Yashwantrao

Chavan
Maharashtra
Open University

Operations Research
Author: Arti Kale
Prof. Ram Thakar

UNIT 1: Introduction to Operations Research

UNIT 2: Linear Programming
UNIT 3: Transportation Problems
UNIT 4: Assignment Problems
UNIT 5: Network Analysis
UNIT 6: Critical Path Method (CPM)
UNIT 7: Inventory Control
UNIT 8: Advanced Linear Programming Techniques
UNIT 9: Advanced Network Analysis Techniques
UNIT 10: Dynamic Programming
UNIT 11: Game Theory
UNIT 12: Queuing Theory
UNIT 13: Simulation Modeling
UNIT 14: Integer Programming
UNIT 15: Non-Linear Programming
UNIT 16: Decision Analysis and Multi-Criteria Decision-making

1
BLOCK I: FOUNDATIONS AND LINEAR PROGRAMMING

UNIT 1: Introduction to Operations Research

1.1 Meaning of Operations Research
1.2 Evolution of Operations Research
1.3 Approaches to Operations Research
1.4 Techniques and Scope of Operations Research
1.5 Managerial Applications of Operations Research
1.6 Role of Operations Research in Decision-making

UNIT 2: Linear Programming

2.1 Introduction to Linear Programming
2.2 Meaning and Characteristics of Linear Programming
2.3 Graphical Approaches and Their Utility
2.4 Simplex Method
2.5 Dual Linear Programming
2.6 Sensitivity Analysis in Linear Programming

BLOCK II: TRANSPORTATION, ASSIGNMENT, AND NETWORK

ANALYSIS

UNIT 3: Transportation Problems

3.1 General Structure of Transportation Problems
3.2 Methods of Initial Allocation
3.3 Degeneracy in Transportation Problems
3.4 Finding Optimal Solutions
3.5 Applications of Transportation Models in Business

UNIT 4: Assignment Problems

4.1 Structure of Assignment Problems
4.2 Methods to Solve Assignment Problems
4.3 Variations in Assignment Problems
4.4 Hungarian Method for Assignment Problems

UNIT 5: Network Analysis

5.1 Introduction to PERT/CPM
5.2 Background and Development of PERT/CPM
5.3 Stages in Application of PERT
5.4 Networking Analysis with PERT
5.5 Comparison of PERT and CPM
UNIT 6: Critical Path Method (CPM)
6.1 Introduction to CPM
6.2 Determination of Critical Path

2
6.3 Determination of Earliest Expected and Latest Allowable Times
6.4 Applications of CPM in Project Management

BLOCK III: INVENTORY, ADVANCED TECHNIQUES, AND

PROGRAMMING

UNIT 7: Inventory Control

7.1 Classification of Inventory Control
7.2 EOQ (Economic Order Quantity) Model
7.3 Inventory Control Systems
7.4 ABC Analysis
7.5 Advantages of EOQ Model in Management
7.6 Just-In-Time (JIT) Inventory Systems

UNIT 8: Advanced Linear Programming Techniques

8.1 Advanced Simplex Method Techniques
8.2 Sensitivity Analysis in Linear Programming
8.3 Post-Optimality Analysis
8.4 Dual Simplex Method

UNIT 9: Advanced Network Analysis Techniques

9.1 Advanced PERT Techniques
9.2 Crash Time and Cost Analysis
9.3 Resource Allocation and Leveling
9.4 Applications of Advanced Network Analysis in Complex Projects

UNIT 10: Dynamic Programming

10.1 Introduction to Dynamic Programming
10.2 Applications of Dynamic Programming
10.3 Solution Techniques
10.4 Real-World Applications of Dynamic Programming

BLOCK IV: GAME THEORY AND QUEUING THEORY

UNIT 11: Game Theory

11.1 Introduction to Game Theory
11.2 Types of Games
11.3 Solution Concepts in Game Theory
11.4 Applications of Game Theory in Management
11.5 Nash Equilibrium and its Applications

UNIT 12: Queuing Theory

12.1 Introduction to Queuing Theory
12.2 Characteristics of Queuing Systems

3
12.3 Models and Applications of Queuing Theory
12.4 Performance Measures in Queuing Systems
12.5 Applications of Queuing Theory in Service Industries

BLOCK V: ADVANCED TOPICS IN OPERATIONS RESEARCH

UNIT 13: Simulation Modeling

13.1 Introduction to Simulation
13.2 Monte Carlo Simulation
13.3 Applications of Simulation in Operations Research
13.4 Advantages and Limitations of Simulation
13.5 Discrete Event Simulation

UNIT 14: Integer Programming

14.1 Introduction to Integer Programming
14.2 Formulation of Integer Programming Problems
14.3 Solution Methods for Integer Programming
14.4 Applications of Integer Programming
14.5 Branch and Bound Method

UNIT 15: Non-Linear Programming

15.1 Introduction to Non-Linear Programming
15.2 Formulation and Applications of Non-Linear Programming
15.3 Solution Techniques for Non-Linear Programming
15.4 Case Studies in Non-Linear Programming
15.5 Convex and Non-Convex Optimization

UNIT 16: Decision Analysis and Multi-Criteria Decision-making

16.1 Introduction to Decision Analysis
16.2 Decision Trees and Payoff Tables
16.3 Multi-Criteria Decision-making (MCDM) Techniques
16.4 Applications of MCDM in Operations Research
16.5 Analytical Hierarchy Process (AHP) and Technique for Order Preference by
Similarity to Ideal Solution (TOPSIS)

4
 Unit 1: Introduction to Operations Research

Learning Outcomes:
 Students will be able to define operations research and describe its significance.
 Students will be able to summarise the historical development of operations
research.
 Students will be able to differentiate between various approaches to operations
research.
 Students will be able to apply operations research techniques to managerial
decision-making processes.
 Students will be able to evaluate operations research's role in improving
organisational decision-making.

