined conditions that flow from the vision of the
organization and that, if achieved, would necessarily result in the at-
tainment of the vision. Strategic goals typically span more than five
years in their achievement, and are each further defined as contribu-
tory strategic objectives.

government/industry partnership A movement away from the ad-

versarial “police officer” role for government versus the “manipula-
tor” role for business, toward a relationship of collaboration toward
common objectives. This relationship will mean an increased sense
of ownership for both groups. Government will need to become a
better coordinator and facilitator of business relationships and infra-
structure, and business will need to become a better customer for
government support through defining its needs in a way that is bene-
ficial to society and useful for government infrastructure designers.
In this way, legislation and funding may be directed into channels
that promote growth and stability for business and the national com-
munity with minimal government intervention.

group process techniques A general set of group behavior process

guidelines including problem solving, consensus building, brain-
storming, and nominal group technique.

group technology An approach to design and/or manufacturing that

uses the common characteristics of parts to gain efficiencies in the
utilization of manufacturing resources (such as machine setups) and
procurement leverage with suppliers by offering the ability to negoti-
ate prices for entire commodities (families)of parts.

growth rate analysis A technique for identifying the rate of growth

in a specific technology over time. A minimum of three points are
used to identify differences along the growth line of the technology,
generally based on the published market size of that technology.
Hawthonte effect A term referring to experiments conducted by Bell
Laboratories in the 1920s, which demonstrated that mere management
interest in employees and their working conditions a f F i i the productivity
of the employees.

hiddafactoly That portion of fictory capacity that is not used due to

waste or poor asset utilization in the areas of availability, duty cycle,
eificiency, and recovery.

hierarcity of imprruement A diagram that helps illustrate the steps re-

quired for continuous improvement in business. To move up the hierar-
chy, an organization must develop an infrastructure to support the various
levels of improvement. These levels of improvement can be applied to
quality, cost, delivery, safety, environment, and morale. These in turn will
lead to improved competitive position and customer satisfiction, and, as
in a building, if the lower levels are not solid, the entire structure will be
weak. See Figure 26.

Figure 26. Hierarchy of improvement.

Standardization
Maintenance
Housekeeping
Employee-Management Trust and Respect
Management Self-Discipline
83
84 histogram

histogrum A vertical bar graph summarizing the variation in a data set.

The histogram’s pictorial nature makes it easier for people to identlfy pat-
terns that are more difficult to discern in a numeric table. See Figure 27.
The histogram is one of the seven basic tools of quality. See also fre-
quency distribution.

hoshin planning A planning technique that derives its name from the
Japanese term hoshin kanri, which means “shining metal” and “pointing
direction.” This planning method uses relatively short-term, high-impact
goals to achieve breakthrough improvements in company performance.
Companies use vision statements to develop their goals, objectives, and
operating strategies. Periodic audits are performed to assess and encour-
age progress toward achieving the envisioned hture state. See also policy
deployment.

house of@@ A graphic tool used in quality function deployment that

displays and organizes the “whats” of defined cusfomer needs and the
“hows” of the manufacturing process, generating a clear picture of the
relationships between what is done in each process of the manufiicturing
and how these individual activities relate to the defined customer needs.
See Figure 28. See also matrix diagram.

Figure 27. Histogram.

18L
16 -
low = 10
high = 40
2 14- range = 30
mean = 21.5
5 12-
g 10-
E
LL
8-
6-
4
2-

FIow Rates
hypothesis 85

Figure 28. House of quality for making instant coffee with a micro-
wave.

Strong Relationship
0 Medium Relationship
A LOW Relationship

\
Process
Variables

Product
Characteristics

Temperature 01.1 I
Flavor 0
hypothesis An unproven theory, supposition, or assertion that is used to
explain certain facts or relationships between variables. Hypotheses are
generally used to provide a foundation for hrther investigation about
possible cause-andeffect relationships.
 I
impe$echn S e e nonconformity.

itnplementatum
’ phase (phase N) The fourth phase of the T Q M process.
(Seethe Introduction of this publication for an overview of all five phases.)
This phase, also called the deploymentphase by some TQM practitioners, is
the portion of a TQM implementation that contains initial pilot activities
and long-term management commitment to the overall TQM process.
The implementation phase of a T Q M program typically involves the fol-
lowing steps:

1. Alignment with the strategic improvement plan. In this step, the

plan-do-check-act cycle is utilized to establish initial pilot programs
focused on the vital few opportunities and to assure that top manage-
ment commitment is supplied in a manner that will provide adequate
funds, personnel, and other resources to enable success. See also stra-
tegic plan.
2. Strengthening the program infrastructure. In this step, training
and team-building activities are used to support pilot initiatives and
to expand the general basis of understanding of T Q M principles and
concepts.
3. Initiating the continuous improvement plan. In this step, the
organization’s vision is fully aligned with the planned and ongoing
T Q M improvement initiatives, such that the organization’s future is
fully a product of the TQM continuous improvement processes.
T Q M process improvement teams become a way of life in the organi-
zation during this step and eventually become integral to the culture
of the company. (See also company culture.) All improvement team
activities are directed toward measurable objectives that are aligned
with organizational objectives and strategic goals and flow from the
company vision and mission. (See also goals, strategic.)

86
in-process testing 87

4. Monitoring pe?jformanceand communicating lessons learned. In this

step, team progress is monitored by executive management, and
progress is measured against objective, time-phased milestones.
Based on successes and Edilures, three things happen: First, successes
are celebrated widely throughout the organization. Second, failures
are traced to root cause so that corrective actions can be docu-
mented and these mistakes can be prevented in the next improve-
ment effort. Third, the strategic improvement plan is adjusted and
modified as required to incorporate the lessons learned from ongo-
ing initiatives.

importance See comparative and relative importance.

important many Juran’s redefining of the trivial many, a term used

in Pareto analysis. See also trivial many.

improvement, breakthrough See breakthrough improvement.

improvement, continuous See continuous improvement.

improvement plan See continuous improvement plan.

in-control process A process that, when measured statistically, con-

sistently produces only variations that can be attributed to common
causes. The presence of this situation is generally judged through
the use of a statistical process control chart. See also control chart,
out-of-control process, lower control limit, and upper control limit.

in-process inspection Inspection of parts within the manufacturing

process, often performed by the machine operator, using manufac-
turing tooling as a medium of inspection. This approach contrasts
with the off-line inspection that detects defects only at widely spaced
intervals between manufacturing processes (usually by an autono-
mous quality inspector), allowing more time and inventory to be ex-
pended before the condition is detected and remedied. See also de-
tection.

in-process testing Testing of manufactured materials, parts, and as-

semblies during the manufacturing process, often by the machine
operator. This approach contrasts with the off-line testing tradition-
88 inadvertent error

ally performed at the conclusion of a series of manufacturing opera-

tions, where detection of the root cause is far more difficult and time
consuming.

inadvertent error See operator-controllableerror.

incursion probability, new technology A calculation used to estimate

the likelihood of new technologies to migrate through any specific
area of the manufacturing company. (Refer to Manufacturing 2000,
W. Duncan, New York: AMACOM, 1994.)

independent variable Any factor that may have an effect on the de-
pendent variable under investigation. Under certain circumstances,
the independent variable may also be called a covariate.

individuals chart See X chart.

inference space A term used to refer to the population or universe

to which the results of an experiment may be generalized. See also
research population.

injluences, competitor See competitor influences.

information reliability See reliability, information.

information systems architecture See architecture, information sys-

tems.

infrastructure management The identification and manipulation of

infrastructure elements. Infrastructure in the business context in-
cludes legislative, political, economic, physical, and social founda-
tions for competitive business operations.

innovation Any significant “quantum leap” or breakthrough

change (especially an improvement) in a method or device. Innova-
tion generally arises from three sources:

1. Demand-Zed innovation, which results from the desire to over-

come labor or material constraints in meeting production de-
mands
intelligence, artiJicia1 89

2. Competitive-led innovation, which results from market demands

for new products and services and the direct pressure of com-
petitors, especially as related to competitive benchmarking
(see also competitor influences)
3. Internally generated innovation, which results from internal re-
search and development, and increasingly from TQM-related
improvement activities

innovation, fostering Fostering innovation in a T Q M environment

is generally a result of using team-based problem solving. Also, the
desire to innovate in order to surpass competitors as experienced in
the competitive benchmarking process. (See also competitor influ-
ences.) Especially in manufacturing organizations, cross-functional
teams have proven very successful in fostering innovations in prod-
ucts as process design. And finally, innovations occur most fre-
quently in environments that are nonthreatening, since risk-taking is
a part of the nature of innovative activity.

inspecting in quality us. designing in quality See detection vs. pre-

vention.

inspection The measurement of products to compare them against

a predefined standard of acceptability. Inspection is generally tar-
geted against a specific set of part characteristics identified as critical
to product performance or marketability.

inspection, point-of-use See point-of-use inspection.

instant pudding A phrase used to describe the incorrect assumption

that quality and productivity improvement can be achieved quickly,
on faith. The “instant pudding” mindset can be a true obstacle to
achieving quality as the process requires ample education and effort.

integrated information systems A set of systems so completely inte-

grated that telecommunications, information processing, electronic
data interchange, and many user interface device controls will be
virtually indistinguishable from one another.

intelligence, artijcial See artificial intelligence.

90 internal customer

internal customer See customer, internal.

internal failure costs Components of the cost of quality that would

disappear if no defects existed in the products prior to shipment to
the customer and if all specifications for the product were correct.
Examples include scrap, rework, retest, downtime, yield losses, and
disposition.

International Standards Organization (ISO) A worldwide federation

of national standards organizations from more than 100 countries.
This organization exists to promote international standardization
and related activities. Headquartered in Geneva, Switzerland, IS0
initiated the IS0 9000 series of quality standards that have become
central to modern T Q M activities in a large number of American
(and other) businesses.

internal data Classification according to a continuum, such as

ounces or inches. Interval data are characterized by an equality of
units and the ability to sensibly divide the units of measure into sub-
units (e.g., hours into minutes). See also continuous data and vari-
ables data.

inventory, buffer See buffer inventory/stock.

involvement See employee involvement.

Ishikawa, Kaoru A quality expert, author, and consultant who has

served as president of Kei-dan-ren Federated Economic Societies and
the Union of Japanese Scientists and Engineers (JUSE). He devel-
oped the Ishikawa diagram (the cause-and-effectdiagram) and is a
recipient of the Deming Prize. He also played a key role in introduc-
ing Deming and Deming’s theories to top management in Japan.

Ishikawa diagram See cause-and-effect diagram.

I S 0 Acronym for International Standards Organization.

I S 0 9000 A set of five international standards pertaining to quality

assurance and quality management developed in 1987 and pub-
lished by the International Standards Organization. The standards
isolated peak distribution 91

were developed to harmonize many national and international stan-

dards in order to facilitate international and intranational contract
establishment and commerce. Because of its breadth, IS0 9000 is
necessarily fairly generic and has been criticized in terms of its useful-
ness. The individual IS0 9000 standards are as follows:

1. I S 0 9000. Quality Management and Quality Assurance Stan-

dards: Guidelines for Selection and Use.
2. I S 0 9001. Quality Systems-Model for Quality Assurance in
DesigdDevelopment, Production, Installation, and Ser-
vicing.
3. I S 0 9002. Quality Systems-Model for Quality Assurance in
Production and Installation.
4. I S 0 9003. Quality Systems-Model for Quality Assurance in
Final Inspection and Test.
5. I S 0 9004.Quality Management and Quality System Elements
Guidelines.

IS0 9000 registration shows that the company has been audited and
approved according to the applicable standard(s) by an independent
IS0 9000 registrar. See also certification, I S 0 9000.

isolated peak distribution In a distribution pattern plotted with ver-

tical bars, a small peaked distribution of values that is separate from
the larger distribution pattern. This phenomenon usually results
from the operation of two or more distinct processes in the data
sampled. See Figure 29.

Figure 29. Isolated peak distribution.

JIT Seejust-in-time manufbcturing.

Jumn, Joseph M . One of the first quality leaders to deal with the broad
management aspects of quality. Juran first introduced the cost of quality
concept in his book The rzualzcy CmhoZ Handbook (New York: McGraw-Hill)
in 1951.Juran advocates structured improvement initiatives, the sense of
urgency, extensive training, and strong upper-level management commit-
ment.

JUSE Union of Japanese Scientists and Engineers.

just-in-time (Irrr) manufacturing A philosophy of waste elimination. In a

just-in-time environment, the underlying operating principle is that any-
thing that does not add value is eliminated or minimized to the greatest
possible extent. In manuhcturing, all inventory, queue time, and labor
that are not required are targeted for reduction. See also valueadded
analysis.
There are several aspects to just-in-time manuhcturing, including:

1. Process-oriented Jlow. The objective of process-oriented flow

is to convert functional layouts of machines in the factory into series
of processes. Layouts are based on the production of product fami-
lies, or commodities. (See commodity analysis.) Process-oriented
flows are superior to traditional functional layouts since they reduce
travel distance, required floor space, and total throughput times.
They also lend support to conversion of operations into cellular
manufacturing.
2. Advanced procurement technology. The objective of advanced
procurement technology is to make suppliers an extension of a com-
pany’s own manufacturing operations. Suppliers directly affect prod-
uct costs, product quality, responsiveness to customer needs, and

92
just-in-time (IIT) manufacturing 93

thereby profitability and viability as a business. In order to maximize

the connectivity between a company’s operations and those of its sup-
pliers, advanced procurement technology directs that companies
move from previously adversarial relationships toward supplier part-
nerships where contracts are tied to many more aspects of perform-
ance than merely price. Quality, delivery performance, and other
measures should be included. In addition, supplier capacity may be
the object of the contract, rather than merely a specific number of
designated parts delivered on a designated date. This cooperative
contracting provides for a widwin situation where supplier work sta-
bility is achieved with maximum production and delivery flexibility
for operations. Electronic data interchange with suppliers is another
important aspect of advanced procurement technology, as are sup-
plier certification programs and involving the supplier in up-front
concurrent engineering processes.
3. Improved design methods. There are two objectives of improved
design methods, with one objective related to the design process and
the other related to the product designed. Just-in-time techniques
may be applied to the process of design engineering to minimize
design work-in-process and design throughput times. Just-in-time
principles should also be applied to the actual design product
through applications of concurrent engineering and value analysis/
value engineering to make the product as efficient and easy to manu-
facture as possible.
4. Enhanced support functions. The objective of enhanced support
functions is to use JIT techniques, such as the elimination of non-
value-added activity and time to eliminate waste in all support areas.
(See value-added analysis.)
5. Employee involvement. The objective of employee involvement
in a JIT environment is to use the skills and knowledge of all employ-
ees effectively in improvement efforts, and to enhance “buy-in” to
improvement initiatives. There are several approaches to employee
involvement, such as quality circles and employee suggestion pro-
grams. See suggestion systems.
6. Enhanced quality. The enhancement of overall product and
process quality levels is an essential element of JIT manufacturing.
In a JIT environment, a central objective is to eliminate defects since
defects are a major cause and result of waste in many resource areas.
JIT efforts, therefore, often include initiatives to ensure that designs
94 just-in-time (IIT) manufacturing

will result in quality products, that process capability will result in

quality products, and so forth. All JIT implementations include a
central emphasis on quality, and most T Q M programs adopted in
manufacturing operations include one or more initiatives that focus
on JIT operations.
7 . Simplajied, synchronousproduction. The objective of simplified,
synchronous production is to convert individual manufacturing ac-
tivities into elements of a continuous manufacturing flow synchro-
nized to end-product demand. Lot sizes are reduced in this process
and materials are “pulled” from operation to operation in manufac-
turing. (See also pull system.)
hken A Japanese term for continuous improvement, founded on the
principles of doing little things better and setting, working toward, and
achieving increasingly higher standards. Masaaki Imai, author of the book
Kaizen: The K q to Japan’s Cmpetdiue Success (New York: Random House,
1986),states that quality deployment is regarded to be the most significant
development to come out of total quality control since the mid-1960s. He
also emphasizes policy deployment as a central aspect to making quality
pervade the entire organization. See also continuous improvement and
breakthrough improvements.

kunbun A Japanese term that is the cornerstone of the just-in-time pull

system. Kanban actually means “to put away and to bring out.” In pull
systems, it often refers to a card or other physical device used to signal the
previous operation that it is authorized to produce the next unit. See also
synchronous production.

kurtosis A shape parameter (y4) that indicates the flatness or “peaked-

ness’’ of a distribution. It is also called “the fourth moment about the
mean, or y4 (population) and g4 (sample).” Normal (Gaussian) distribu-
tions are referred to as Mesokurtic in nature; mso indicating an intermedi-
ate level of peakedness. Distributions that are higher in the peak and lower
in the tails than a Mesokurtic distribution are called Leptolzurtic. Distribu-
tions that are lower in the peak and higher in the tails than Mesokurtic
hnctions are called Phkurtic distributions. Underlying population distri-
butions that are Mesokurtic possess a theoretical y4 value of 0.0. Theoreti-
cal population functions that are Leptokurtic possess positive y4 values.
Platykurtic fbnctions reflect negative y4 indices. See Figure 30. See also
normal distribution and skewness.

