Introduction To Operations Research
Introduction To Operations Research
Introduction To Operations Research
TO
OPERATIONS RESEARCH
Evaluation Scheme
Number of credits : 3
Component Assignment slot Marks
Individual Assignments End of sessions 10 and 15 15 + 15
and case analysis (group work) - 2
Decision making situations are not mutually exclusive and complexities arise in decision
making due to several factors,
• political, economic, social, technological ethical, limitation of resources, values system,
risk attitudes and knowledge of decision maker etc.
Approach to decision making
Inductive - Heuristic
• Rule of thumb
• Common sense
• Snap judgement
• Gut Feeling
Area of OR Application
National Planning and Budgeting
Defense Services
Industrial Establishment and Private Sector Units
R & D and Engineering
Business Enterprise
Operational Research Methodology
OR approach to problem solving consists of six steps
1. Formulate the problem
2. Determine the assumptions (model building) and formulate the problem in mathematical
framework
3. Acquire the input data
4. Solve the model formulated and interpret the results
5. Validate the model
6. Implement the solution obtained
1. Formulate the problem
The first step in OR is to develop a clear and concise statement of the problem.
Logically, we cannot expect to get a right answer if the problem is identified incorrectly
One problem may be related to other problem, therefore an analysis be made as to how the
solution to one problem is likely to affect the problem or situation in general
Usually, we cannot deal with all the problems and has to focus only on few of them, that
whose solutions are likely to result in greatest profit increase or least cost
once the problem has been identified, it is categorized as being standard or special problem
The standard problems are known as programmed problems
Programmed problems are well structured in their solution strategy, ex. assignment of
workers to jobs, fixing product mix,
Special problems are unique and non recurrent in nature
2. Model Building
Once the problem is defined, the next step is to build a suitable model.
A model is a theoretical abstract or simplified representation of a real life problem
Models can be classified as: physical models and symbolic models
a) Physical models – Is a physical or schematic representation of the real thing. There are two
types of physical models : Iconic and Analogue
• Iconic Models - Are scaled up or scaled down version of a particular thing that the represent.
ex. A model aeroplane in a wind tunnel, model sun and its planets in a planetarium, maps,
drawings,
• Analogue Models – The analogue models use one set of properties to represent another set.
ex, an electrical network model may be used as an analogue model to study the flows in
transportation system.
b) Symbolic Model – Many real life problems can be described by symbolic models or
mathematical forms.
• These are most general and abstract type of models and OR uses symbolic models
• Symbolic models can be verbal or mathematical. ex. perimeter: p = 2L+2W
2. Model Building (cont.)
• Mathematical models can be broadly classified as;
• Deterministic models – In which all parameters in the mathematical formulation are fixed
at predetermined values so that no uncertainty exists
a) Probabilistic models – Some or all the basic characteristics may be random variables
(capable of assuming different values with a given probabilities)
• Probabilistic models are also termed as stochastic or chance models.
• Mathematical models comprise three basic components :
1. Decision variables – Represents those factors where a choice could be made. Variables can
be manipulated and therefore are controllable by the decision maker.
2. The result Variables – Indicates level of effectiveness of a system, or out put of the system,
termed as dependent variables, are beyond the control of decision maker
3. Un-controllable variables – ex. profit function of a firm making two products can be stated
as:
P = p1x1+p2x2 , P = total profit, x1 and x2 are the number of units sold (independent
variables), p1 and p2 is the profit per unit of the product respectively (un-controllable
variable)
3. Obtain in-put data
Once an appropriate model has been formulated, the next step is to obtain the data to
be used in the model as in-put
• A number of sources, including company reports and documents, interviews with the
company personnel etc.
4. Solution of Model
Solution to a model implies determination of a specific set of decision variables that
would yield a desired level of output.
• A solution, which does not satisfy all the constraints. Since an infeasible solution fails to
to meet one or more system requirements, it is an unacceptable one.
• Only feasible solutions are of interest to the decision maker
It involves testing the structural assumptions of the model to ascertain their validity
Usually validity of the model is tested by comparing its performance with the past data
available in respect of the actual system.
6. Implementation
The final step is implementation of results.