Structure:
1.1 Meaning of Operations Research
1.2 Evolution of Operations Research
1.3 Approaches to Operations Research
 Knowledge Check 1
 Outcome-Based Activity 1
1.4 Techniques and Scope of Operations Research
1.5 Managerial Applications of Operations Research
1.6 Role of Operations Research in Decision-making
 Knowledge Check 2
 Outcome-Based Activity 2
1.7 Summary
1.8 Keywords
1.9 Self-Assessment Questions
1.10 References / Reference Reading

5
1.1 Meaning of Operations Research
Operations Research (OR) is a field concerned with using methods to improve our
decision-making processes. It uses different mathematics, statistical models, and
techniques to uncover the patterns and tendencies.
Operations Research can be described as offering a structured and even mathematical
decision-making model, which is essential in today’s world, where data controls
everything. Operations Research was born during World War II when military
commanders and other leaders tried to develop scientific solutions that would enhance
the general communication, transport and planning processes in their military
strategies.
Today, it has emerged as an important tool in various fields, such as the industry at
large, commerce, engineering, health care, and civil services, for efficiency and
economic improvement. Operations research is the appropriate way of setting up and
solving a problem to acquire the best solution for mathematics.

1.2 Evolution of Operations Research

Operations Research has its roots in as early as the first half of the 20th century. It is
the process that took place in the course and after the Second World War. Here is a
chronological overview:
 Early Beginnings: Operations Research can be traced back to its roots in the
early 20th century, and it had its share of management gurus, such as Frederick
Taylor, whose work delayed the application of scientific principles in
management.
 World War II: Operations Research found its way mainly during the Second
World War, where its most important advances emerged. Military operations
required optimization of resource use and thus created operational research teams
that involved mathematicians, scientists, and engineers, which further shaped OR.
 Post-War Development: From the war onwards, many techniques were applied
to industrial planning, hence the formation of OR societies and journals.
 Technological Advancements: After the introduction of computers in the 1950s
and 1960s, the use of OR became more efficient, and it was now possible to solve
some of the problems through simulations and optimization.

6
 Modern Day: Today, OR is present in many enterprises and serves as one of the
tools for managing decision-making in such sectors as logistics, SCM, finance,
healthcare, etc. The upcoming and current advancement of big data and analytics
in OR further increases its range of uses and fields of practice.

1.3 Approaches to Operations Research

Overview of Operations Research can be done in more ways based on the type of the
problem and the available data. The main approaches include:
 Descriptive OR: This can be explained as a process of defining and creating an
understanding of the context of the system at the current time. This involves
frequently gathering data, studying it, and trying to find a pattern or trend. Some
examples are statistics and data mining. Statistics Data Mining Techniques like
statistical analysis and data mining are commonly used.
 Predictive OR: It is the process of analysis of past events as a way of predicting
events in the future. Methods like regression analysis, forecasting, and simulation
models are helpful in outcome prediction and evaluating the effectiveness of the
varied aspects.
 Prescriptive OR: This is the most advanced level of problem-solving, where one
has to find the optimal solution to a certain problem. It is an optimization method
that includes linear programming techniques, integer programming techniques,
and network models to give operational strategies.
 Hybrid Approaches: Descriptive, predictive, and prescriptive approaches can be
integrated to tackle complicated issues more effectively.

 Knowledge Check 1
Fill in the Blanks.
1. Operations Research employs various __________ methods to help make
better decisions. (mathematical)
2. The roots of Operations Research can be traced back to the early ________
century. (20th)
3. During World War II, OR teams included mathematicians, scientists, and
_________. (engineers)

7
 4. Predictive OR uses historical data to make ________ about future events.
(predictions)

 Outcome-Based Activity 1
List one example of a real-world problem that could be solved using Operations
Research techniques.

1.4 Techniques and Scope of Operations Research

Operations Research utilizes a variety of techniques to solve problems. The scope of
OR is vast, encompassing numerous methods and applications. Some key techniques
include:
 Linear Programming (LP): LP is used to optimize a linear objective function
subject to linear constraints. It is widely used in resource allocation, production
planning, and transportation problems.
 Integer Programming (IP): Similar to LP, but the solution variables must be
integers. It is used in situations where discrete decisions are needed, such as
scheduling and layout planning.
 Simulation: This involves creating a model of a real-world system and
conducting experiments on it to understand behaviour and evaluate the impact of
different strategies. It is useful for complex systems where analytical solutions are
difficult.
 Network Models: These include techniques like the shortest path problem,
maximum flow problem, and project scheduling using PERT/CPM. They are used
in transportation, logistics, and project management.
 Queuing Theory: This is used to study waiting lines and optimize service
processes. It is applicable in areas like customer service, manufacturing, and
telecommunications.
 Game Theory: This includes analyzing relationships and games between different
decision-making entities. It can be applied to economic sciences, business
management, and even warfare strategies.
 Decision Analysis: This involves decision-making with respect to risks involving
two or more parties. Methods are decision trees and utility theory. The scope of
OR is broad and can be applied to various domains, such as:

8
 o Manufacturing: They include the efficient coordination of production
calendars, supply chain, and surge protection.
o Transportation and Logistics: For organizations engaging in delivery
services, factors such as route optimization, fleet management, and supply
chain design.
o Healthcare: Improving patient flow, resource allocation, and scheduling of
medical staff.
o Finance: Portfolio optimization, risk management, and pricing strategies.
o Public Services: Optimizing emergency response, urban planning, and policy
analysis.

1.5 Managerial Applications of Operations Research

Operations research is useful in offering managers some analytical tools and
techniques for problem-solving. The various functions are:
 Production and Operations Management: OR techniques play an important
role in the process of maximizing production, minimizing costs and introducing
advancements. Some of the categories include production planning and control,
inventory management and control, and quality assurance.
 Supply Chain Management: OR is used to design and manage supply chains,
ensuring the right products are delivered at the right time and cost. Techniques
include transportation models, network design, and demand forecasting.
 Marketing and Sales: OR is useful in the process of market segmentation, the
development of appropriate pricing models and sales volume prediction. It helps
the non-linear partitioning of customers and marketing communication into high-
profit segments.
 Human Resource Management: The OR techniques are used in the areas of
workforce planning, scheduling and performance appraisal. It is very essential to
determine the right number of employees to enlist and increase their efficiency.
 Finance and Budgeting: Risk analysis and assessment, investment planning,
financial analysis, budgeting, and portfolios are other applications of OR. Staking
principles such as linear programming and simulation help in decision-making on
investment.

9
 Project Management: OR has resources that can help manage time for projects,
resources for projects, and risks that are likely to occur with the projects. Risk
management tools like PERT/CPM and critical path analysis assist in the timely
and cost-effective accomplishment of projects.

Real-World Example:
OR can be used by a retail firm to determine the right inventory levels needed at
different outlets in a retail chain. Using linear programming, one can find the right
combination of quantities to order for each store and reduce the inventory cost, as
well as be in a position to ensure the company replenishes stock frequently enough to
meet the demand of each store. This results in an elevated use of resources and
improved customer satisfaction.

1.6 Role of Operations Research in Decision-making

Decision-making is one of the few areas in which operations research can make a
significant contribution. It offers a neat and easy-to-measure way of bringing
solutions to numerous examples and formulating improved decisions. Here are some
key roles of OR in decision-making:
 Problem Identification and Structuring: OR contributes to defining the actual
problem to be solved with the help of the outlined methodology. It extends the
concept of problem formulation and is defined as the process of understanding the
goals, boundedness, and important factors that are in place.
 Model Building: OR involves constructing mathematical models that represent
the real-world problem. These models help analyze the problem and explore
different scenarios.
 Solution Generation: Depending on the chosen method of OR, either the global
or the near-global solution is obtained. This includes techniques such as
optimization, simulation, decision analysis and other related techniques.
 Evaluation of Alternatives: Decision-making is dealt with by OR in the sense
that the latter gives instruments for assessing various decision options. It focuses
on the evaluation of the consequences of the strategies provided and the
measurement of the effects of the strategy on the goals.

10
 Implementation and Monitoring: Monitoring and control are also supported by
OR to ensure that the selected solution delivers its intended performance. It
involves monitoring organizational metrics and finding out the cause of their
performance, either positively or negatively.
 Continuous Improvement: OR can be used to support ongoing improvement of a
system providing feedback and information. It works in the capacity of a
diagnostic tool to pinpoint areas that might require enhancement and enhance the
process on a continuous basis.

Real-World Example:
One application of OR is in an airline booking system where the best flight schedules
and the most appropriate crew to attend to passengers' needs are determined. In this
way, the use of mathematical methods such as integer programming will help the
airline to create effective schedules with low costs and more aircraft usage. It results
in better performance and customer satisfaction, which are always the goals of any
organization.

 Knowledge Check 2
State True or False.
1. Linear Programming is used to optimize non-linear objective functions.
(False)
2. Simulation involves creating a model of a real-world system and conducting
experiments on it. (True)
3. Operations Research is rarely used in finance and budgeting. (False)
4. Decision analysis helps evaluate and compare different decision alternatives
under uncertainty. (True)

 Outcome-Based Activity 2
Identify a decision-making scenario in your daily life where you could apply
Operations Research techniques.