95
96 kurtosis

Figure 30. The general forms of kurtosis.

Mesokuiic Curve
(intermediate)

Leptokurtic Curve

Platykurtic Curve
latent defcts Defects that are not identified until they are experienced
during customer ownership. Latent defects are generally caused by inade-
quate design margin, product complexity, and/or low process capability.
The solution to latent defects most often lies in increasing the ratio of
design margin to process capability.

LCL See lower control limit.

Z
ed teum, business See discussion under planning phase (phase II).

lead time The time ahead of a specified milestone at which the prior
activity must be started. In manuhcturing, there are several segments of
lead time. For example, material requirements planning systems usually
contain manuhcturing lead time to produce the product, supplier lead
times (to procure parts for the product), and move and queue times for
material handling between production operations. See also cycle time.

Ze&, team A member of a team with a vested interest in the team's

success and one who is willing to help the team stay focused.

k&s role The responsibilities of the leader:

1. Should be content focused

2. Is accountable for outcomes, assignments and record keeping,
and progress checks
3. Should work with the team hcilitator to discuss meeting outcomes
and meeting agenda design (who will do what, etc.)
4. Is a full-fledged team member
5. Can serve as the team's coach regarding content
6. Should be the primary contact with the team sponsor

97
98 leadership

leadership The ability to create a vision, clearly communicate that vision

to others, and then motivate those others to achieve the envisioned state.
Leadership is an absolutely critical element of any T Q M implementation.
See also communication and motivation.

leadershipprocesses The primary processes performed by top-level man-

agement leaders in a TQM environment. According to Svenson et al. in
The @d?y Roadmap (New York AMACOM, 1994),these processes include
strategic planning, operational planning, budget control, empowerment,
and results measurement.

lqbtokurtic See kurtosis.

Zistaing to the customer See customer mapping process.

little q See big Q little q.

LL Seelowerlimit.

logistics, &sign fw See design for logistics.

loss function, Taguchi See Taguchi loss function.

lot In a TQM context, a quantity of parts to be accumulated for purposes

of sampling. In the more common manufacturing context, the term lot is
used to refer to a group of parts or product (often physically in one con-
tainer) traveling together through the manuhcturing process.

lot sire The quantity of product or number of units (or parts) within a
lot. See lot.

Low (George M.)Award See NASA Award.

lower c m Z limit ( E L ) The lower boundary, drawn as a horizontal

dashed line on a control chart, indicating the value above which sample
(statistical)values may fluctuate when the process is in a state of control.
A full and proper assessment of control requires consideration of process
performance with respect to the upper control limit and laws of probabil-
ity in addition to the lower control limit. See control limit and in-control
process.
LSL 99

lowet.limit (LL) A term used by different practitioners to refer (ambigu-

ously) to lower specification limit, lower process limit, or lower control
limit.

lower process limit The threshold below which exactly .135%of the mea-
sures, scores, values, or observations M.

lower speciication limit (LSL) The point below which any measured
value of a critical characteristic will result in a defective product in that
it no longer conforms to specifications. See also process capability and
confbrmance.

LSL S e e lower specification limit.

 M
machine shop See flexible manufacturing.

Malcolm BaZdrige Award See Malcolm Baldrige National Quality

A d .
Malcolm BaZdrige National Quality Award An annual award to recognize
U.S. companies that excel in quality management and quality achieve-
ment. The award was created by Public Law 100-107 in 1987 and sup-
ported by the Foundation for the Malcolm Baldrige National Quality
Award, established in 1988. The award is named for Malcolm Baldrige,
who served as secretary of commerce from 1981 until his tragic death in a
rodeo accident in 1987. The award is administered by the American soci-
ety for Quality Control, under contract to the National Institute of Stan-
dards and Technology (NIST). The award has three eligible categories:
manuficturing companies, service companies, and small businesses. Up to
two awards may be given in each category annually. Companies participat-
ing in the award process submit applications that include completion of
the award examination application and review by a board of examiners.

MAN See metropolitan area network.

management, daily See daily management.

management by objective (MBO) An approach to management of per-

formance through the establishment of annual objectives for each em-
ployee and rating the performance of the employees at the end of the
performance period (typically this is an annual process cycle) based on
performance to those objectives. Note that this approach is not an element
of total quality management,but is an “old school” methodology that is in
contrast with a quality philosophy. See also 14 points, especially point 11.

management by polity See policy deployment.

100
master manual 101

mamgenmt-cmMh ermr See operator self-control.

mamgenmt of constraints See constraint management.

‘ ,and machines
manpower, methods, materials See four M’s.

manufacturing resource planning (MRP n) A software system composed

of multiple autonomous and semi-autonomous modules, including part
master files, a bill of materials module, master scheduling, material re-
quirements planning, procurement management, shop floor activity con-
trol, inventory management, and distribution requirements planning.
Very large MRP I1 systems have also grown to encompass previously sepa-
rate software modules such as accounts payable and accounts receivable,
and an engineering database.

manufactzlring styks See batch manuhcturing, cellular manuhcturing,

and flexible man&cturing.

mappngprocess, custonter See customer mapping process.

market shre, importance of The value of this indicator of market posi-

tion. According to many leading business analysts, market share is an es-
sential measure of company health, and one of the most powehl predict-
ors of the company’s future. Buzzell and Gale, in their book The PIMS
Principhs (New York: Free Press, 1987),say:

The 1960s and 1970s brought a dawning realization that

market share is key to a company’s growth and profitability.
The 1980s have shown just as clearly that one factor above
all others-quality-drives market share. And when supe-
rior quality and large market share are both present,
profitability is virtually guaranteed.

master manual A method of documenting the T Q M program in

order to lend consistency and structure throughout the different
functional areas and processes of the organization. The master man-
ual is actually a set of manuals. One manual is kept in each depart-
ment. Another is kept that summarizes the manuals in all the depart-
ments within a division, and so on, up to the overall company master
manual. The manual should provide continuity of structure by de-
102 master scheduling

picting the flow-down of company vision and mission statements

through policies and practices down to individual operating proce-
dures. It should also provide the linkage between T Q M improve-
ment team initiatives and overall company strategy. Finally, the mas-
ter manual system should provide clear documentation of the
company quality system, supporting the company’s efforts in the
areas of Malcolm Baldrige National Quality Award application and/
or I S 0 9000 registration. A fairly comprehensive outline of master
manual contents and uses is provided in The TQM Trilogy, by Maho-
ney and Thor (New York: AMACOM, 1994).

master scheduling The overall production schedule that drives the

production of the manufacturing organization through the loading
of end-product delivery dates (known as independent demand) into the
material requirements planning system. The accuracy of the master
production schedule is essential to effective manufacturing materials
planning.

materialjow See parts flow.

material handling, automated See automated material handling.

material requirements planning (MRP) A computerized system of

planning the purchase and/or production of parts from raw material
through end-item delivery. The MRP system uses a structured bill of
material for the product, and applies lead time offsets to determine
when each component must be procured or fabricated in order to
support assembly and delivery of the finished product. MRP also of-
fers the option to apply any of a number of lot sizing alternatives and
general inventory policies (such as reorder points, safety stock levels,
and scrap factors) to material and production planning activities.
MRP is generally offered by software suppliers as one module of a
complete software system called manufacturing resource planning,
which is much broader in scope.

materials, designer See designer materials.

matrix, decision See decision matrix.

matrix diagram The central portion of a quality function deploy-

ment house of quality diagram. This matrix compares the needs of
measurement, evolution of business 103

the customer to the activities performed within the company. The

relationships between activities performed and customer needs are
identified as weak, medium, or strong in order to target improve-
ment efforts.

MBO See management by objective.

mean An arithmetic average. In a statistical context, the mean refers

to the central tendency of a distribution, around which the scores or
values of a sample set are distributed. See also arithmetic mean.

mean sums of squares The value calculated by dividing the sum of

squares from the target factor by its degrees of freedom. The result
is an unbiased estimate of the variability of a factor in a given popula-
tion.

mean time between failures (MTBF) The average time between a re-
pairable product’s failures for a specific unit of measure (e.g., operat-
ing hours or cycles).

mean time toJirst failure (MTTFF) The average time interval be-
tween the first usage of a repairable product and the first failure of
that product for a defined unit of measure (e.g., operating hours or
miles).

mean time to repair (MTTR) The average length of time required

to restore the machine or system to an operable state after it has
become inoperable and repairs have been initiated.

means, analysis of See analysis of means.

measurement, evolution of business A look at business measures over

time. Traditionally, business measures were all financially based and
offered only a historical view of business performance within a fairly
narrow frame of reference. In the 199Os, however, business measures
have become broader in order to become more predictive of future
performance and more relevant in terms of the competitive position
of the company within its industry. Business measures are becoming
higher-level metrics with more comprehensive scope, such as market
share. (See market share, importance of.) In addition, measures are
104 measurement error

growing more tightly linked from top to bottom throughout the com-
pany and focusing increasingly on process-oriented measurement
criteria.

measurement error The difference between measured values and

actual values for the dimension measured. See also true value.

measurement process The combination of equipment (instruments,

gauges, fixtures, etc.), standards, procedures, operators, and envi-
ronmental characteristics operating on an input (product or process
to be measured) and producing an output (measurement).

median The middle of a set of data points, or the point that sepa-
rates the data into halves. Fifty percent of the data values fall below
the median, and 50% fall above it. The median is also referred to as
the separation boundary between the second and third quartile of
data.

median and range ( X - R ) chart A control chart that uses the me-
dian and the range of sample values recorded in order to determine
whether a process is in control or if an assignable cause is present
and should be addressed. See in-control process.

mesokurtic See kurtosis.

metric See tactical metrics.

metropolitan area network (MAN) A computer and telecommunica-

tions network that connects the individual local area networks
(LANs) into a larger interactive network.

middle management’s role in TQM A look at the part middle man-

agement plays in T Q M implementation. The role of middle manage-
ment in a T Q M environment is to communicate and consistently/
effectively manage the conversion of executive-level strategies
through T Q M improvement initiatives. This linkage is often missing
in failed and failing T Q M programs. Middle management is respon-
sible for chartering the specific improvement initiatives, monitoring
team-level progress, communicating successes and challenges to the
executive level, securing required resource deployment from execu-
MTTR 105

tive levels, and sharing “best practices” developed by individual im-

provement teams with the entire company. See also communication
and chartering a team.

mission statement A statement based on the vision, the guiding val-

ues and principles of the company. The mission statement is a funda-
mental element of policy deployment. (See also organizational val-
ues.) Among the components listed in an effectively constructed
mission statement are the organization’s:

1. Business (what the organization does)

2. Customers (who the organization’s customers are)
3. Plans for the future (where the organization is going over the
coming decade)

mode The score or value that occurs most frequently in a set of ob-
servations.

motivation The reasons why people take certain action. In a T Q M

environment, the foundation for motivated employees is put into
place with effective development and deployment of the vision and
strategies. Once this foundation is in place, motivation becomes a
function of communication, reward systems, individual empower-
ment, and the charisma of the leadership involved. See also reward
and recognition system.

moving range chart See RM chart.

M R chart Moving range chart. See RM chart.

MRP See material requirements planning.

MRP ZZ See manufacturing resource planning.

MTBF See mean time between failures.

MTTFF See mean time to first failure.

MTTR See mean time to repair.

106

mu (p) The mean of a population or universe.

multivariate charts A type of chart that depicts multiple types of

variation simultaneously. (For example, multivariate charts may de-
pict the level of variation over time and also the level of variation
from piece to piece.) This chart helps the user take a good prelimi-
nary step in narrowing the field of potential causes of variation for
designing an experiment.
 N
N The size of a finite population or lot.
n The number of units represented in a sample (sample size).

NASA Award The National Aeronautics and Space Administration

(NASA) Quality and Excellence Award. This award was created in 1985 to
heighten the importance of quality for NASA's contractors. It was recently
renamed the George M. Low Award, in honor of a former NASA deputy
administrator.This award is usually given in two categories: large business
and small business. The categories covered include performance achieve-
ments (including such areas as cost, schedule, and quality) and process
achievements (including such areas as commitment, communication,
planning, training,and awarddrecognition.)

National Instit& of Standards and Technology (NISQ A group within

the US. Department of Commerce that distributes copies of the Malcolm
Baldrige guidelines to the public and governs the Malcolm Baldrige Na-
tional Quality Award. The address is:
National Institute of Standards and Technology
Route 270 and Quince Orchard Road
Administration Building, Room A537
Gaithersburg, MD 20899

natural language processors See artificial intelligence.

natural tolerance The range corresponding to upper process limit-

lower process limit for a process or population distribution within
which the middle 99.73% of the scores will fall.

natural work group A group of people that share a common work

area and regularly spend most of their working time together.

107
108 natural work team

natural work team A natural work group that has a clearly defined
mission, usually continuous improvement.

NDE See nondestructive evaluation.

NDT See nondestructive testing.

needs, customer See customer needs.

networking phase (phase V ) The final phase of the T Q M process.

(See the Introduction of this publication for an overview of all five
phases.) In this phase, the successes of the internal T Q M process
improvements are fully exploited and outside organizations (espe-
cially suppliers) are brought into the improvement activity. The net-
working (or diversification) phase is comprised of five steps:

1. The strategic improvement plan is communicated with outst.de organi-

zations. The two most important outside entities with whom this infor-
mation should be shared are customers and suppliers. It’s important
to share the information with customers to ensure that what the com-
pany is planning will achieve the ultimate goal of satisfying customer
needs. It’s critical to share the plan with suppliers, especially key sup-
pliers, in order to bring their improvement plans into alignment and
generate opportunities for shared savings from these improvement
activities.
2. “Lessons learned” and current examples of TQM operational suc-
cesses should be shred with these outside organizations. This step bolsters
customer confidence and promotes brand loyalty. It also demon-
strates clearly to suppliers that what you are asking them to do really
does work and that you have proven this in a “live” work environ-
ment.
3. Customers and suppliers should be visited on a regular basis. Strong
linkages from the customer through the supplier must be developed
to improve responsiveness along the value-adding chain and to antic-
ipate the customers’ needs.
4. The TQM process must be institutionalized. This step involves
making the T Q M process into the formal operating process of the
company through adopting the necessary changes in policies, proce-
dures, standard practices, purchase contracts, customer require-
ments documentation, and so on.
normal distribution 109

5. The process must be developed into a continuous loop. In other

words, the process is repeated over and over again, institutionalizing
continuous improvement. This will require the constant resetting of
goals, continuous learning on the part of everyone in the organiza-
tion, and continuous benchmarking activity.