The impact of decision may cut across various segments of the organization, and factors
like resistance to change, desired to be consulted and informed, motivation so on may
come in the way of implementation.
Skills and expertise is needed to tackle the implementation issues.
Tools / Techniques of Operational Research
1. Linear Programming
2. Transportation Problems
3. Assignment Problems
4. Decision Theory
5. Markov Chain
6. Queuing Theory
7. Simulation Models
8. Games Theory
9. Sequencing
10. Replacement Problems
11. CPM and PERT Analysis
12. Forecasting Techniques
1.Linear Programming (LP)
This technique is used to find a solution for optimizing a given objective.
• Objective may be maximizing profits or minimizing costs.
Objective function and boundary conditions are linear in nature.
LP techniques solve Product-Mix and Distribution problems of enterprise.
Its also used to allocate scarce resources in optimum manner in problems of scheduling,
product mix, etc.
2.Transportation Problems
Transportation problems deals with transportation of a product
From a number of sources to number of destinations
With limited supplies
With specified demands
At the total transportation cost.
The main objective of transportation is to schedule shipment from sources to
destinations in such a way so as to Minimize the Total Transportation Cost.
3. Assignment Problems
It is a special type of linear programming problem.
It deals in allocating various resources or items to various activities in a one to one basis in
such a way that the time or cost involved is minimized and the sale or profit is maximized.
e.g. Manager may like to know which job should be assigned to which person so that
all jobs can be completed in the shortest possible time.
4.Decision Theory
Decision Theory is primarily considered with decision making under the conditions of:
• Risk and
• Uncertainty
5. Queuing Theory
This theory deals with the situations in which queue is formed, e.g. customers waiting for
services, machines waiting for repairmen, and aircrafts waiting for landing strips, etc.
If the Queue will be long the cost will be high due to long waiting hour.
This technique is used to analyze the feasibility of adding facilities and to access the
amount and cost of waiting time.
This calculations can then be used to determine the desirable number of service facilities
6. Simulation Model
A simulation model is a mathematical model that calculates the impact of uncertain inputs
and decisions we make on outcomes that we care about, such as profit and loss,
investment returns, environmental consequences, and the like. Basic building blocks are,
a. State of the system (e.g No of customers in a queue)
b. Possible state that can occur
c. Possible events (arrival rate and service rate)
d. Simulation clock (records the passage of time)
e. State transition (generated by various kinds of events)
7. Markov Chain
Formally, a Markov chain is a probabilistic automaton. The probability distribution of state
transitions is typically represented as the Markov chain’s transition matrix. If the Markov
chain has N possible states, the matrix will be an N x N matrix, such that entry (i, j) is the
probability of transitioning from state i to state j. Additionally, the transition matrix must be
a stochastic matrix, a matrix whose entries in each row must add up to exactly 1
8. Game Theory
Game theory attempts to take into consideration the interactions between the
participants and their behavior to study the strategic decision-making between rational
individuals.
It tries to find out the actions that a “player” should perform which would maximize
his chances of success mathematically and logically.
Actions by everyone involved directly alter dynamics of the game, and hence the
players are all interdependent.
The games can be broadly classified into two categories: zero-sum and non-zero-sum.
In zero-sum games, the loss of one is gain of another.
9. Sequencing
Models have been developed to find a sequence for Processing Jobs so that the total
elapsed time for all the jobs will be minimum.
The models also help to resolve the conflict between the objectives of maximizing
machines utilization and complying with predetermined delivering rates.
10. Replacement Problems
This Theory is concerned with situations that arise when some items such as machines,
men, etc. require replacement due to their decreasing efficiency, failure or break-down.
• Sooner or later all the equipment are required to be replaced because of:
• Obsolescence
• Discovery of New Technology
• Better Design of Equipment
Similarly, a manager fails to understand the complex working of OR. Thus there is a gap
between the two.
• Management itself may offer a lot of resistance due to conventional thinking.
Moreover, a fairly good solution at present may be more desirable than a perfect OR
solution available after sometime.
The computational time increases depending upon the size of the problem and accuracy
of results desired.
Next Topic Linear Programming