11
1.6 Summary
 OR allows a systematic and quantitative approach that is essential, especially in
present-day decision-making processes. It was originally established during World
War II and has since become popular in business engineering, health and public
services.
 OR allows the field a systematic and quantitative approach in a manner that is
essential in present-day decision-making processes. Originally established during
World War II, it has since become popular in business engineering, health and
public services.
 The evolution of OR could be dated back to the early part of the twentieth century,
and there was more progress made during the Second World War when it was
applied to enhance warfare and supply chain management. After the war, it was
extended to the Industrial applications for the formation of societies and journals
related to OR.
 Key OR techniques include linear programming for optimizing linear objective
functions, integer programming for discrete decision-making, and simulation for
modelling real-world systems. Network models and queuing theory are also
widely used.
 OR is used in the marketing and sales areas for marketing segmentation and
pricing strategies in the human resource management functions that involve
workforce planning and in the finance function for financial control and risk
assessment. The other area where OR is useful is during the scheduling of the
project and the identification of resources that will be required.
 OR helps to make better decisions or brings more structure and a quantitative
decision-making method to tackle many issues. It can be used in identifying
problems, analyzing problems coming up with models, finding solutions, and
comparing solutions.

1.7 Keywords
 Operations Research (OR): A discipline that uses advanced analytical methods
to help make better decisions.
 Linear Programming (LP): A mathematical technique for optimizing a linear
objective function subject to linear constraints.

12
 Simulation: A method that involves creating a model of a real-world system to
conduct experiments and understand its behaviour.
 Decision Analysis: Techniques used to evaluate and compare different decision
alternatives under uncertainty.
 Optimization: The process of making something as effective or functional as
possible.

1.8 Self-Assessment Questions

1. Define Operations Research and discuss its significance in modern decision-
making.
2. Trace the historical development and evolution of Operations Research.
3. Compare different approaches to Operations Research and provide examples of
their applications.
4. Explain the key techniques used in Operations Research and their scope.
5. Discuss how Operations Research can be applied in managerial decision-making
with relevant examples.

1.9 References / Reference Reading

 Kapoor, V. K., and S. D. Sharma. Operations Research: Theory and Applications.
Sultan Chand & Sons, 2022.
 Taha, Hamdy A. Operations Research: An Introduction. 10th ed., Pearson
Education, 2017.
 Kanti Swarup, P. K. Gupta, and Man Mohan. Operations Research. Sultan Chand
& Sons, 2020.
 Hillier, Frederick S., and Gerald J. Lieberman. Introduction to Operations
Research. 10th ed., McGraw-Hill Education, 2020.
 Ravindran, A. Operations Research and Management Science Handbook. CRC
Press, 2016.

13
 Unit 2: Linear Programming

Learning Outcomes:
 Students will be able to understand the basic principles of linear programming.
 Students will be able to identify the characteristics of linear programming models.
 Students will be able to apply graphical methods to solve linear programming
problems.
 Students will be able to analyze dual linear programming and its applications.
 Students will be able to conduct sensitivity analysis in linear programming.

Structure:
2.1 Introduction to Linear Programming
2.2 Meaning and Characteristics of Linear Programming
2.3 Graphical Approaches and Their Utility
 Knowledge Check 1
 Outcome-Based Activity 1
2.4 Simplex Method
2.5 Dual Linear Programming
2.6 Sensitivity Analysis in Linear Programming
 Knowledge Check 2
 Outcome-Based Activity 2
2.7 Summary
2.8 Keywords
2.9 Self-Assessment Questions
2.10 References / Reference Reading

14
2.1 Introduction to Linear Programming
Linear programming (LP) can be described as an analytical tool whose aim is to assist
in the decision-making process that seeks to arrive at the optimal solution under
specific conditions of a particular target function. It is used in areas like business,
economic analysis, engineering, and military practices. Linear programming aims to
obtain an optimal solution that denotes the highest profit or, vice versa, the lowest
cost while meeting the restrictions.
Definition
Linear programming is defined as the way to find the best result in a mathematical
model with orthogonal requirement vectors, which correspond to linear forms. Linear
programming is the problem of finding the minimum of a linear objective function on
the condition that it satisfies linear equality and inequality constraints
Historical Background
The historical background of linear programming can be traced back to the period
between 1940 and 1950, which was the Second World War. The military initially used
it to solve complex resource allocation problems. The method was formalized by
George Dantzig in 1947 with the introduction of the Simplex algorithm, which
remains one of the most widely used methods for solving linear programming
problems.
Applications of Linear Programming
Linear programming is used in various domains:
 Business and Economics: For optimizing resource allocation, production
schedules, and transportation logistics.
 Engineering: In areas such as network design, system optimization, and
energy management.
 Military: For planning and logistics, including the optimal deployment of
resources and personnel.

2.2 Meaning and Characteristics of Linear Programming

Meaning
The goal of optimization involves using mathematical techniques like linear
programming to maximize profit or minimize expenses while working within
predetermined restrictions. The objective function and these restrictions are both
linear.

15
Characteristics
1. Linear Objective Function: The function that needs to be maximized or
minimized is linear. For example, maximizing profit or minimizing cost,
represented as Z = c1x1 + c2x2 +…+ cnxn.
2. Linear Constraints: The restrictions or limitations on the decision variables
are expressed in the form of linear inequalities or equations. For example, a1x1
+ a2x2 + … +anxn ≤ b.
3. Non-negativity Restriction: The decision variables cannot be negative, i.e.,
x1, x2,…, xn ≥ 0.
4. Feasible Region: The set in which all the feasible solutions that meet the
constraints have formed a region called the feasible region.
5. Optimal Solution: A point within the constraint boundary that gives the
highest or the lowest score depending on what one is looking for in the
objective function.