NIST See National Institute of Standards and Technology.

nominal data Discrete data with only two classifications. For exam-
ple, numeric values that represent open circuit vs. closed circuit, us-
able vs. unusable, and occurs vs. does not occur. The control charts
used for nominal data are the np-chart and the c-chart.

nominal group technique A technique generally used by problem-

solving teams following brainstorming activity. This approach uses
ideas jotted down silently and independently on slips of paper, after
which the ideas are read off by each individual, and the group priori-
tizes the ideas to reach consensus on the approach to be taken.

nominal value The customer-defined or related optimum condition

as measured for a critical quality characteristic. See also target value.

nonconformity The failure of a product or service to fulfill a speci-

fied requirement. Also, the presence of an attribute or characteristic
that, due to count, frequency, or both, is found objectionable by the
customer or consumer. See also defect.

nondestructive evaluation (NDE) The evaluation of a product that

does not negatively affect the serviceability of the product.

nondestructive testing (NDT) The testing of a product that does not

negatively affect the usability or serviceability of the product. In sta-
tistical analysis, a nondestructive test is one where the true value asso-
ciated with the specimen evaluated is reobtainable; that is, it does
not change as a result of time, testing, or both. See also destructive
testing.

normal distribution A symmetric, unimodal, bell-shaped distribu-

tion of continuous data where skewness (y3)and kurtosis (y4) both
equal zero. See Figure 3 1. See also bell-shaped curve.
110 np-chart

Figure 3 1 . Normal distribution.

np-chart A control chart, based on the approximation of the bino-

mial distribution, depicting the number of defective units per sam-
ple over time (used when the sample size remains constant from sam-
ple to sample).
 0
objective Statement of a specific result that is to be accomplished within
a designated period of time.

OC curve See operating characteristic curve.

ODNODZF See office document architectudoffice document inter-

change format

o@ce document arehitecturelojce document interchnge format (ODA/

ODW) An information technology standard for office document formats
that is in the early stages of acceptance in U.S. markets. It is expected
that, with International Standards Organization backing, ODA will gain
widespread support in Europe before it attains a sigdicant level of accep-
tance within the United States.

open systems intercon- (OSI) A standard that will be followed by

supplien who offer integrated information systems throughout the sec-
ond half of the 1990s. More formally known as the International Stan-
dards Organization (ISO) Open Systems Interconnection model.

operating c h r d (OC)curve A graph usehl in many areas of qual-

ity and statistics and also in other areas. In this graph, the vertical axis
represents the probability of an event occurring or of taking some action
(such as accepting a shipment or rejecting a hypothesis). The horizontal
axis represents the magnitude or amount of a characteristic or property
upon which the occurrence of the event (or the decision) is based. In
quality activities, OC curves are commonly used to indicate the changing
probability of accepting a lot or a shipment under increasing “true” defect
rate contamination of the lot. OC curves are also used commonly in statis-
tics to describe the probability of observing a particular statistical value
(such as a given mean difference) under increasing amounts of “true”

111
112 operating plan

population mean difference (also called effect size).Because the probabili-

ties represented on OC curves are frequently dependent on sample size
and effect size, they are helphl in determining the proper sample size for
different kinds of investigations.

qber-utingplun See business plan.

operation controlfeedbmk The feedback resulting from mechanisms de-

veloped to measure the success of improvement initiatives in overall oper-
ations performance. This feedback provides clear evidence of the level of
improvement or degradation in operations, and when effectively devel-
oped, links the visible changes to specific TQM initiatives underway in the
business. Feedback should include the information that is deemed to be
critical and necessary to monitodmanage the improvement efforts. It
should be formatted in a clear and structured manner that is sensible
and usehl to the management audience. It should also include thresholds
defining when corrective action will occur.
The primary source of operation control feedback is process-level
metrics (the metrics developed to reflect the performance of the individual
processes of the business). Secondary sources of operation control feed-
back include suggestion systems, quality surveys, and audits (such as Bal-
drige criteria audits and I S 0 9OOO criteria audits).

operational&jniti.n The definition of terms, variables, constructs, or

concepts by the means or methods used to measure or manipulate them.
The purpose of an operational definition is to allow one to bring the term,
variable, construct, or concept into quantitative empirical reality. There-
fore, it is important to make explicit and public the specific steps and
actions taken to measure, assess, or manipulate an event or phenomenon
of interest. This is accomplished by the operational definition. As an exam-
ple, one could operationally define general employee satisfiction in terms
of the number of days each employee is absent from work, whereas an-
other could define employee satishction in terms of the score of an em-
ployee satisfiction survey. The strength of a given material might be oper-
ationally defined as the force, in pounds, required to break it; another
could define it in terms of its resistance to penetration; and yet another
could define it in terms of hours of use in a particular application before
Mure.

operutm-cmtroZlubkkm Error that operators can monitor, detect, and

work to prevent. (See also detection vs. prevention.) Standard operating
operator-controllable error 113

procedures (SOPs) and standardization efforts need to be designed to

minimize operator-controllable error. In effea, we must plan to reduce
human error with our SOPs.
There are three types of operatorcontrollable errors: inadvertent,
technique, and conscious.

1. Inadvertent errors, which must be planned for in the creation

of SOPs and work methods. Inadvertent errors are the result of the
natural incapacity for humans to pay attention 100% of the time. Dis-
tinguishing features of inadvertent errors are:
A. Unintentional. The operator does not want to make the
error.
B. Unwitting. At the point of making the error, the operator
has no knowledge that he or she has just made an error.
C. Unpredictable. No one knows beforehand just when a
worker will make an error, or the type of error that will be
made. Also, no one knows who will be the next operator to
make an error.
2. Technique errors, which arise because the operator lacks some
essential technique, knowledge, or skill needed to avoid making the
error. Distinguishing features of technique errors include:
A. Unintentional. The operator wants to do good work.
B. Selective. The errors are confined to those error categories
for which the missing technique is essential.
C. Consistent. Operators possessing the technique can consis-
tently avoid making the errors and the operators who lack
the technique will consistently make the error.
D. Witting and unwitting. Technique errors can be of both
kinds. Operators may realize at the time that they have
just committed an error, but they lack the technique to
prevent it from happening (e.g., a golfer who knows when
he or she is consistently slicing the ball). The opposite is
also true in that an operator may not realize that his or
her lack of technique may result in premature product
failure or waste further downstream or at the customer
site.
E. Unavoidable by the unazded operator. Operators lacking the
technique do not know what should be done differently
from the current procedure. Therefore, the error can go
on and on without some assistance to eliminate it.
114 operator self-control

The evidence of technique errors is the presence of a consistent dif-

ference in error rates among workers. These consistent differences
in error rates are selective to specific error types. Within technique
errors, there are two subcategories of errors:
A. Best practice. This occurs when a minority of operators pos-
sess the technique, knowledge, or skill that is essential to
avoid making the error.
B. Damaging practice. This occurs when a minority of opera-
tors lack the technique, knowledge, or skill that is essential
to avoid making the error.
3. Conscious errors, which operators know they are making and
usually intend to keep making.
Distinguishing features of conscious errors include:
A. Intentional. The error is the result of a deliberate intention
by the operator to commit the error.
B. Witting. At the time the error is made, the operator knows
that an error has been made.
C. Consistent. The operators who cause conscious errors usu-
ally do so on a continuing basis.

operator self-control The knowledge and action to meet all quality,

safety, and productivity standards. To reach such a state, the opera-
tors must be supplied with the following:

1. The means of knowing what is expected of them. What are they

supposed to do? This knowledge can come in the form of training,
specifications,samples of good and bad product, standard operating
procedures, instructions, etc. It is important, though, that this
knowledge is clear and unequivocal.
2. The means of knowing what their actual PerfOrmance is. Are they
doing what they should do? This knowledge can come in the form of
control charts, process measurements, inspections, information
from daily management boards, feedback from downstream pro-
cesses, and information from the customer feedback system, etc.
What is important here is that operators know whether or not they
are meeting their assigned responsibilities. (See also feedback, cus-
tomer.)
3. The means for regulation. If the actual performance does not
meet the desired performance, how will the operators change what
 organizational structure and kwels 115

they are doing? Regulation can come in the form of reaction plans,
product disposition plans, and process adjustment procedures, etc.

If all these needs and criteria have been met, without exception, then
the operators are in a state of self-control. They have all the means
needed to do good work and any errors are operator-controllable.
(See operator-controllable errors.) If any of these criteria or needs
have not been met, then the managerial job has not been completed.
Therefore, all resulting errors are considered management-control-
lable.

ordinal data Quantitative data that provide a classification accord-

ing to order. The magnitude of the ordered values may not, how-
ever, represent uniform differences. Ordinal data generally are one
of three types:

1. Count data, such as number of scratches or number of inclu-

sions.
2. Subjective scale data (also called rank data), derived from re-
sponses to an evaluative scale. For example:
rated background 1 2 3 4 5
' -
pick up off a hot line t
none slight moderate heavy extreme

3. Low-resolution data, commonly obtained from measuring a

continuous underlying function with a gauge or device having
limited measurement resolution.

The control charts used for ordinal data are the c-chart and the
u-chart. See also attribute data.

organizational structure and levels The three conceptual levels of

organizational structure that have distinct and separate roles in a
T Q M implementation:

1. Upper-level or executive management. Sets strategy, monitors

progress, and manages improvement efforts at the highest
level. (See also executive's role in TQM.)
2. Middle management. Defines and develops improvement proj-
116 organizcrtional values

ects around the processes of the company. (See also middle

management’s role in TQM.)
3. Working level (teams and individuals). Focuses on implemen-
tation of the improvement initiatives.

organirational values The principles that create an organization’scul-

ture. (See company culture.)These are often most clearly evidenced in
general operating characteristics such as internal competition vs. inter-
nal teamwork; emphasis on protecting proprietary data vs. open rela-
tionships with customers and suppliers; continuous improvement vs.
“if it ain’t broke, don’t fix it.” Often, in environments where T Q M im-
plementations are not successful, one strongly contributing reason is
that the working values of the company are different from the ones
formally stated in company literature and by the company’s executive
leaders. Since one of the most powerful influences on company culture
is the example set by upper-level management, this problem can only
be remedied when company executives begin to demonstrate a per-
sonal commitment to TQM principles in their daily business activities.

OSI See open systems interconnection.

out-of-control process A process in which the characteristic(s)
under statistical evaluation are not in a state of statistical control due
to the effects of special causes of variation. See Figure 32. See also
in-control process, lower control limit, and upper limit.

out-of-spec, out-of-specification Terms indicating that a unit(s) does

not meet conformance requirements. See also specification.

overall equipment effectiveness See asset utilization.

ownership Having the legitimate power and authority to carry out

a designated action item. It is important to note that under some
circumstances legitimate ownership may exist without the ownership
actually being accepted by the individual involved. Another situation
can arise when individuals assume ownership without any legitimate
power or authority. The bottom line in real terms is that when a
person or group assumes responsibility for the performance of a
process, and has the ability to effect change in that process (whether
that power is legitimate or not), there is ownership. This concept is
very important. Many times, T Q M implementations are ruined by
ownership

w
f

t
Process in Control

...........................................

...........................................
I I

Points Outside Control Limits

Run
I I

.........................................

.........................................

(8 points in a row above or bebw center)

............................................

Trend
(6 consecutive increasing or decreasing)

l
........................................... I
I

...........................................
W

I
-
Figure 32. Patterns indicating possible out-of-control conditions.

1 ' '
Cyclical Effect
I..........................................---

PY54 .-----.-----.--~.
---------.--..-
1

Lack of Variability
(15 consecutive points in center 1/3)

, 1 1 1

Excessive Variability
1

......................................

appointing individuals or teams to improve a process when the team

or individual has no interest and refuses to become actively involved.
In these situations, the ownership is official, but inconsequential. As
a result, no improvement occurs and usually T Q M is blamed for the
failure. See also employee involvement.
Alternating Values
1

(8 consecutive points outside center 1/3)

....................................

(14 consecutive alternating points)

I

............................................

............................................
1 1
,

1 1
117

1
pchurt A control chart, based on the normal approximation of the bino-
mial distribution, reflecting sample percentages or proportions of defec-
tive units. When the process is found to be in a state of control, the pro-
cess average (p-BAR)can be employed to estimate n (pi), the percentage
or proportion defective in a population. See Figure 33.

p p p o r t h The ratio of the number of units in which at least one event

of a given classification occurs to the total number of units sampled. (A
unit is only counted once even if several events of the same classification
are encountered.)

paradigm A popular term referring to a shared set of assumptions about

how things work. Joel Barker, most commonly considered to be the man
who popularized the use of this term, defines paradigms as “a set of rules
and regulations (written or unwritten) that does two things: (1) it estab-
lishes or defines boundaries;and (2) it tells you how to behave inside the
boundaries in order to be successfid.” (Purudzgm,The Business OfDiScovmng
the Fukure, New York: Harper Business, 1993.)

parameter In statistics, the parameter is a measure computed from all

observationsin a given population. Population or universe parameters are
designated by Greek letters. See also descriptive statistics.

Pureto unaEysis An analysis that identifies and differentiates the vital few
from the trivial many (or the “important many” as Juran suggested they
may in bct be). Pareto analysis was originally developed by Vilfredo Par-
eto, a nineteenthcentury Italian economist who discovered this relation-
ship in the context of wealth and its distribution among the general popu-
lation. The principle was defined in its broader context by Juran in 1950.
In almost any population, 2Wo of the population represent 8Wo of the
value in that population. When ranked in descending order, it becomes

118
 Pareto chart 119

12 - 1 " " 1 " " 1 " " 1 " " 1 " " 1 " " 1
-

10 - -
'
10.31%

-
C

2?
2 - -
. 1.11%

0 - I l . . . . l . . . . I . . . . l . . . . l . . . . l
-
0 5 10 15 20 25 30
Week
relatively easy to identlftr the most sigdicant items to control. For exam-
ple, when performing cycle counts of inventory, the highest dollar value
items should be counted most frequently. In most cases, this means that
counting just 2Wo of the part numbers involved will cover 80% of the
dollars in inventory. Similarly, when identlfylng where to focus improve-
ment efforts, a Pareto analysis is ofien used to iden* the relatively few
causes that generate most of the problem conditions.

Pureto chart A graph (usually a bar chart) that depicts the values of char-
ted Occurrences in descending order, identdjmg those that represent
roughly 20% of the occurrences and roughly 80% of the value of the
occurrences measured. The ranking of defect causes from most to least
120 Pareto diagram

sigmficant in this manner has become such a popular approach to focus-

ing targeted improvement efforts that it is now designated one of the
seven basic tools of quality. See Figure 34.

Pareto diagram See Pareto chart.

partsJow A term describing the general flow of parts and material

through a manufacturing fictory. The flow can generally be described as
either process oriented or as a functional flow. (See functional layout and
process-oriented flow.) The same principles that govern an analysis and
improvement effort in this area also apply to businesses other than manu-
ficturing, such as processing insurance claims. Most often, hnctional lay-
outs and flows result in h r more non-value-added time and material in
flow than process-oriented layouts and flows contain. See also value-added
analysis and flow charts.

parts d i o n (PPM) The number of defects or errors per million

operations performed or per million products produced. Three sigma
( 2 3a) equates to 2,700 defects or errors per million, and six sigma
( f 6a)equates to 3.4 d e f m or errors per million.

P D U cycle See plando-check-actcycle.

PDCA discipZine A strategy used to achieve b e u g h improvements

in safety, quality,morale, delivery, cost, and other critical business objec-
tives. The PDCA philosophy is the glue that binds together policy deploy-

Figure 34. Pareto chart.