2.3 Graphical Approaches and Their Utility

The best strategy for obtaining solutions in two decision variables while solving linear
programming problems is to use the graphical method. It also makes it easier to
determine the ideal solution to the problem and identify the viable region.
Steps in the Graphical Method
1. Formulate the Problem: Determine the objective with its function or criteria
and the constraints.
2. Graph the Constraints: For these constraints, plot them so that you will be
able to identify the feasible region on the coordinate.
3. Identify the Feasible Region: Find out the region that contains all the
constraints, showing all feasible solutions.
4. Plot the Objective Function: Produce the objective function line and shift it
vertically to identify the best point in the feasible region.
5. Determine the Optimal Solution: This will mean trying to find the value at
which the objective function is highest or lowest.
Example
Consider a manufacturer who produces two products, A and B. The profit per unit for
product A is Rs.40, and for product B is Rs.50. The manufacturer wants to maximize
profit subject to the following constraints:

16
  Product A requires 1 hour of machining and 2 hours of assembly per unit.
 Product B requires 3 hours of machining and 1 hour of assembly per unit.
 The manufacturer has 30 hours of machining time and 20 hours of assembly
time available per week.
Objective Function:
Maximize Z = 40x1 + 50x2
Constraints:
x1 + 3x2 ≤ 30 (Machining time)
2x1 + x2 ≤ 20 (Assembly time)
x1, x2 ≥ 0 (Non-negativity)

 Knowledge Check 1
Fill in the Blanks.
1. Linear programming is a mathematical technique designed to help in the
decision-making process for optimizing a particular ________. (objective
function )
2. The development of linear programming dates back to the ________ during
World War II. (1940s)
3. One of the characteristics of linear programming is the non-negativity
restriction, which means that decision variables cannot be ________.
(negative)
4. The graphical method of solving linear programming problems involves
plotting the constraints and identifying the ________ region. (feasible)

 Outcome-Based Activity 1
Draw a simple linear programming problem with two variables on a graph, plot
the constraints, and identify the feasible region.

2.4 Simplex Method

To find the optimal solution to linear programming problems, use the Simplex
Method, a systematic approach. When dealing with problems involving more than
two choice variables, it is especially helpful.
Steps in the Simplex Method

17
 1. Formulate the Problem: Define the objective function and constraints.
2. Convert to Standard Form: Ensure all constraints are in equality form by
adding slack variables.
3. Set Up the Initial Simplex Tableau: Create an initial tableau representing the
objective function and constraints.
4. Identify the Pivot Element: Determine the entering and leaving variables
based on the highest coefficient in the objective function row.
5. Perform Row Operations: Use row operations to transform the tableau and
update the values.
6. Repeat: Continue the process until all coefficients in the objective function
row are non-negative.
7. Extract the Optimal Solution: The final tableau provides the values of
decision variables that maximize or minimize the objective function.
Example
Consider a linear programming problem with the objective to maximize Z = 3x1 +
2x2, subject to the constraints:
x1 + x2 ≤ 4
x1 + 2x2 ≤ 6
x1, x2 ≥ 0
Standard Form:
x1 + x2 + s1 = 4
x1 + 2x2 + s2 = 6
s1, s2 ≥ 0
Initial Simplex Tableau:
Basic Variable x1 x2 s1 s2 RHS
s1 1 1 1 0 4
s2 1 2 0 1 6
Z -3 -2 0 0 0
Iteration Steps:
1. Identify the Pivot Column: Select the most negative coefficient in the Z row.
2. Identify the Pivot Row: Divide the RHS by the pivot column to find the
smallest positive ratio.
3. Perform Row Operations: Update the tableau.

18
 4. Repeat: Continue until all Z row coefficients are non-negative.
Final Simplex Tableau:
Basic Variable x1 x2 s1 s2 RHS
x1 1 0 1 -1 2
x2 0 1 -1 1 2
Z 0 0 1 1 10
Optimal Solution:

2.5 Dual Linear Programming

Dual linear programming involves transforming a given linear programming problem
(known as the primal problem) into another related problem (known as the dual
problem). Solving the dual problem provides insights into the original problem and
can sometimes be more efficient.
Formulation of Dual Problem
For a primal problem in the form:

The dual problem is formulated as:

19
Properties of Dual Problems
1. Weak Duality: The dual problem's objective function always has a value that
is larger than or equal to the primal problem's objective function.
2. Strong Duality: If the primal has an optimal solution, then the dual also has
an optimal solution, and the optimal values of their objective functions are
equal.
3. Complementary Slackness: If a primal constraint is binding, the
corresponding dual variable is zero, and vice versa.
Example
Consider the primal problem:

The corresponding dual problem is:

By solving the dual problem, we can obtain the same optimal solution for the primal
problem, demonstrating the strong duality property.

2.6 Sensitivity Analysis in Linear Programming

Sensitivity analysis is sometimes referred to as post-optimality analysis. It is the study
of how variations in the right-hand values of the constraints or the coefficients of the
objective function affect the optimal solution to a linear programming problem.
Importance of Sensitivity Analysis
1. Decision-making: Helps in understanding the impact of changes in input
parameters on the optimal solution.
2. Resource Allocation: Assists in making informed decisions about resource
reallocation.

20
 3. Robustness Check: Ensures that the solution remains feasible and optimal
under varying conditions.
Components of Sensitivity Analysis
1. Objective Function Coefficients: Examining the effect of changes in the
coefficients of the objective function on the optimal solution.
2. Right-Hand Side Values: Analyzing the impact of changes in the constraint
values on the feasible region and optimal solution.
3. Constraint Coefficients: Studying how changes in the coefficients of the
constraints affect the solution.
Methods of Sensitivity Analysis
1. Graphical Method: For problems with two decision variables, graphical
methods can be used to visualize the changes in the feasible region and
optimal solution.
2. Simplex Method: Using the final simplex tableau, sensitivity analysis can be
performed to determine the allowable range of changes for the coefficients and
constraints without altering the optimal solution.
Example
Consider a linear programming problem with the objective to maximize Z = 2x1 +
3x2, subject to the constraints:

After solving the problem using the Simplex Method, the optimal solution is x1 = 2, x2
= 4 with Z = 16.
To perform sensitivity analysis, we analyze the effect of changing the coefficient of
x1x_1x1 in the objective function from 2 to 2.5:
New Objective Function:
Z = 2.5x1 + 3x2
Re-solving the problem with the new objective function, we find that the optimal
solution remains the same, but the value of Z changes to 18.

 Knowledge Check 2
State True or False.

21
 1. The Simplex Method is only useful for problems with two decision variables.
(False)
2. Dual linear programming involves transforming a given linear programming
problem into a related problem. (True)
3. Sensitivity analysis is used to study how changes in the coefficients of the
objective function or constraints affect the optimal solution. (True)
4. The final simplex tableau does not provide any information about sensitivity
analysis. (False)

 Outcome-Based Activity 2
Perform sensitivity analysis on a given linear programming problem by changing
one of the objective function coefficients and observing the effect on the optimal
solution.

2.6 Summary
 Linear programming (LP) is a mathematical technique designed to optimize a
specific objective function under given constraints. It originated during World War
II for military resource allocation and was formalized by George Dantzig in 1947
with the Simplex algorithm.
 LP is used across various fields, such as business, economics, and engineering, for
tasks like resource allocation, production scheduling, and transportation logistics.
The primary goal is to find the best possible outcome within the given constraints.
 Linear programming is bent on the optimization of linear functions, where
optimization desire is slimmer maximization or minimization of the objective
function concerning a group of linear constraints of equalities or inequalities.
Some of the characteristics include linear objective function and constraints, non-
negative constraint, feasible region, and optimal solution.
 This involves the graphical solution of LP problems with two decision variables:
the plots of constraints of a problem, the feasible region, and the optimal solution,
as achieved by shifting the line of the objective function up or down parallel to
itself.
 The Simplex Method is a systematic procedure for finding the optimal solution to
LP problems with more than two decision variables. It involves converting the

22
 problem to standard form, setting up an initial tableau, performing row operations,
and iterating until all objective function coefficients are non-negative.
 Dual linear programming involves transforming a given LP problem (primal) into
a related problem (dual). Solving the dual can provide insights into the primal
problem and is sometimes more efficient. The dual problem's formulation mirrors
the primal with its own objective function and constraints.
 Techniques that can be used in sensitivity analysis include graphical solution
techniques that are available for problems with two decision variables and the
Simplex solution for more complicated problems. One example focuses on how
the coefficients of the objective function affect the location of the optimal solution
to illustrate the concept of sensitivity analysis.

2.7 Keywords
 Objective Function: The function in a linear programming problem that needs to
be maximized or minimized.
 Constraints: The restrictions or limitations on the decision variables in a linear
programming problem.
 Feasible Region: The area on a graph where all constraints overlap, representing
all possible solutions to a linear programming problem.
 Simplex Method: A systematic procedure for finding the optimal solution to a
linear programming problem with more than two variables.
 Sensitivity Analysis: The study of how the optimal solution to a linear
programming problem changes in response to variations in the coefficients of the
objective function or the right-hand side values of the constraints.

2.8 Self-Assessment Questions

1. What is the primary objective of linear programming?
2. Explain the steps involved in the graphical method for solving linear
programming problems.
3. Describe the Simplex Method and its importance in solving linear programming
problems.
4. How is the dual of a linear programming problem formulated, and what is its
significance?

23
5. What are the key components of sensitivity analysis in linear programming?

2.9 References / Reference Reading

 Taha, Hamdy A. Operations Research: An Introduction. 10th ed., Pearson, 2017.
 Ravindran, A., Don T. Phillips, and James J. Solberg. Operations Research:
Principles and Practice. 2nd ed., Wiley India Pvt. Ltd., 2016.
 Sharma, J. K. Operations Research: Theory and Applications. 6th ed., Macmillan
Publishers India, 2018.
 Hillier, Frederick S., and Gerald J. Lieberman. Introduction to Operations
Research. 10th ed., McGraw-Hill Education, 2014.
 Srinivasan, G. Operations Research: Principles and Applications. 2nd ed., PHI
Learning Pvt. Ltd., 2017.