1most
significant
77-,
Categories (e.g., defect types)
significant
percent chart 121

ment and daily management into a cohesive set of processes. Break-

through improvements are best achieved and sustained by individuals and
teams using the scientific method z+s it is expressed in the plandecheck-
act cycle broken down into easily implemented steps. See Figure 35. See
also total quality management.

PDCA management discipline Fosters the use of the plandocheck-act

cycle in the implementation of continuous improvement, standardiza-
tion, and innovation structure so that planning and feedback become an
integral part of management systems and procedures. See Figure 36.

PDCA strategy See PDCA discipline.

percent chart See p-chart.

Figure 35. The plan-do-check-act discipline within policy deploy-

ment.

ACT identify
- Adopt Targets
- Abandon and Goals
- Adjust
Identify
Strategies
to Achieve
Act Plan theTwrets
and Goals

Review Check Do Train in the

Progress: ldentied
Strategies
- Against the
Targets & Goals
- Within the Undertake the
Strategy improvement
Activity
Figure 36. Evolution of the PDCA management discipline and modern management PDCA cycle.

1. Determine goals and targets. Act Determine Plan

P 2. Determine methods of reaching goals.
- Take
appropriate
goals and
targets.
Determine
action. methods of
3. Engage in education and training. reaching

D 4. Implement work.
goals.
Engage in

c 5. Check the effects of implementation.

Check the effects
of implementation.
education and
training.

Implement
Check work. Do
A 6. Take appropriate action.

Adapted from Ishikawa.

plan-do-check-act (PDCA) cycle 123

pet.ceptual map See customer map.

PERT See p q p m evaluation and review technique.

PERT chart A chart produced by the PERT (program evaluation and

review technique) process that reflects the sequential interdependencies in
major activities. PERT charts, also called arrow dhgrams, are often used
to manage large projects.

phase I, p.eparationphase See preparation phase (phase I).

phase IZ,planningphase See planning phase (phase II).

phase IZZ, assessment phase S e e assessment phase (phase III).

phase W,implementaturn
* phase See implementation phase (phase IV).

phase V , nelworiiingphase See networking phase (phase V).

PIT Process improvement team. See also planning phase (phase 11).

plan, business See business plan.

planao-checkact(PDC4) cycle As described in PDCA management dis-

cipline, this cyclic approach is comprised of four steps: plan, do, check,
and act. Though often used in problem solving, as illustrated in the follow-
ing example, PDCA has other applications (e.g., in a project to implement
new software):

1. Plan. After one or more factors are identified as causal to a

process problem, possible solutions are brainstormed and the most
attractive solution approach is selected. Then the implementation is
planned.
2. Do. The solution is implemented. This is usually done in a
pilot setting as opposed to beginning immediately with the entire
population of possible applications, in order to minimize risk while
the effectiveness of the solution is proven effective.
3. Check. Confirm and document the effects of the action taken.
This is frequently accomplished through process control chart data
124 planned grouping

and through before-and-after studies. If the solution proves ineffec-

tive, the team returns to the plan stage and begins again.
4. Act. This involves deploying the solution across the entire
population of possible applications, once it has been proven effective,
and making it a permanent feature of ongoing process operations.
Documentation is updated and training is performed as needed to
ensure that the change is fully integrated into daily operations.

The PDCA cycle is sometimes called the Shewhart cycle and the
Deming cycle-Walter Shewhart developed the concept and described
it in his book Statistical Methods from the Viewpoint of Quality Control
(Washington, D.C.: The Graduate School, Department of Agricul-
ture, 1939), and W. Edwards Deming popularized the concept in
Japan.

planned grouping This term refers to those methods associated with

improving the sensitivity of experimental designs through the use
of specially designed experimental patterns. Examples include block
designs, matched pair designs, and repeated measures designs.

planning, extrapolation See extrapolation planning.

planningphase (phase ZZ) This is the second phase of a T Q M imple-

mentation program. (See the Introduction of this publication for an
overview of all five phases.) This phase consists of these basic steps:

1. Securing of commitment. Secure commitment of the estimated

resources required in order to validate the commitment level of the
company to these changes and lay the foundation for real progress.
2. Establishing the TQM implementation organization and infrastruc-
ture.
A. A business lead team is established and a facilitator is as-
signed to lead regular meetings. The lead team’s role is
to lead the policy deployment activities throughout the
organization. This involves constructing a workable plan
and committing adequate resources over a sufficient pe-
riod of time to see the company through the implementa-
tion process. The lead team is comprised of senior-level
executives from the entire organization, including repre-
planning phase (phase ZZ) 125

sentatives of organized labor when they are a part of the

operation. They should provide a sense of legitimacy, im-
portance, and structure to the T Q M program.
B. A steering committee that represents a cross-section of the
company’s employee base is established. The role of com-
mittee members is to facilitate the implementation of
T Q M throughout the organization. They do this by pro-
viding both horizontal linkage between improvement
teams and senior management, and horizontal linkage (in
a process form) between functional organizations. The
steering committee identifies those improvement projects
that are likely to have the greatest impact on the organiza-
tion and the greatest likelihood of success. Committee
members then charter the process improvement teams to
perform these improvements, providing guidance, re-
sources, and technical assistance as required. They set
specific targets and metrics for improvement as each team
is chartered and monitor progress toward those targets,
taking corrective action as needed. They also provide
regular, comprehensive status on overall improvement
progress to the business lead team. (See chartering a
team.)
C. Process improvement teams (PITs) are established, char-
tered, and guided by the steering committee, as described
in the previous text. They are comprised of process ex-
perts from each of the functional areas involved in the
process that is targeted for improvement. Teams may also
include suppliers and customers. PITs are temporary in
nature and disband when projects are complete.
3. Strategic planning and policy deployment.
A. Mission statements are reviewed when they exist and ad-
justed as required to reflect the importance of meeting
customer needs and utilizing TQM principles in that pur-
suit.
B. A vision of the future state of the business is generated-a
vision that is compelling and desirable to all company
stakeholders.
C. Company goals and objectives are defined or modified to
reflect the mission and vision statement(s) generated in
126 planning phase (phase ZZ)

the previous steps. The goals and objectives should pro-

vide specific targets for improvement initiatives and tacti-
cal planning and should incorporate the anticipated im-
pact on the bottom line (return on investment). No fewer
than three and no more than seven goals and objectives
should be stated to retain adequate focus through the
course of the initial round of T Q M implementation. The
output of this process should be the backbone of the com-
pany’s operating or business plan.
4. Development ofthe implementation plan. Planning should then be
done to determine which specific processes will need to be improved
to attain the specific goals and objectives defined previously. Plan-
ning will identify timing and steps involved in the deployment of
process improvement teams, and high-level monitoring and report-
ing milestones. Planning should also include a cursory costbenefit
analysis and an estimation of required resources for the completion
of each major improvement initiative.
5 . Awareness training.
A. The impacts of a total quality philosophy are discussed
and described in detail, based on information from case
studies of similar implementations and popular T Q M
press. The costs as well as the potential benefits are out-
lined so that expectations and anticipated commitments
are understood as clearly as possible.
B. Principles and practices of T Q M are then described, usu-
ally by an outside consultant who has led other organiza-
tions through the process. A basic understanding of the
T Q M tools and techniques and their applications is real-
ized in this step, so that management understands the ter-
minology and process elements that will become a regular
part of the company’s culture.
C. The importance and role of the customer are reviewed
and explored in some detail, focusing on the (internal and
external) customer as the definer of quality, and as an inte-
gral part of the improvement process.
D. Commitment and ownership are established by involving
all of the participants of both the business lead team and
the steering committee in the development and imple-
mentation planning activities. Specific improvement tar-
poka-yoke 127

gets and time frames are tied directly to the reward struc-
ture of these employees, and they are required to state
their commitment or withdraw from the process. See also
reward and recognition system.

plateau distribution A distribution of values measured and plotted

in chart form, which looks like a hill with a flat top and no really
discernible peak. This form of distribution typically has slight tails on
each end of the distribution and is most often the result of multiple
bell-shaped distributions with centers roughly evenly distributed
throughout the range shown on the chart. See Figure 37. See also
kurtosis.

platykurtic See kurtosis.

PLC See programmable logic controller.

point-of-use inspection Inspection performed by the user (down-

stream internal customer) at the point where the part is used (typi-
cally where the part is installed in the next assembly) in a manufactur-
ing operation.

Poisson distribution A probability distribution for discrete random

variables that can take integer (whole) values (ordinal data). See Fig-
ure 38. It takes the form

poka-yoke A Japanese term that refers to foolproofing a design such

that all ambiguity is removed and it becomes virtually impossible to

Figure 37. Plateau distribution.

128 policy

Figure 38. Poisson distribution for average = 1, 5, 10, 20.

--- 1
- 5

-
mmmm 10
20

0.0 10 20 30 40 50
X

set up a machine or produce a part or an assembly incorrectly. This

is often accomplished through designing the components of the tools
and the assemblies so they will fit together only in the proper orienta-
tion and sequence.

policy A statement of principle that exists to direct a specific course

of action by the employees of a company.

policy deployment One of three integrated quality management sub-

systems supporting total quality management (TQM). It is an inte-
grated strategic planning process in which a company develops and
deploys three to five major initiatives, focusing resources on a critical
few (usually one to three) companywide breakthrough areas re-
quired to improve competitive position or to counter threats in a
sociaVeconomic environment. Also referred to as hoshin planning.
See Figure 39. See also planning phase (phase 11) and breakthrough
improvements.
precision 129

Figure 39. The fundamental elements of policy deployment.

Visions, Values,
Beliefs, Guiding
Principles

Corporate/
Company
Mission
Statement

Policies Deployed
to Organization
Corporate/ Annually
Company
Strategic Plan

Strategic
Product-Market Business or
Analysis Operating Plan

Input from Corporation or

Customer Quality Company
Assurance System Focal Point
Chart

population A collection of all identifiable observations in accordance

with a set of rules or boundaries. Also called a target population or
universe, the term population refers to the aggregate of all observations
of interest to the researcher. In this case, the term of interest refers to
the total set of observations to which the researcher wishes to general-
ize the results of the research study. See also research population.

PPM See parts per million.

precision The relative degree of agreement among repeated and

reproduced individual measurements or test results for one or more
measurement processes or systems. The components of variation for
precision are associated with repeatability and reproducibility. The
degree of precision in a measurement process is generally designated
as the standard deviation of measurement error, or uE.See also accu-
racy.
130 predictive maintenance

predictive maintenance Within a TQM environment, the use of an

array of statistical models and methods intended to signal corrective
action to a process before associated critical product variables can
move out of control. In a more limited context, predictive mainte-
nance is also used to describe the performance of preventive mainte-
nance in a predictive, rather than reactive, mode of operation.

preparation phase (phase I) The period when executives examine

the nature of their businesses and the problems that must be over-
come in order to become more competitive in their markets. (See the
Introduction of this publication for an overview of all five phases.)
T Q M is one such strategy that may be applied across industries.
Nonetheless, executive-level management must consider the range
of possible strategies in light of their business environment and de-
cide for themselves whether T Q M is the approach that they should
embrace. Management must understand the magnitude of change
(not merely physical and process changes, but the far more difficult
cultural changes as well) that must occur in a successful T Q M imple-
mentation. This self-examination is best done with the help of a top
consulting team who is trusted by executive management and has a
demonstrated track record with a broad range of improvement strat-
egies. See also consultant’s role in TQM.
It is not uncommon for the top management team to establish
or reestablish its vision and mission statements for the company dur-
ing this phase. At the very least, they should be carefully reviewed as
management tries to determine the future goals of the company and
why they have been selected.
Finally, the culmination of the preparation phase should be com-
mitment to an approach, and if management decides to proceed with
TQM, the next four phases of implementation should be under-
taken. Management must understand, however, that commitment to
T Q M means that the entire philosophy and culture of the company
will necessarily change. (See company culture.) The company must
orient itself toward process-based management, process control, and
continuous process improvement with one overriding objective: sat-
isfaction of customer needs, and eventually, satisfying those needs
even before the customer clearly understands what they are! The
leadership of T Q M implementation cannot be delegated. TQM im-
plementation must be led by top management in a lead-by-example
manner.
prioritizing improvements 131

prevention Avoidance of nonconformances in products and services;

not allowing an error or defect to occur. It takes many forms depend-
ing on the application involved, but all of its forms involve reducing
variability. See also detection vs. prevention, prevention costs, non-
conformity, and conformance.

prevention costs Those costs incurred because of action taken to

prevent errors or defects, lowering or eliminating the other two cate-
gories of the cost of quality: appraisal costs and failure costs. Pre-
vention costs include education and training costs, design review
costs, and process control-related costs. See also education and train-
ing (for total quality).

preventive maintenance All actions performed in an attempt to re-

tain an item in its specified condition by providing systematic inspec-
tion, detection, and prevention of developing failures. See also de-
tection vs. prevention.

price (as relates to cost) See cost-price model.

prioritizing improvements It is common during the planning phase

of a TQM implementation to identify several potential improvement
projects to be undertaken, all of which offer some benefit to the com-
pany. Prioritizing these opportunities correctly involves mapping
each potential improvement to the associated critical business per-
formance metrics, as defined with the identification of critical busi-
ness processes. A matrix may be constructed, listing opportunities
along the Y axis, and critical business performance metrics along the
X axis. Then each opportunity may be rated by the business lead
team as having a high, medium, or low impact on each metric. A
further refinement that adds value to this analysis is rating each busi-
ness metric in terms of its relative importance. In this manner, a me-
dium impact improvement on a metric with a value o f “ 10” would be
undertaken before an opportunity that was determined to have a
high impact on a metric that is only valued at “2,” assuming that
higher numbers rank more highly in terms of their importance. In
this manner, the cumulative positive effect of each opportunity may
readily be compared to all other opportunities, and priorities should
become clear. See also comparative and relative importance.
132 Prize, Deming

Prize, Deming See Deming Prize.

problem-solving strategy (PSS) Working toward a solution to a prob-

lem. While there are several problem-solving approaches in use by
TQM practitioners, a good example is the technique employed by
Florida Power & Light when seeking the Deming Prize. This ap-
proach follows the steps of the PDCA process:

Plan
1. Reasons for improvement. Adopt team mission that defines the
improvement theme (general problem). Document impor-
tance and why problem should be worked on. Schedule prob-
lem-solving activities.
2. Current situation. Understand the nature of the problem. Se-
lect appropriate components of the theme (subproblem) for
improvement. Set an improvement target.
3. Analysis. Identify and verify the root causes of the problem or
undesired state.

Do
4. Countemeasures. Select and implement countermeasures that
will counteract the root causes of the problem.

Check
5. Results. Check the effectiveness of the countermeasures. De-
termine if problem correction or desired change has been
achieved. If not, return to step 3.

Act
6. Standardization. Prevent the problem from recurring. Stan-
dardize and achieve control to maintain the gains. (See also
standardization and standard operating procedure.)
7. Future plans. Plan what to do about any remaining problems
and evaluate the team’s effectiveness.

process Any repetitive series of events. Generally, the use of the

term process (versus activity or operation) implies that the events are
operating in a state of statistical control and are therefore affected by
process capability 133

only common causes of variation, rendering the output of the events

stable and predictable. See also control (statistical).

process, business See business processes.

process, support See support processes.

process capability The degree to which the process (output) meets

(will meet) specifications or requirements on an ongoing basis. Capa-
bility refers to the future performance of the process, not to its past
or present performance, although past or current data are used in
making the assessment of capability. The prediction of future per-
formance requires stability and predictability in the process output.
Therefore, actual or true capability cannot be assessed until the pro-
cess has been documented to be in a state of control. If the process is
not in a state of control, any calculation of capability must be called
potential capability, and in most cases should not be calculated at all.
See also in-control process and out-of-control process.
Proper calculations of capability consider both process variabil-
ity (spread) and process central tendency (center). In many cases,
variability and closeness to a proper target, as opposed to being cen-
tered between the specifications, must be considered. It is for these
reasons that there are at least three commonly used indices of capa-
bility: C,, Cpk,and Cpm.Since the C, index accounts only for variabil-
ity (and not centrality or location), one or both of the other indices
are used along with C,.
C, is an index that is the ratio of the spread across the specifica-
tion (total tolerance or upper specification limit minus the lower
specification limit) to the spread across the process distribution (the
center 99.73% of the process distribution, also called the natural tol-
erance). For a normal distribution, the spread across the middle
99.73% is represented as six sigma or 6a (which can be estimated
from control charts by R-bar/d2or S-bark,). This value can also be
obtained by subtracting the lower natural process limit (LNPL) from
the upper natural process limit (UNPL). For the nonnormal case, the
spread across the middle 99.73% of the process distribution is ob-
tained by distributional fitting methods. In either case the following
holds for C,:
(USL - LSL)
c, =
(UNPL - LNPL)
134 process capability

and for normal distributions, the simplification for C, is:

cp= (USL6a
- LSL)

C, can take on values from 0 to + 00.