24
 Unit 3: Transportation Problems

Learning Outcomes:
 Students will be able to define the general structure of transportation problems.
 Students will be able to apply methods for initial allocation in transportation
problems.
 Students will be able to identify degeneracy issues in transportation problems.
 Students will be able to determine optimal solutions for transportation problems.
 Students will be able to evaluate the applications of transportation models in
business.

Structure:
3.1 General Structure of Transportation Problems
3.2 Methods of Initial Allocation
 Knowledge Check 1
 Outcome-Based Activity 1
3.3 Degeneracy in Transportation Problems
3.4 Finding Optimal Solutions
3.5 Applications of Transportation Models in Business
 Knowledge Check 2
 Outcome-Based Activity 2
3.6 Summary
3.7 Keywords
3.8 Self-Assessment Questions
3.9 References / Reference Reading

25
3.1 General Structure of Transportation Problems
Transportation problems belong to a family of linear programming problems that
involve the transportation of resources from several sources to several recipients in
the cheapest manner. These problems are designed to solve one common goal of
minimizing the overall transportation cost subjected to supply and demand.
Comprising three broad parts, it usually presents the following general pattern:
 Components of Transportation Problems
o Sources (Supply Points): Export origins, which refer to sources of goods.
Each source has a fixed supply capacity for a particular product.
o Destinations (Demand Points): Locations where goods are received. Each
destination has a specific demand requirement.
o Cost Matrix: It is a table that represents the cost of transporting goods from
every source to destination.
 Formulation of Transportation Problems
Transportation problems can be formulated mathematically using the following
notation:
o xij: Number of units transported from source i to destination j.
o cij: Cost per unit of transporting from source i to destination j.
o ai: Supply available at source i.
o bj: Demand at destination j.
The objective is to minimize the total transportation cost:

Minimize
Subject to:

o Supply constraints: for each source i.

o Demand constraints: for each destination j.
o Non-negativity constraints: xij ≥ 0.
 Balanced vs Unbalanced Problems
o Balanced Transportation Problem: Total supply equals total demand (∑ai =
∑bj).
o Unbalanced Transportation Problem: Total supply does not equal total
demand. Dummy sources or destinations are added to balance the problem.

26
Example of a Transportation Problem
Consider a manufacturing company that needs to transport its products from three
warehouses to four retail stores. The supply at each warehouse and the demand at
each store are known, and the transportation cost per unit between each warehouse
and store is given. The company needs to determine how many units to transport from
each warehouse to each store to minimize the total transportation cost while satisfying
the supply and demand constraints.

3.2 Methods of Initial Allocation

To solve transportation problems, an initial feasible solution is needed. Several
methods can be used for initial allocation:
 North-West Corner Method
This method starts at the top-left (north-west) corner of the cost matrix and
allocates as much as possible to the corresponding cell until either the supply or
demand is exhausted. The process then moves right or down to the next cell and
repeats until all supplies and demands are met.
Steps of the North-West Corner Method
1. Start at the north-west corner cell (i.e., the first cell of the cost matrix).
2. Allocate as many units as possible to this cell, which is the minimum of the
supply of the row and the demand of the column.
3. Supply and demand are adjusted by subtracting the allocated units.
4. If you are out of stock, you can move down to the next row. If all products
under a column are sold out, feel free to proceed to the next column
immediately.
5. Repeat this process until all supplies and demands are satisfied.
 Least Cost Method
It seeks to select the cell with the lowest cost where, as much as possible, supply
and demand constraints are met. It proceeds to select the next least cost cell up to
the point that all supply and demand are satisfied.
Steps of the Least Cost Method
1. Determine the cell with the least cost in the cost matrix.
2. Designate as many units as possible in this cell, and it equals the previous
cell’s value of the minimum row supply and column demand of the cells.

27
 3. Supply and demand are adjusted by subtracting the allocated units.
4. Cross out the row or column where the supply or demand is exhausted.
5. Repeat the process until all supplies and demands are satisfied.
 Vogel's Approximation Method (VAM)
By calculating the difference between the lowest and second-lowest prices in a
row or column, VAM analyzes penalties for all of the columns and rows.
Subsequently, the cell with the lowest cost receives the maximum possible
allocation, and the row or column with the largest penalty is chosen. Until all
supplies and demands have been met, this process repeats.
Steps of Vogel's Approximation Method
1. Find out the penalty for each row and column by subtracting the lowest cost
from the second-lowest-cost.
2. Determine the row or column with the highest penalty.
3. Allocate as many units as possible to the cell with the lowest cost in the
identified row or column.
4. Supply and demand are adjusted by subtracting the allocated units.
5. Cross out the row or column where the supply or demand is exhausted.
6. Recalculate the penalties and repeat the process until all supplies and demands
are satisfied.
Example: Initial Allocation
Consider a transportation problem with the following cost matrix, supplies, and
demands:
D1 D2 D3 Supply
S1 8 6 10 20
S2 9 12 13 25
S3 14 9 16 15
Demand 15 25 20
Using the North-West Corner Method, the initial allocation is:
 Allocate 15 units to x11 (S1 to D1).
 Allocate 5 units to x12 (S1 to D2).
 Allocate 20 units to x22 (S2 to D2).
 Allocate 5 units to x23 (S2 to D3).
 Allocate 15 units to x33 (S3 to D3).