The Cpk index considers not only process variability, but also the
degree of centering. Cpk can be understood as decreasing the value
of C, as a function of the degree to which the process average devi-
ates from the center of the specification limits. Indeed, an initial for-
mula for Cpk was:

Ip, - spec mzdpointl

C, = C, ( 1 - k ) where k =

i"' LsLl
Another more popular and easy-to-follow calculation for C,k in-
volves the use of z-scores calculated for the specification limit values
relative to the process distribution. A z-score is calculated for the USL
called Z, and for the LSL called Z,. The absolute values of these two
z-scores are taken and the smaller of the two is used in calculating
Cpk as follows:

And an alternative formula for Cpk is:

(p - Nearest Specification Limit)

cpk =
3a

The above calculations for C are predicated upon the condition

p:
of normality in the process distribution. If this is not the case, the
above values will be incorrect and one should use an alternative such
as the equivalent Cpk.
The equivalent c p k index (called C ") can be obtained if one is
pk.
able to estimate the percent of observations expected to be out of
specification for the particular nonnormal process distribution. In
process capability index 135

that case, the largest proportion (percent) out of specification is de-

termined, that proportion is taken to the unit normal table, and an
estimated z-score is retrieved and treated as an estimate of the mini-
mum z in the following formula:

The value of Cpk (and CpkE)relative to c, is given in the following

expression: 0 5 Cpk 5 C,. Some practitioners use a negative Cpk value
when the process average is outside of the specification limits, but in
actuality Cpk is undefined in such a case.
C,, is an index of capability that takes into consideration both
process variability and deviation of the process average from the tar-
get (or nominal) value. Therefore, it is an index that reveals total
variability or deviation from the target or nominal. When the target
is centered between the specification limits, C,, gives information
about all of the process data relative to the target and in that sense is
more informative than c p k , although Cpk is usually calculated.
C,, is computed as follows:

where No is the nominal or target value.

The value of C,, relative to C, is 0 5 C,, 5 C,. It should be
noted that C,, is almost always less than c p k , but there are cases when
C,, could be greater than c p k .
The capability ratio (CR) is the spread across the process distri-
bution divided by the spread across the specification and is the recip-
rocal of the C, index.

pocess capability index A calculated value that provides an indica-

tion of the degree to which the process output is expected to meet
specification. Most index values are defined in such a way that a
marginally capable process has an index of 1.O and increasing index
values indicate increased capability as with C,, Cpk,C,:, Ppk, and Ppm,
etc. (An index of less than 1.0 means that the process is not capable.)
136 process change

However, the capability ratio (CR) indicates increased capability with

values smaller than 1.O. See also process capability.

process change See change.

process characteristics See critical process characteristics.

process/equipment reliability A look at the probability that equip-

ment or a process will function as intended. Manufacturing systems
and mechanical equipment are subject to failures that may be classi-
fied into three basic groups:

1. Znfant mortality. Usually associated with new or renewed

equipment, the result of improperly installed equipment, or
extremely low-life components due to manufacturing varia-
tions.
2. Random. Unreliable design, insufficient redundancy, unex-
pected and unpredictable occurrences, or product misuse.
3. Wearout. Predictable and limited to age or cycle.

See also reliability and reliability modeling.

process improvement team (PIT) See planning phase (phase 11).

process-orientedflow An approach to the layout of the workplace

that organizes the work in a way that reflects the processes that the
work goes through. Similar to thefocusfactory approach, the principle
of process-oriented flow is based on generic routings for entire fami-
lies of parts. This approach is in sharp contrast to the functional
layout, which is more traditional, especially in manufacturing com-
panies. Traditional functional approaches group similar equipment
in the same geographic areas of the workplace, which requires the
work to be moved repeatedly over long distances in order to perform
the operations required for transformation. The process-oriented
approach locates all of the equipment and tools required through the
production routing in close physical proximity, regardless of equip-
ment type. This allows for minimal throughput times and work-in-
process inventory. See also just-in-time manufacturing and parts
flow.
product characteristics 137

process-oriented plant layout See process-oriented flow.

process planning, automated See automated process planning.

process value analysis (PM) A phase in the implementation of activ-

ity-based costing in which a process is studied and analyzed to deter-
mine the level of value added. This would likely involve:

Observing a process and recording a listing of each process

step.
Identifying the resources required to complete each step (e.g.,
time, manpower, and floor space).
Assessing each step to determine which steps add value to the
product or service rendered.
0 Assessing what level of value added is for the process as a
whole.

process variables The various conditions and settings at the opera-

tion that influence product quality. Process variables may include:

time temperature machine speed

pressure clearances tooling dimensions
tension flow settings tooling surface conditions
recipes/formulas dwell times ramp times
roller positions tooling heights vacuum pressures

See also critical process characteristics and quality tables.

procurement costs See total cost concept in purchasing.

procurement technology, advanced See advanced procurement tech-

nology.

product characteristics The various conditions and measurements

of the product that describe the quality of the product. Product char-
acteristics include such elements as:

size weight color

strength hardness print-to-cut
138 product market analysis

surface condition defect level register

resistivity turn on voltage moisture of product

See also critical product characteristics and quality tables.

product market analysis See strategic product market analysis.

production validation A series of tests and appraisals performed to

assure that initial production parts will meet the design intent. Typi-
cally, production validation includes the following elements:

1. Ensure that all requirements are included in engineering

specifications.
2. Conduct functional tests and durability tests.
3. Complete all requirements before initial shipment.
4. Include all sample sizes and acceptance criteria.
5. Obtain concurrence from manufacturing and quality control
through feasibility reviews.
6. Rerun the first five steps when processes or designs change.

projit, short-term A metric that has motivated management to oper-

ate under a short time horizon when it comes to profit planning.
Many CEOs and CFOs believe that capital investment diminishes the
price of company stock, is too costly, and leads to unfavorable depre-
ciation requirements. These executives think that capital investment
is detrimental to their own careers and should be minimized. The
result of this approach is undercapitalized companies with large cash
reserves in paper investment vehicles and deliberate, forced obsoles-
cence of the business. This is particularly true in manufacturing com-
panies.
T Q M is at odds with this philosophy, not because T Q M advo-
cates large capital investments (it usually does not), but because the
heart of T Q M is a long-range view that quality will bring about
growth through continuous improvement. That improvement often
causes the need for capital investment in an incremental manner.
Furthermore, more significant expenditures are sometimes made to-
gether with just-in-time implementations, moving equipment, and
reconfiguring factory floor operations. This conflict between short-
term profitability and long-term growth and sustainability is a long-
standing problem and is only exacerbated when senior-level manag-
program evaluation and review technique (PERT) 139

ers do not recognize the long-range nature of a commitment to

TQM.

profound knowledge, a system of A phrase coined by Dr. W. Edwards

Deming to describe the underlying foundation from which the 14
points follow as application and to provide a working theory in order
to understand and optimize organizations. This system is made up of
four component parts, each related to the others: appreciation for a
system, knowledge about variation, theory of knowledge, and psy-
chology.

program evaluation and review technique (PERT) A project manage-

ment tool that produces a flow chart depicting the sequences and
interdependencies of critical activities in the project. The output
chart from this process enables users to quickly identify the critical
path (the longest sequential string) of activities through the project,
keeping attention focused on the most important progressive steps.
(See critical path method.) The PERT process includes the following
steps:

1. Identify all of the signijicant activities in the project. Then establish

milestones that represent their completion.
2. Identify the sequence an which these activities must bepe$ormed. This
is usually accomplished most easily by working backward,
identifying interdependencies and completion times for each
step. For example, before the last step is performed (usually
packaging or shipping), determine what other steps must be
completed. Before these steps are performed, see what other
steps must be completed, and so on.
3. Develop t h e j o w chart. The activity completion milestones are
each represented by numbered or labeled “bubbles,” with
connecting lines representing the dependencies between the
activities. Each line should be identified with a number that
represents the time to do this work, or the entire diagram
may be plotted on a time-based scale, with the weeks called
out along the bottom of the chart.
4. Count the number of weeks in each “string” of interdependent activi-
ties. (Actually the time increment is irrelevant. It could be
minutes, weeks, months, or years.) This way the critical path
may be identified for the attention of the project managers.
140 programmable bgic controller (PLC)

5. Maintain the chart. Since the actual work time required always
varies from estimated times, the actual critical path will shift
back and forth between lengthening and shortening strings
of activity.

programmable logic controller (PLC) A computer used to control

process variability through machine adjustment based on preloaded
computer instructions. PLCs are usually RISC (reduced instruction
set computing) based and are applied to machines performing repet-
itive work in industrial environments.

project plan development The most important step in the planning

phase (phase 11) of a T Q M implementation. The project plan must
be comprehensive, realistic, and reflective of the goals and objectives
of the company as it focuses on satisfying customer needs. The devel-
opment of project plans will flow from policy deployment when de-
ploying the strategic plan of the company.

PSS See problem-solvingstrategy.

pull signals Devices used to signal the prior operators that they are
authorized to produce another unit. Pull signals may be as sophisti-
cated as electronic messages on a computer screen or as simple as an
empty spot in the previous operation’s outgoing area.

pull system A way of managing shop floor (or even office-level) activ-
ity that minimizes work-in-process and dramatically improves
throughput time by eliminating interoperation queues. A pull system
requires two things: a pull signal and a fixed upper volume limit. A
fixed upper volume limit means that operators must stop producing
parts whenever they have not received their (pull signal) authoriza-
tion to produce more. If only two pull signal devices are provided
between operations, then a maximum of two parts will exist in flow
between those operations at any given point in time. Pull signals are
sometimes called kanbans. See also push systems.

push systems The antithesis of the pull system. Push systems allow
production to continue based on a predetermined schedule. Push
systems launch orders into the production system on a scheduled
interval and assume that they will come out the end of the process at
push systems I41

the end of the designated throughput times. Even the best closed-
loop push systems are much less responsive to in-process variation,
and therefore much less effective for controlling production and
work-in-process than pull systems.
QA See quality assurance.

QF’ See quality function deployment.

QIS See quality improvement strategy.

Quality The never-ending reduction of variation around a customerde-

fined target in the absence of defects. Customers most often define quality
in terms of the product or service’s attributes that are most important to
them. The most readily perceivable evidence of quality is the satishction
of customer requirements the first time and every time the product or
service is used. See also big Q little q.

q d i t y , average outgoing See average outgoing quality.

quality, c m f m m e See conformance quality.

quulity, dejinitims of The ways quality is defined. Juran defines quality

as fitness for use and Deming defines it as follows: “Improvement of the
process, which increases uniformity of output of product, reduces rework
and mistakes, reduces waste of manpower, machine-time, and materials,
and thus increases output with less effort.” The most important thing to
remember about definitions of quality is that they are only relevant to
the extent that they provide value to the customer, as perceived by the
customer.

quality,&sign See design quality.

QualiEy (as related to cost and schedule) See costfscheduldquality rela-

tionship.

142
quality circles 143

Quality assurance (QA) All activities essential to ensuring that a product

or service will meet the needs of an internal or external customer. See
customer, external, customer, internal, and customer needs.

quuliiy assurance, customer See customer quality assurance.

Quality assumnce, total See total quality assurance.

4uality assumnce organhtim, functions of The functions of the quality

assurance organization shift dramatically with the implementation of
TQM. There are several categories of these changes, including:

1. Qydiiy’s target. Where the QA organization’s target shifts from

conformance to requirements to minimal variation from target
values.
2. @a& planning. Where the function of QA shifts from defining
the minimum requirements for product and process measure-
ments to quality function deployment
3. Dejnzng acceptabilzty. Where the shift is away from defining what is
acceptable and toward continuous improvement.
4. Cost reduct.ian. Where emphasis shifts away from reducing Mure
costs toward reducing overall process variation in order to drive
down total costs of nonquality.
5. I n s p e h . Where the emphasis shifts away from inspection by
the QA organization toward operator-based process control with
occasional QA audits.
6. & s w q . Where the responsibility for quality shifts away from
QA and toward every individual for the quality of his or her own
work.

Quality c h u r m See satisfier.

Quality circles Groups of people who meet regularly (sometimes with

their supervisors) to i d e n q and solve quality-related problems within
their processes. Quality circles reached their peak in terms of quantity in
the 1980s. More recently, these efforts have shifted into other mediums
such as process improvement teams and natural work groups. The pri-
mary activities performed by the quality circle include detailed process
definition (including identification of the internal and external custom-
ers), elimination of non-value-added activity in the processes, and estab-
144 quality control

lishment of performance baselines and targets. Quality circles typically are

composed of three to nine employees who work in the same geographic
area, most often operating equipment within a defined sequential process.

quality control The operational methods and activities used to hEll

quality requirements, usually through and including the evaluation of
data. Quality control, in the “old school,” was the name given to the de-
partment responsible for product inspection. See also statistical quality
control.

quality control circles See quality circles.

quality diss&jers/wWs See dissatisfier.

Quality function dgblovmat (QFD) A method used to translate customer

needs into appropriate technical requirements for every stage of product
development-fiom the concept stage through distribution. The QFD
process begins with a step often called “listening to the voice of the cus-
tomer.” The American Supplier Institute has defined QFD as “a system
for translating consumer requirements into appropriate company require-
ments at each stage from research and product development to engineer-
ing and manuficturing to marketindsales and distribution.
QFD uses a table called the house of quality, which has several
“rooms” defining the relationships between design requirements and cus-
tomer requirements (the “what” room); the relationship between what the
company does and its satisfiction of design requirements (the “how”
room); objective target values (the “how m u c h room); and both engi-
neering and customer competitive assessments.
QFD is usually implemented in four phases:

1. Product plunning. Design requirements are matched to customer

needs, allowing the critical design requirements to be identified.
2. Part d+!qm&. Part characteristics are matched to critical design
requirements identified in step 1, allowing the critical part charac-
teristics to be identified.
3. Process planning. Process operations are matched to critical part
characteristics, allowing key process operations to be identified.
4. Production planning. Key production operations are matched
against production requirements to i d e n e the operations that
quality management values 145

must be brought into tight process control in order to result in

customer satisfaction.

See also quality tables.

quality improvement strategy (QIS) A disciplined plan for the appli-

cation of statistical tools for bringing a critical product characteristic
into a state of control, capability, and/or acceptability.

quality level, acceptable See acceptable quality level.

quality loss function A parabolic approximation of the monetary

cost and quality loss that results when a quality characteristic (satis-
fier) does not meet its target value. The cost of this deviation in-
creases quadratically as the quality characteristic moves farther away
from its target. The quality loss function was first introduced in this
form by Genichi Taguchi, and the formula used to compute it varies
depending on the type of quality characteristic being observed. See
Taguchi loss function.

quality management, absolutes of See absolutes of quality manage-

ment.

quality management systems Systems composed of effective, compre-

hensive plans; strong T Q M leadership at every level of the organiza-
tion that has “bought in” to T Q M as a way of life; discrete metrics
that accurately reflect process health and provide the basis for con-
tinuous improvement; and active continuous improvement toward
customer satisfaction. See also planning phase (phase 11) and imple-
mentation phase (phase IV).

quality management values Among the values most successful T Q M

organizations hold are:

1. Open, honest communication with employees, customers,

and suppliers
2. Absolute faith in the concept that customer satisfaction
equates to quality, and that quality equates to success
3. Measurable, discrete performance goals at every level, bench-
marked against the best in that business
146 quality planning, advanced

4. Realism, and the recognition that T Q M is a lifetime commit-

ment, not a quick-fix plan for improving short-term profit-
ability

See also organizational values.

quality planning, advanced See advanced quality planning.

quick-change devices Devices used in manufacturing setup reduc-

tion activities to eliminate the adjustments required during setup
changeovers. Among the devices most commonly used on manufac-
turing equipment are quick-change couplings, clamps, locator pins,
and fixed-position stops for dies and fixtures.
R chart Range chart. A common type of control chart used with contin-
uous (or variables) data to evaluate the stability of the range (R) of sub-
p u p samples with subgroup size less than 8. The results may also be
used to estimate the overall process variance. The R chart should be ac-
companied by an X chart or an % (median) chart to show central ten-
dency.

R,,,chart Moving range chart. A control chart that usually accompanies

an X chart and reflects variation in the process by plotting the range
between each value and the previous value.

RAB See Registrar Accreditation Board.

radar chart See spider chart.

m d m suntpling The selection of size n units from a population where

each sample has an equal chance of being selected.
. .
mru&m&wn The process of assigning experimental units to treatment
conditions, or the order of testing, in a purely chance manner. Typically,
randomization is assodated with three experimental design functions:

1. Random sampling or selection

2. Random assignment
3. Randomization of order

RMR studies See gauge repeatability and reproducibility (studies).

mngechart SeeRchart.

147
148 ratio data

ratio data An interval data scale, structured such that the ratio between
any two measures results in a meaningfbl metric. See also continuous data
and variables data.

reaction p h n Accompanied by an associated control chart, a plan that

details what the empowered process owner is to do, check, and act upon
should the monitored process go “out of control.” Specific reactions corre-
sponding to the nature of the out-ofcontrol conditions, which may ap-
pear, should be identified in the reaction plan whenever possible. See also
out-of-control process.

recovery Good output from the operation divided by the total output
from the operation. Also known as yield.

red bead experiment An experiment developed by Dr. W. Edwards Dem-

ing to demonstrate that managers should not rank employees based on
previous performance because observed performance daerences may
well be attributed to the system, not the employees. In this experiment,
4000 beads (20% red, 8Wo white) are placed in a jar and stirred, and then
six people each select a sample of 50 beads. Each person’s goal is to pro-
duce white beads, because the customer will reject red beads. When every-
one has a sample, the red beads are counted. The limits of the variation
between the six individuals that can be attributed to the system are calcu-
lated and each will likely hll within these calculated limits. The calculations
illustrate that there is no real evidence that one person will be a better
performer than another in the future, so it would be a waste of manage-
ment’s time to try to determine why, for example, Susan produced eight
red beads and Bob produced twelve. Instead, the red bead experiment
demonstrates that management should work to improve the system,
allowing everyone to improve performance.

Regktrur Accreditation Board (RAB) A private, nonprofit organization

that ensures the competence of quality systems registrars. Their method
involves providing accreditation to international standards for the regis-
trars.

regktration, I S 0 9000 The presence of a supplier in the registration list-

ing of an accredited registrar, following the conferring of “registered”
status by that registrar. See also I s 0 9000 and certification, I s 0 9000.
reliability modeling I49

registration, (ISO) registrar The presence of a registrar in the listing of

accredited registrars maintained by the Registrar Accreditation Board.

relationship diagram A chart developed to assist in the problem-solving

process and one of the seven basic tools of quality. The diagram depicts
the relationships between naturally grouped sets of problems and ideas.
In using a relationship diagram, the process improvement team:

1. Writes each problem (or symptom) in a circle.

2. Draws arrows between the circles showing the influence of one
idea on the others (Solid line arrows mean strong influence and
are counted as a value of “1.” Dotted lines represent weak influ-
ence, and are represented by values of “3’).
3. Ranks the circles by the cumulative influence value emanating
from each circle, and the circles with the highest value are targeted
first for resolution.

See also affinity diagram.

reliability The probability that a system, production center, ma-

chine, or component will perform its intended function (in a stated
condition or environment) for a predetermined period of time with-
out failure. Reliability of a system is indicated by R,. See also process/
equipment reliability.

reliability, information Information reliability has a direct impact

on the customer’s perception of the quality of the product when it
misleads customers as to the use of the product, the maintenance of
the product, the specifications of the product, or other factors that
may become dissatisfiers. See also expectations, customer.

reliability, tooling See tooling reliability.

reliability modeling Application of different mathematical models to

identify the probability of failure in a system. Reliability may be
thought of as a “bathtub curve,” when it is viewed as a function of
time, spread over the life cycle of the system. See Figure 40. During
the startup phase of any system, the failure rate is high, and comes
down a steep curve until it flattens out sharply during the mature
useful Zfe portion of the system’s life cycle, then accelerates steeply
150 reliability prediction

1 2 3
Infant + - 4 Random -b 4 Wearout
Mortality Failures Failures

Failure
Rate
A.

again in the later wearout period at the end of the system’s life cycle.
Since the middle period or useful life period is much longer than
either of the other two periods, the graphing of this failure probabil-
ity curve produces the “bathtub” look, and hence the name. At these
different stages, and depending on the nature and quantity of the
equipment involved in the process, different mathematical models
may be utilized to determine the reliability level of the system. The
exponential model, the reliability series, the reliability of parallel sys-
tems series, and the k-out-of-n system are among the reliability mod-
els employed. See also process/equipment reliability.

reliability prediction Quantitative assessments of reliability. These

predictions are used by engineers to assess the reliability of systems
and subsystems, facilitate trade-offs between reliability and other fac-
tors, and forecast system and subsystem performance over time. In
the case of existing systems, the reliability of the system may be as-
sessed by establishing a base failure rate, breaking this failure rate
down into manageable elements, and adjusting the elements to re-
flect changes. In the case of completely new designs, the approach
used will require a combination of some of the following elements:

1 . Failure modes and effects analysis

2. Comparison of new design components with similar compo-
nents and failure modes of existing designs
3. Laboratory testing
4. Subcomponent testing
resistance, planning for 151

repeatability The ability of a measurement process to show close

agreement among a number of consecutive measurements of the
same unit being measured under the same operating conditions. The
variability of measurement within a single measurement system.

replication The acquisition of two or more observations under a set

of identical experimental conditions, or the collection of multiple ob-
servations under a set of identical experimental conditions, or the
collection of multiple observations at each of the treatment conditions
tested. The primary purpose of replication is to provide quantifiable
and accurate measurement of the precision of the design, where pre-
cision relates to the concept of experimental error.

reproducibility The ability of a measurement process to show close

agreement among repeated measurements of the same unit being
measured by multiple measurement processes or systems contribut-
ing to a single process stream. The variability of measurement error
between measurement systems.

requirements planning See distribution requirements planning.

research population That portion of the population or universe

available for sampling.

resistance, planning for Identifying anticipated areas of resistance

to T Q M implementation and broad-brush planning for dealing with
that resistance. This planning should occur in the first two phases of
the T Q M implementation. Specific planning should also be done in
the third and fourth phases, as process improvement teams are char-
tered and deployed. (See chartering a team.) Resistance to change is
often the single most commonly cited reason for the failure of T Q M
programs, after lack of top management support. Resistance to
change may be most effectively mitigated in the following ways:

1. Proactive management that communicates honestly and

openly about the changes in expectations and the importance
of those changes. See also communication.
2. Changes in measurements, particularly performance mea-
surements, in a way that supports the new ways of doing busi-
ness.
152 resolution

3. Movement from vertical organization to horizontal, process-

based organization. This will more effectively communicate
that the process, and control of variability in the process, are
now of paramount importance to management.
4. Empowerment and involvement of employees at every level
of the organization in the development and deployment of
change. See also employee involvement.

resolution The smallest increment or unit of measure available from

a measurement process.

resource infrastructure The network of people and systems from

which resources are drawn for process improvement efforts. These
resources include people, equipment, materials, facilities, dollars,
time, and information.

resource matrix A matrix that compares available resources (typi-

cally in man-hours) to the time required to perform functional re-
sponsibilities, achieve strategic objectives (focal point items), and
maintain critical tactical items.

response variable See dependent variable.

reward and recognition system A component of employee involve-

ment. Employee ideas and efforts (e.g., problem-solving, quality im-
provement) should be recognized, regardless of their outcome. This
recognition may consist of:

1. A personal visit by management to the employee’s work area

to congratulate him or her and express sincere appreciation
2. Selection for presentation at a quarterly quality review ses-
sion
3. Posting the employee’s name on the company bulletin board
in the implementation area, so that others can express their
appreciation for the employee’s efforts
4. Monetary awards, trips, and so forth, for the suggestions that
generate the best improvement results, or that enhance em-
ployee safety

risk The probability of making an error while testing a hypothesis.

See also Type I error and Type I1 error.
run 153

risk sharing An approach to new product development that in-

volves suppliers’ assuming some part of the risk in the new product
development and launch by investing in their own start-up costs in
return for some end-product profits. See also strategic alliances.

risks of TQM The risks involved with implementing T Q M in any

business. Among the most frequently experienced risks are:

1. The drag on resources required to implement such a com-

prehensive program while still running the day-to-day busi-
ness.
2. The long time horizon for return on investment. (See profits,
short-term.)
3. The fear of change, resulting in some near-term upheaval,
dissension, and disruption. (See planning for resistance.)
4. Disintegration occurring between processes as only a few are
improved at a time.

Recognizing and anticipating these factors will go a long way

toward enabling management to effectively reduce the risks associ-
ated with these factors.

robustness Aspects or characteristics of a product or process that

remain relatively stable (exhibit minimum variation levels), even
though factors that influence operations or usage, such as environ-
ment and wear, are constantly changing. See also stability.

root cause The most fundamental reasons (causes) that require-

ments are not met. Identifying root causes often requires “peeling
back” the reasons first expressed when looking at the ultimate symp-
tom of a problem. For example, poor paint finish may be caused by
problems with paint adherence. Problems with paint adherence may
be caused by hydraulic fluid leaks. Hydraulic fluid leaks may be
caused by loose connections. Loose connections in hydraulic lines
may be caused by poor tightening methods or tools.

run An out-of-control condition identified on a control chart as

eight or more consecutive points on one side of the center line (i.e.,
at least eight consecutive points above or eight consecutive points
below the center line). A run indicates that a special cause of varia-
154 run chart

tion is present that has caused a shift in the process average. See also
out-of-control process.

run chart A graph depicting consecutive measures plotted around

the center line.
s-chrt A standard deviation chart. A type of control chart involving the
subgroup standard deviation, s, which is used to evaluate the stability of
the variability within the process. Usually accompanies an X-barchart
and is used when the subgroup is greater than or equal to 8.

sumpk A portion or subset of a population or to take a sample.

sumpZe size The number of experimental units, objects, or subjects in a

sample or subgroup.

sumpling,random See random sampling.

sumpling error The expected and quantifiable discrepancy between a sta-

tistic and its associated parameter due to the sample size and the variabil-
ity of the population fi-om which the sample was drawn. Also sometimes
called chance.

SAPPHO project A study conducted by the University of Sussex in

Brighton, Great Britain, which paired successhl innovators with unsuc-
cessfd innovators who were seeking similar innovations. The study
showed that the most signficant difference between success levels of the
innovations developed was the level of attention paid to identlfylng and
understanding cusfomer needs. SAPPHO is an acronym for ScientificAc-
tivity Predictor fkom Patterns with Heuristic Origins.

satisfaction, customer See customer satisfaction.

satisJim A quality characteristic. Any feature (property or attribute) of

the product, material, process, or service that is required to achieve fitness
for use.These characteristics exist in several categories, including:

155
156 scatter diagram

1. 72chnobpa.l (e.g., hardness, acidity)

2. Psrchological (e.g., beauty, status)
3. Time-oriented(e.g., reliability,maintainability)
4. Crmtractual (eg., guarantee provisions)
5. Ethical (e.g., service shop honesty, courtesy of salespeople)
scatter diagram A graph used to analyze the relationship between two
variables. Two sets of data are plotted, with the Y axis representing the
variable predicted (dependent variable) and the X axis representing the
variable making the prediction (independent variable). The scatter dia-
gram shows the nature of possible relationships between the two variables.
To determine ifa relationship exists, the person performing the evaluation
must know about the variables because an apparent relationship between
two variables does not prove a relationship exists and the data are fi-e-
quently analyzed using statistical tools. See Figure 41. The scatter diagram
is one of the seven basic tools of quality.

schedule (as related to cost and quality) See cost/scheduldquality rela-

tionship.
SDCA cyck Standardize-do-check-act cycle. Similar to the plando-
check-act cycle, except that the plan element is replaced with a standard-
ize element.

1. Standardize. A standard operating procedure (SOP) is devel-

Figure 4 1 . Scatter diagram.

a
a. a

a:

Part Size
setup time reduction 157

oped, usually by a natural work group or by a cross-functional team,

after one or more solutions are identified as effective in resolving a
process problem or to reduce variability.
2. Do. Implement the standard operating procedure (SOP), on
a pilot basis if possible.
3. Check. Check the effectiveness of the SOP: Is it being followed
and is it clear and easy to understand? Confirm and document the
effects of the action taken. This is usually accomplished through
process control chart data and through before and after studies.
4. Act. Return to the standardize phase and make adjustments to
the SOP if necessary. If the SOP is all right, implement it across all
applicable areas.

See also daily management.

selection, employee See employee selection and training.

selfmanaged and selfdirected teams A group of people empowered

to manage their own work functions. The group is typically com-
posed of all the individuals who perform work making up a defined
business process.

sequential convergence See convergence, sequential.

serpentine cells A popular layout for cellular manufacturing opera-

tions that deploy operators inside U-shaped work cells that are con-
nected end-to-end in a serpentine manner. See also U-shaped cells.

setup time reduction An important part of reducing manufacturing

lot sizes in order to reduce cycle times and work-in-process inven-
tory. Setup time reductions were made popular as part of the just-
in-time manufacturing movement in the late 1970s and early 1980s.
One of the most famous practitioners in this area is Shigeo Shingo,
who developed a technique called Single Minute Exchange of Dies
(SMED). Reduced setup times result in smaller economic order
quantities (EOQs),and thereby allow fewer parts to be manufactured
in a given lot. There are several approaches to setup time reduction,
such as dedicated setups, quick-change devices, and fixed-position
mechanisms to reduce or eliminate adjustments.
158 seven basic tools of quality

seven basic tools of quality The fundamental tools that organizations

use to help understand their processes, which leads to their improve-
ment. These tools are:

1. Cause-and-effect diagrams
2. Check sheets
3. Control charts
4. Flowcharts
5. Histograms
6. Pareto charts
7. Scatter diagrams

These tools are described in more detail in separate entries

within this text.

seven quality control tools See seven basic tools of quality.

shape (distributional) The visual appearance of measurements plot-

ted on a histogram. Distributional shape is generally described by
measures or indices of symmetry and peakedness. Specific shapes or
distributions are described by name in other entries within this text
(i.e., bell-shaped, plateau, double-peaked, and truncated). See also
skewness and kurtosis, two shape parameters.

shareholder Owner of a company’s stock.

shareholders vs. stakeholders Shareholders are individuals who

maintain financial ownership positions in a company; partners or
stockholders. They are a subset of stakeholders. Shareholders typi-
cally have a financial interest in the company, but little or no interest
or involvement in day-to-day operations. Stakeholders are all of
those individuals and organizations who have a significant interest in
the business, not merely in the financial returns of the business.
There are several categories of stakeholders, including:

1. Government. Who is interested in company compliance to all

regulations and laws concerning the environment, employee
safety, tax reporting and collection, securities regulations,
and related areas such as transportation.
ShewhatVDerning cycle 159

2. Shareholders. Who are interested in the return on monies in-

vested in company stock.
3. Executive management. Who are interested in progress toward
strategic goals and objectives, and short-term business per-
formance, which is likely to have a direct impact on their per-
sonal financial stability and career growth.
4. Customers. Who are interested primarily in product or service
quality and value. This includes product or service features,
price, financing terms, and availability.
5. Employees. Who are interested in their own compensation, job
security, personal safety, career growth, and satisfaction in
their work.
6. Suppliers. Who are interested in a stable and growing demand
(market) for their products and services, and profitability
within their product lines.
7. Community. Who is interested in good stewardship and contri-
bution by the company. The community appreciates the con-
tribution of companies to the local tax base and social contri-
butions (participation in fund-raisers, etc.). The community
also wants companies to be good stewards of the local envi-
ronment and resource base.

Shewhart, WalterA . The man who created the control chart, who is
also known as the father of statistical quality control. Shewhart,
ASQC’s first honorary member, described the basic principles of SQC
in his book Economic Control of the Quality of Manufactured Products
(New York: Van Nostrand, 1931).

Shewhart control charts Control charts (X-bar, median, individuals,

R, RM,s-, np-, p-, c-, and u-charts) named for the late Dr. Walter
Shewhart, who described the basic principles of control chart theory
in his book Economic Control of the Quality of Manufactured Products
(New York: Van Nostrand) in 1931. See also average and range chart,
s-chart, np-chart, p-chart, c-chart, RM chart, u-chart, and X chart.
Shewhart/Deming cycle An improvement methodology attributed to
Shewhart and Deming by Svenson et al. in The Quality Roadmap (New
York: AMACOM, 1994). The methodology includes four phases:

1 . Assessment. Data gathering and analysis to identify baseline

performance and opportunities for improvement.
160 Shingo Prire

2. Design and testing. Producing a validated improvement that

is ready to be deployed, using standard design and testing
methodologies.
3. Deployment. The implementation of the improvement across
the spectrum of possible applications is developed.
4. Integration. Integration of the improved processes with other
processes throughout the organization.

Shingo Prize The Shingo Prize for Excellence in Manufacturing,

conferred in order to promote world-class manufacturing and to rec-
ognize companies that excel in productivity and process improve-
ment. Established by the late Dr. Shigeo Shingo, the prize is spon-
sored and managed by the Utah State University Partner’s Program.
The major categories of evaluation in consideration of the Shingo
Prize are:

1. Total quality and productivity management culture and in-

frastructure. (See company culture.)
2. Manufacturing strategy, process, and systems.
3. Measured quality and productivity.
4. Customer satisfaction.

short-term projit See profit, short-term.

sigma The standard deviation of a population.

simultaneous convergence See convergence, simultaneous.

six sigma A term coined by Motorola to emphasize the continuous

reduction of process variation. For a normal distribution, the proba-
bility that a value falls within +- 6 a of the mean (p)is 99.99966%. If
k 6a falls inside the specification limits, then there are only 3.4
defects per million parts produced from the designated process.

six-sigma quality program A program developed by Motorola to em-

phasize process variation reduction and defects. This program con-
tains these key ingredients:

1. A superordinate goal of total customer satisfaction

2. Common, uniform quality metrics for all business areas
special causes 161

3. Identical improvement rate goals for all business areas

4. Goal-directed incentives for both management and em-
ployees
5. Coordinated training in why and how to achieve the goal

The program is organized in this sequence:

1. Identify the product or service to be provided.

2. Identify the customer(s)of the product or service.
3. Identify and communicate the need to provide a product or
service that satisfies the customer. See also communication.
4. Define the processes performed to satisfy the customer.
5. Make the process mistake-proof and eliminate all wasted ef-
fort.
6. Ensure continuous improvement through process control.

skewed distribution A distribution of data that, represented on a

histogram, is manifested in a peak that is off-center from the mid-
point of the range and tapers off much more sharply on one side of
the peak than the other side. See Figure 42. See also skewness.

skewness A descriptor of shape that indicates the degree to which a

distribution is nonsymmetrical. Also called the third moment, or y3
(population)and g3 (sample). See also kurtosis and skewed distribu-
tion.

SMED An acronym referring to single-minute exchange of dies, a setup

reduction technique developed by Dr. Shigeo Shingo.

SOP See standard operating procedure.

span of control The scope of responsibility over which an employee

or organizational unit has direct control.

SPC See statistical process control.

special causes Specific and unusual external circumstances that re-

sult in unpredictable process variation. They are relatively few in
number. Special causes are also called assignable causes. See also com-
mon causes and process control.
162 specification limits

Figure 42. Skewed distributions.

specification limits Thresholds that define acceptable ranges of the

product or service characteristics or attributes for identifying con-
formance. In the manufacturing environment, this is often synony-
mous with blueprint tolerance. They should always reflect engineer-
ing-defined, and ultimately customer-defined, requirements for a
process or product variable.

specifications Identified requirements to which a designated service

or product must conform. See also conformance.

spider chart A chart used to depict and communicate process per-

formance against that of a benchmarked company along multiple
process attributes. The performance of the benchmarked company
on each attribute is defined as a point on the perimeter of a polygon.
A scale emanates from the center of the polygon to each point. The
sponsors 163

current performance of the company is plotted along each scale, and

the points between baseline measurements are then connected,
showing the performance of the company against the company
benchmarked. Also called radar chart. See Figure 43.
sponsors Management leaders who are responsible for chartering
process improvement teams, acting as advocates for those teams, as-
sisting the teams in obtaining needed resources, assisting the teams
by clearing roadblocks, and providing rewards and recognition to
successful teams. It is not recommended that the sponsor be a mem-
ber of the team. See chartering a team. See also reward and recogni-
tion system.

Figure 43. Spider charthadar chart.

Customer
Satisfaction

Human Resource
Utilization

- The top company performance in that category

-
II Average performance of top companies in that category
Status of your company’s performance
164 sponsor’s role

sponsor’s role The responsibilities of the sponsor:

1. Charge the team with a clear mission of what is to be accom-

plished.
2. Tell the team members why the mission is important to the
operation of the business.
3. Select the initial team members and in many cases the team
leader.
4. Work with the team leader to clarify the mission.
5. Work with the team to answer questions, help get past road-
blocks, etc.
6. Provide resources to the team as necessary.
7. Help to remove organizational obstacles and barriers to team
success.

SQA See supplier quality assurance.

SQC See statistical quality control. See also Shewhart, Walter A.

SQI See supplier quality improvement.

stability A measure of the dependability or consistency of a process

over time. A process is stable if it is in control, as depicted on a process
control chart. See also out-of-control process.

stages of change As described by Svenson et al. in The Qmlity Road-

map (New York: AMACOM, 1994), there are four stages of change in
typical TQM implementations:

Stage I , getting started. Management begins to map the existing

enterprise and gather data about process performance.
Stage 2, targeted improvement. Initial forays are made by way of
pilot improvement projects.
Stage 3, extension to all processes. Improvements are spread across
all potential applications within the company.
Stage 4, continuous improvement. Continuous process improve-
ment becomes a way of life at the company. (See continuous
improvement.)
statistical methods 165

stakeholders All those inside and outside the company who have a
significant interest in company performance. See shareholders vs.
stakeholders.

standard deviation A measure of variability that describes the

spread of scores in a data set around the mean. Technically, the pop-
ulation standard deviation can be defined as the square root of the
arithmetic average of the squares of all deviations from the mean in
any frequency distribution. u = population standard deviation; s =
sample standard deviation.

standard deviation chart See s-chart.

standard operating procedure (SOP) A written document that de-

scribes the methods and procedures a worker uses to run an opera-
tion. Manufacturing SOPs typically include descriptions of the pro-
cess along with the procedures that outline the following: power-up,
setup, shutdown, operation, process control, adjustment, documen-
tation, housekeeping, routine maintenance, simple troubleshooting,
and diagnosis. Simple diagrams, charts, check sheets, flow diagrams,
and visual aids may be a part of the SOP. Note: There are also SOPs
for nonmanufacturing processes (e.g., an SOP for filing a medical
claim) that do not outline power-up, maintenance, etc.

standardization Usually means implementing a standard operating

procedure; could also mean converting a distribution to a standard-
ized normal distribution (Z scores).

standardize-do-check-act cycle See SDCA cycle.

standardized control charts Control charts on which the observed

values are translated into standardized values that represent the
number of standard errors between the observed value and the pro-
cess average.

statistical methods Methods used to describe data and to identify

levels of variation and their causes. Statistical methods may be di-
vided into basic and advanced statistical techniques. Basic statistical
methods include the graphic representation of data, basic descrip-
tive statistics, control chart theory, process control and capability
166 statistical process control (SPC)

procedures for variables and attribute data, and underlying distribu-

tional analysis. Advanced statistical methods encompass statistical in-
ferences, significance tests, and design and analysis of experiments.
Beyond the techniques and methods listed above, the fields of
measurement, reliability, supplier quality assurance, procurement,
testing, and others have their own applications and tools for collect-
ing, analyzing, and interpreting statistical data in support of decision
making in day-to-day business operations. See also data collection
and seven basic tools of quality.

statistical process control (SPC) The application of statistical meth-

ods to identify and reduce process variation. Typical SPC applications
use SPC charts to track process performance in light of control lim-
its. The use of these charts allows operators to monitor the perform-
ance of their processes in order to detect changes (out-of-control con-
ditions) that may lead to unacceptable conditions or failure to capture
improved conditions if left unchecked or process improvements that
should be standardized. SPC charts are also used to determine when
to make adjustments in order to continually reduce process variabil-
ity toward nominal values. See also out-of-control process and de-
tection vs. prevention.

statistical quality control (SQC) A discipline that involves applying

basic and advanced statistical methods and tools to process-related
data in order to identify the critical process variables or root causes
and to systematically reduce variation or eliminate problems. The
objective of SQC is to ensure that all customer needs are satisfied
through stable, constantly improving processes in the absence of de-
fects. The basic tools of SQC are listed under seven basic tools of
quality. These tools, as well as advanced methods, are explained in
more detail in other portions of this text.

statistics, descriptive See descriptive statistics.

steering committee See discussion under planning phase (phase 11).

stock, buffer See buffer inventory/stock.

strategic alliances Alliances between manufacturers and their key

suppliers that provide leverage in gaining access to foreign markets
strategic product-market analysis 167

and offer the advantage to those suppliers of local labor content in

the end item manufactured. See also risk sharing.

strategic goals See goals, strategic.

strategic intent A broadly based goal required to meet a competitive

threat or achieve company goals. These intents result from long-term
strategic analyses and plans. For example, “Achieve perferred sup-
plier status with all critical customers” might constitute a strategic
intent.

strategic objective A short-term, specific outcome required to

achieve a strategic intent. Usually, multiple strategic objectives must
be achieved to meet a given linked intent. Further, any given set of
strategic objectives must be collectively exhaustive and additive, as
related to the strategic intent to which they are linked. That is, if all
strategic objectives are met, it must be implicitly correct that the in-
tent will be met.
Strategic objectives of limited scope and term, with metrics and
time-based milestones, are called focal points. See also focal point
chart.

strategic plan A fundamental element of policy deployment, the

strategic plan describes in broad terms the strategic emphases or
areas of focus by which the organization intends to achieve:

1. Continuously improving levels of customer satisfaction

2. Continuously improving competitive position and profit-
ability

The strategic plan typically encompasses a three- to five-year pe-

riod and in TQM environments is developed in the planning phase
(phase 11) of the T Q M implementation. It provides several highest-
level long-term milestones against which company performance and
enterprise-level progress may be monitored.

strategic product-market analysis An analysis that provides the basis

for strategic and operating planning activity. The specific structure
of the analysis varies from company to company, but the content gen-
erally includes:
168 subgroup

1. Current practices and processes. An assessment of the status of

key critical process characteristics as they currently exist, a
forecast of needs and demands of the current customer base
compared to anticipated future demands, and a technical
benchmarking comparison to current and potential competi-
tors. (See also competitor influences.)
2. Internal and external forces analysis. An analysis of all significant
internal and external forces affecting the products and ser-
vices of the company, with associated forecasts.
3. Growth/penetration assessment. An assessment of the potential
for growth and penetration of new markets for core compe-
tencies, products, and services.

The strategic product-market analysis should be performed late

in the preparation phase (phase I) or early in the planning phase
(phase 11) of the T Q M implementation, before the strategic planning
activity.

subgroup A set of units or observations obtained by subdividing a

larger group of units or observations.

suggestion systems A means by which employees (and stakeholders)

can offer their ideas. The purpose of an employee suggestion system
is to provide an opportunity for individuals with the company to con-
tribute to achievement of the company’s strategic goals. (See goals,
strategic.) It resides in the employee involvement element of the
T Q M model, is focused on the employee, and is generally adminis-
tered by first-line supervision.
The suggestion program within the T Q M model is unlike stand-
alone suggestion programs in that it is focused on specific topics for
improvement that have been identified as a part of policy deploy-
ment, and it operates within the department’s span of control, mak-
ing it easier to administer. In addition, TQM-based suggestion pro-
grams support suggester-led evaluations of the suggestion, and
orient the suggestions toward specific results identified as ways to
improve overall processes and systems. Those offering suggestions
play an active role in implementing their suggestions.

supplier A person, group, or organization (internal or external)

from whom goods or services are obtained.
supplier quality improvement 169

supplier cost of quality accounting systems One component of the

supplier quality assurance system model, which provides the de-
tailed data required to analyze the total cost (as opposed to price) of
products from suppliers. See also cost of quality and total cost con-
cept in purchasing.

supplier quality assurance Confidence that a supplier’s product or

service will meet its customers’ requirements. A data-based relation-
ship is developed between the customer and the supplier that en-
sures that products and services delivered satisfy (and eventually ex-
ceed) customer needs. A component of the total quality assurance
part of the T Q M model, which includes supplier selection and quali-
fication, cost of quality, and specification approval and review sys-
tem. See Figure 44. See also customer satisfaction.

supplier quality improvement As an element of T Q M , supplier qual-

ity improvement programs should involve three major elements:

Figure 44. A suggested model for a supplier quality assurance

system.

Supplier cost of
Selection and Quality
Qualification Accounting
System

Critical Characteristic
Approval and Review
System
170 supplier selection and qualijcation

1. Including key suppliers in the design review process

2. Requiring suppliers to adopt a management philosophy of
continuous improvement, founded in T Q M principles
3. Inviting key suppliers who have adopted a T Q M approach to
become risk-sharing partners in new development projects
where they are significant stakeholders
supplier selection and qualification One component of the supplier
quality assurance system model, dedicated to the analysis and selec-
tion of supplier, faciliation and improvement responsibilities, and
purchased productkomponent analysis. Among the elements con-
tained in the supplier selection and qualification system are:
1. The ongoing supplier review system (including on-site sup-
plier support and assistance, and on-site SQA performance
evaluations)
2. The internal supplier assessment and qualification system (in-
cluding an assessment of price and delivery performance, as
well as an evaluation of the supplier’s product quality as it
relates to downstream process variation)
support processes Processes that support the primary value-adding
processes of the company. These processes include human resources,
accounting, and information technology management. See also
value-added analysis.
suruey-feedback process The process developed and used by a com-
pany to continuously receive feedback from its customers and sup-
pliers regarding its performance. (See also feedback, customer.)
This process is an important source of information in the first two
phases of a T Q M implementation and is also critical to the continu-
ous improvement effort contained in the last phase of T Q M imple-
mentation. The process is designed differently in virtually every com-
pany, but the most important aspects are the same: the process must
yield accurate, usable, and timely information about the perceptions
of customers and suppliers concerning the quality of the company.
synchronousproduction An approach used in just-in-time manufac-
turing that synchronizes the rate of production from end item deliv-
ery all the way back through supplier production to customer de-
mand. In other words, products are built at the same rate at which
they are sold. See also uniform work loading.
tactical metrics Measurements that are tracked to maintain critical tacti-
cal items-hose things that must be maintained while working on strate-
gic objectives and focal point items. For example, if reducing delivery time
is a strategic objective, a critical tactical objective may be to maintain inven-
tory. Hence, inventory would be a critical item and h y s inventgr may be a
critical tactical metric.

Taguchi, Genichi A Japanese quality leader whose work in the 1950s

popularized the idea that poor quahty results in a loss to society. He is
also h o u s for his work with orthogonal arrays and hctional hctorial
experiments.

Taguchi loss function A parabolic representation that estimates the qual-

ity loss, expressed monetarily, that results when quality characteristics
(satisfiers) deviate from target values.
The cost of this deviation increases quadratically as the characteristic
moves M e r fi-om the target value. This model is named for Genichi
Taguchi,who won the Deming Prize in 1961 for his paper based on this
topic. See Figure 45.

Figure 45. Taguchi loss function.

LSL target USL

171
172 Taguchi quality loss function

T w h i qudity loss function See Taguchi loss function.

tdy sheet See check sheet.

targetpopulation See population.

target value The value given to a process owner to “hit” in ongoing

production and control. If possible, the target will be equivalent to the
nominal value for the product or process characteristic. Where nominal
is closer to one specification than another, the target may be different
from the nominal value to avoid the production of nonconforming units.

team, chartering a See chartering a team.

team, cross-functional See cross-functional team.

team, steering S e e steering teams.

team member’s role The responsibilities of the team member:

1. Help ensure balanced participation among all team members.

2. Attend team meetings.
3. Identdj opportunities for improvement of team interaction.
4. Carry out action items (assigned tasksjresponsibilities).
5. C w individual understanding of the team mission.
6. Follow the team’s ground rules.
7. Learn the structured process the team uses.
8. Recognize and communicate the need for training.
9. Ensure closure on items.
10. Share responsibility for team members’ performance.

teams Groups of people formed for the specific purpose of improving

quality or solving problems. The teams may be natural work p u p s or
cross-functional in membership, depending on the nature of the subject
matter they are formed to address. See sponsors;facilitator;leader, team;
and team member‘s role. See also cross-functional team and problem
solving.

technique e r r o ~ S e e operator-controllable error.

tool wear I 73

techmhgy migration paths The most likely paths of migration of a new

technology through an organization. These paths are predominantly
downstream through the internal suppliercustomer structure, but may
vary depending on several other hctors, as described in Munufatwing
2000 (New York, AMACOM, 1994.)

telecommuting Performing work from a remote location via telecommu-

nications and/or computer equipment. Typically, this involves the use of
personal computers, telephone lines, and a telecommunications device
(modem)on each end that translates the computer’s language into trans-
mittable signals and back again.

TEQA See European Quality Award.

testing See destructive testing and nondestructive testing.

Theory X A theory of behavior that suggests and acts on the belief that
people are basically lazy,find work repulsive, and prefer to be directed in
their work activities.

Theory Y A theory of behavior that hypothesizes and acts on the belief

that people are self-motivated,enjoy work, and prefer to think for them-
selves, remaining at least semi-autonomous in their work activity.

Theory 2 A theory of behavior that emphasizes fBir treatment of custom-

ers, production of quality products, and other traitdvalues that would
generally be regarded hvorably.

third+* registration Certification by an independent registrar as com-

pliant with one or more I s 0 9000 standards. See registration, Is0 9000.

throughput time See cycle time.

tolerance The maximum acceptable deviation from specified target

value. The amount of tolerance is listed in the specification or on the
blueprint for a quality characteristic.

tool wear A condition under which the variability of the process at any
one point in time will be less than the distribution over the life of the tool.
174 tooling reliability

toding reliubdiiy The probability that a given tool will produce parts
within control limits.Poor tooling reliability is usually symptomatic of a
more hndamental problem. It is important to look for the more hnda-
mental problem (e.g., machine vibration) and solve it in order to maintain
long-term process control.

total cost concept in purcMng The concept that the true cost of a pur-
chased item is the cost associated with procurement and use of the item.
See also supplier cost of quality accounting systems.

totalpredictive maintenurn See predictive maintenance.

total p.aluctive maintenance (TPM) A representation of overall equip-

ment effectiveness using mathematical models of losses (associated with the
manuhcturing process) due to equipment failure,setup and adjustment,
idling and minor stoppage, reduced speed, process defects, and reduced
yield.

totalquality assumnce (QA) One of three interlocking technologies sup-

porting the management subsystems of TQM. Customer driven, it sup-
ports daily management, the PDCA management discipline, and policy
deployment. It focuses on prevention rather than detection through de-
sign and process control, and may be subdivided into the components of
supplier quality assurance, statistical quality control,and customer qual-
ity assurance. See also detection vs. prevention, quality improvement
strategy, and problem solving.

totalQualily control(w) An approach to quality that focuses on assur-

ing the quality of all the inputs to a system, including materials, informa-
tion, labor, and equipment. TQC (also referred to as SQC) is often re-
garded as the subset of TQM from which the plandocheckact cycle
approach emerged.

total quuliiy management (QM) A term initially coined by the Naval Air
Systems Command to describe its Japanese-style management approach
to quality improvement. Since then, the term TQM has taken on many
meanings. Simply stated, TQM is a system by which continuous improve-
ment of all value-adding processes performed by the organization may be
achieved. The customer determines if value has been added based on his
or her satishction. (See customer satisfaction and valueadded analysis.)
training 175

TQM is predicated on the participation of each organization member in

improving products, processes, services, and the company culture. The
methods used in TQM have been developed from earlier work by quality
leaders such as W. Edwards Deming,Armand V. Feigenbaum, Kaoru
Ishikawa, and Joseph M. Juran.

total quality management system (QMS) The consistent, structured ap-

plication of management commitment, leadership, customer focus, total
participation, and systematic analysis to an organization’s processes in
order to achieve customer satidkction.

TPM See total productive maintenance.

TQA Seetotalqualityassurance.

QC See total quality control.

Q M See total quality management.

Q M S See total quality management system.

tracking,attributes See attributes tracking.

training An educational tool used to develop and improve skills, which
must be present for quality improvement. Training for TQM practitioners
is tailored to the needs of the individual, based on:

1. Roles. Management roles are very different from floor-level pro-

cess improvement team member roles, so their skills requirements
differ.
2. Timing.It is best to administer training when there is a clear and
present need for the training.
3. Costlavailabilzty. Both help determine the feasibility of training ini-
tiatives.

Among the most frequently included training for TQM practitioners

are the following topics:

0 The seven basic tools of quality

0 Statistical process control
176 trend

Benchmarking
0 Quality function deployment
0 Problem-solving strategy
0 Quality improvement strategy
Industrial statistics
0 Design of experiments
Team effectiveness
Basic T Q M awareness
0 New roles for T Q M leaders (managers and supervisors)

See also education and training (for total quality) and employee
selection and training.

trend A series of six or more measurements that represent progres-

sively increasing or progressively decreasing values. Nonrandom be-
havior. A trend indicates an out-of-control condition and the pres-
ence of an assignable (special) cause. See also special cause, process
control, and out-of-control process.

trivial many In the context of Pareto analysis, a reference to the

majority (roughly 80%) of occurrences in a population of occur-
rences representing 20% of the value of the population. @ran re-
ferred to this group as the important Many since they often, in fact,
hold value.) See also vital few.

true value The theoretically correct and actual value of the charac-
teristic being measured.

truncated distribution On a histogram, an asymmetrical shape with

the peak at or near one end of the data range, and with the data
range falling off abruptly at one end, while the other end of the dis-
tribution tapers off gently. Truncated distributions often result from
abnormal effects on the data gathered, such as screening the data by
employing 100% inspection and reworking the defective parts be-
fore their values are counted in the distribution. See Figure 46.

tweaking Action taken to compensate for common causes of varia-

tion, which often has the effect of increasing rather than decreasing
the variation.
Type ZZ error 177

Figure
- -
- -
-- -
-

Type Z error An incorrect decision to reject a hypothesis when it is

really true. See also alpha.

Qpe IZ error An incorrect decision to accept a hypothesis that is

really false. See also beta.
 U
u-chart A control chart based on the normal approximation to the Poi-
sson distribution, which depicts the defect count per inspection unit.

U s h a p d cells In cellular manuhcturing, work centers configured in

horizontal U shapes, with operators occupying positions within the U. This
configuration allows operators to more easily move between positions, ad-
justing the number of operators in a cell based on the work load present
at a given time. The U-shaped layout is illustrated under cellular manu-
Eacturing.

UCL See upper control limit.

unexplained variation S e e experimental error.

u n i f i i w k loading Matching the cycle times of individual operations

within the production routing, so that production occurs as if to a com-
mon drum beat. This is synonymous with synchronous production, ex-
cept that synchronous production goes even further, matching produc-
tion to end item delivery rates.

universe See population.

uppet. control limit (UCL) The upper boundary, drawn as a horizontal

dashed line on a control chart, indicating the value below which sample
(statistical)values are expected to filldue to common causes of variation.
A full and proper assessment of control requires consideration of process
performance with respect to the lower control limit and the laws of proba-
bility in addition to the upper control limit. See control limit and in-
control process.

178
USL 179

upper limit (UL) A term used by different practitioners to refer (ambigu-

ously) to upper specification limit, upper process limit, or upper control
limit.

upper process limit The threshold above which exactly .135% of the mea-
sures, scores, values, or observations fill.

upper specification hmit (USL) The threshold above which the character-
istic beiig measured is considered to be defective, because it no longer
conforms to specifications. See also conformance.

USL See upper specification limit.

value The result of the benefit one receives from a product or service
relative to what was expended to receive or acquire it. Generally value is
the ratio of benefit to cost, and more specifically,value is the ratio of qual-
ity to cost or quality to price. Cost in this case is not just the purchase price,
but refers to all of the costs the customer incurs for the privilege of using
a product or service. To increase value, one can increase the benefit to
the customer through increased quality or by more effectively meeting
expectations. Value is also increased by reducing the price or cost to the
customer. Those organizations able to both increase quality and decrease
(total)cost to the customer benefit greatly.

value, adding See adding value.

value, n o m i d See nominal value.

value, turget See target value.

value, true See true value.

v a l u e d d anaZysis An analysis performed on a process to determine

which elements of that process add value, and how much throughput
time and work-in-processresults from that non-value-added activity.

value&ingprocess An activity that transforms an input into a more

valuable output to be used by a customer. The customer may be part of
the organization performing the activity or may be an external customer.
See also adding value.

value anaEysislvalue t?ngiwering A structured technique developed by

Value Analysis, Incorporated, in 1963, which formally analyzes processes
and designs for the sole purpose of reducing cost and waste in those pro-
cesses and designs.

180
vision (company/corporate) 181

values, organirational See organizational values.

variability Changes in data,a characteristic, or a function caused by one

of four hctors: assignable or special causes, common causes, tweaking,
or seasonalllong-termtrends.

variable, degendentor response See dependent variable.

variable, process See critical process characteristics.

variubh data Quantifiable, measurable, continuous data. Variable-

based control charts include X-bar, R, median, X, s, and moving range
(RM).
variance A measure of dispersion that is equal to the squared standard
deviation.

variance, anaEysis of See analysis of variance.

variation, envinmment’s role See environment and Variation.

variation, unexplained See experimental error.

vm$cation, &sign See design verification.

uirtzud reality A particularly vivid form of computer simulation that gen-

erates the feeling that a person is physically at some remote location
through the use of devices that provide multiple sensory inputs (visual,
audio, tactile, etc.).

vision, deuelopment of The development of the vision and vision state

ment. This process should be performed by the company’s top-level man-
agement and/or owner@).Visions result from people answering tough
questions about exactly what business(es) they desire in five to ten years,
why, and generally how they plan to get there. They must envision their
size, location, and operating methods. They must base all of these ele-
ments on an even more fundamental and important question: What is the
purpose of this company?

z i r h (contpa~/cmfmra&) The company’s purpose for existence, the

standards and behaviors or guiding principles that the company exhibits,
182 vision statement

and the overall strategy the company intends to use to achieve its objec-
tives. See vision statement and vision, development of.

vision statemat A hndamental element of policy deployment. The vi-

sion statement describes the underlying premises on which the company
is based, and more importantly, describes in a compelling way where the
company is going to be at some hture point. It is critical to gaining shared
goals and objectives within the company, and to generating enthusiastic
participation in broad, sweeping changes such as T Q M implementation.
Impa&l vision statements create employee commitment, not just compli-
ance.

vitalfm A term emerging from the Pareto analysis, which refers to the
minority (roughly 2Wo) of any population, which represents the majority
(roughly 800/0)of the value of the population. See also trivial many.

voice of the customer See customer mapping pracess.

W M s See wide area networks.

waste Any expenditure of resources (including time) that does not add
value in terms of a customer-satisfjmg quality characteristic (satisfier).
See also valueadded analysis.

WeibuZZ distributim/phts Graphic representations of the Weibull fimdy

of distributions,described by three parameters: shape, scale, and location.
The shape parameter (13) establishes the pattern of the curve; the scale
parameter (a)is related to the peakedness of the curve; and the location
parameter (y) is the origin, or starting point, and has a dehult value of
zero. See Figure 47.

0.4 0.8 1.2 1.6 2.0

183
184 wide area networks (WANs)

wide area networks (WANss) Computer networks similar in function to

today’s local area networks (LANs),where multiple users are supported by
servers on a host computer, except that the scale is much greater. WANs
developed over the next several years will connect multiple LANs and
other remote hcitities with voice, data, and image transmission.

W P See work-in-process.

work content study A detailed study of the specific tasks performed in a

designated process. A work content study identifies all tasks as value-add-
ing or non-value-adding, identifies the time and resources required for
each task, and identifies the skills required for the performance of each
task. It is typically performed as part of a throughput time-reduction effort
in just-in-timemanuthcturing programs, but is also very effective in non-
manufacturing businesses. See also valueadded analysis.

work-inp.ocess W P s ) Any inventory against which direct labor has been

applied and has not yet been completed. One of three categories of inven-
tory: raw material, work-in-process,and finished goods.

WWU class As good as, or better than, anyone else in the business in
which a company is involved. “Good” in this context means “able to iden-
and satisfy the needs of the customer.”

world-class manufacturing As good as, or better than, any other com-

pany offering the same types of manuhctured products. See also world
class.
 R c h r t See median and range chart.

& RM c h r t Individuals and moving range chart. See X chart and &
Chart.

X-bar and R & R) c h r t See average and range chart.

X-bar and stundurd deuiaeion & s) c h r t s Variables control charts con-

taining the data that would normally appear on an X-barchart and a
standard deviation chart. These charts are used most often for high-vol-
ume processes and when large sample sizes are involved (n 2 8 per sub-
group).
X-bar chrts Sample average control charts.

X c h r t Individuals chart. Variables control chart that plots individual

measurements (vs. subgroups) for each time period.

185
ykZd Seerecovery.

186
2 score The number of standard deviations separating a given value
fiom the mean of a distribution. A standardized value used to transform
a continuous data set into a standard normal distribution (a normal distri-
bution with a mean of zero and a standard deviation of one).

zero defects A quality philosophy that advocates a focus on error-free

process performance, resulting in defect-free products and/or services.

187