28
 Using the Least Cost Method:
1. The least cost cell is x12 with a cost of 6. Allocate 20 units.
2. The next least cost cell is x21 with a cost of 9. Allocate 15 units.
3. The next least cost cell is x22 with a cost of 12. Allocate 5 units.
4. The next least cost cell is x31 with a cost of 14. Allocate 10 units.
Using Vogel's Approximation Method:
1. Calculate penalties for each row and column.
2. Allocate to the cell with the lowest cost in the row or column with the highest
penalty.
3. Adjust supply and demand and recalculate penalties.
4. Repeat the process until all supplies and demands are met.

 Knowledge Check 1
Fill in the Blanks.
1. The main components of transportation problems include sources,
destinations, and the __________. (cost matrix)
2. In a balanced transportation problem, total supply __________ total demand.
(equals)
3. The North-West Corner Method starts at the __________ corner of the cost
matrix. (top-left)
4. Vogel's Approximation Method calculates penalties by finding the difference
between the __________ and second-lowest costs in a row or column. (lowest)

 Outcome-Based Activity 1
Draw a cost matrix for a transportation problem with three sources and three
destinations, and identify the initial allocation using the North-West Corner
Method.

3.3 Degeneracy in Transportation Problems

Degeneracy occurs when the number of filled cells in the transportation table is less
than m + n – 1, where mmm is the number of sources and n is the number of
destinations.

29
 Identifying Degeneracy
Degeneracy can be identified when, during the initial allocation or subsequent
iterations, there are fewer allocations than required. This situation can cause the
simplex method to cycle endlessly.
 Handling Degeneracy
To handle degeneracy, a small value (ε) is introduced in the vacant cells to ensure
that the number of allocations equals m + n − 1. This small value does not affect
the final solution but helps in proceeding with the algorithm.
Example: Handling Degeneracy
Consider the same cost matrix with supplies and demands balanced. If, after initial
allocation, only m + n – 2 cells are filled, add ε to a vacant cell with the least cost.
Practical Implications of Degeneracy
In real-world scenarios, degeneracy might represent situations where multiple
transportation routes have the same cost or capacity constraints are equally tight
across various routes. Handling degeneracy correctly ensures that the solution
remains feasible and optimal, avoiding potential misallocations or cycling issues
in the algorithm.

3.4 Finding Optimal Solutions

Once an initial feasible solution is obtained, the next step is to optimize it. The most
common methods for finding the optimal solution are the stepping stone method and
the MODI (Modified Distribution) method.
 Stepping Stone Method
The stepping stone method evaluates the potential of improving the current
solution by examining each empty cell (i.e., cells not used in the initial allocation).
The method involves:
o Constructing a closed path (loop) starting and ending at the empty cell.
o Alternatively, adding and subtracting a trial value along the path.
o Calculating the net change in cost.
o Updating the solution if the net change in cost is negative.
Steps of the Stepping Stone Method
1. Select an empty cell to evaluate.
2. Trace a closed path that starts from and returns to this empty cell.

30
 3. Assign alternating plus and minus signs along the path.
4. Find out the net change in total transportation cost by adding the costs at the
plus positions and subtracting the costs at the minus positions.
5. If the net change is negative, the current solution can be improved by adjusting
the allocations along the path.
Example: Stepping Stone Method
Using the previous example, assume the initial allocation is complete. To check
for optimality:
1. Select an empty cell (e.g., x13).
2. Form a closed path including x13, x12, x22, and x23.
3. Calculate the net change in cost.
4. Adjust the allocation if the net cost decreases.
 MODI (Modified Distribution) Method
The MODI method provides a systematic way to check optimality and improve
the solution:
1. Determine the opportunity cost for each empty cell.
2. Identify the cell which has the most negative opportunity cost.
3. Adjust the allocation to reduce total cost.
Steps of the MODI Method
1. Calculate ui and vj for each source and destination.
2. Identify the opportunity cost for each empty cell: cij − (ui + vj).
3. Identify the cell which has the most negative opportunity cost.
4. Adjust the allocations based on the cell with the most negative opportunity
cost.
Example: MODI Method
1. Find ui and vj values using the initial allocation.
2. Determine the opportunity costs for empty cells.
3. Adjust the allocation based on the most negative opportunity cost to minimize
the total transportation cost.
Practical Application of Optimization Methods
In practice, to solve large-scale transportation problems by applying these
methods, some instrumental and software tools in companies exist. These tools
enable automatic computation and testing, which makes the management of
intricate logistics networks possible.
31
3.5 Applications of Transportation Models in Business
Transportation models have various applications in business for optimizing logistics
and supply chain operations.
Supply Chain Management
Transportation models help various businesses minimise transportation costs and
improve overall supply chain efficiency. Companies can determine the best routes and
methods for shipping goods from warehouses to retail stores or consumers.
Production and Distribution Planning
In transportation model activities, companies can match production dates with
distribution strategies to make sure that products are there at certain points in time at a
certain place, and this, in effect, helps to eliminate time wastage.
Inventory Management
Transportation plays a significant or rather crucial role in ensuring that stocks are
controlled in the right manner to avoid high holding costs or low stock-out times.
Entrepreneurs can optimize the flow of these utilities and the use of their storage
areas.
Example: Real-World Application in India
A leading FMCG company in India utilises transportation models to optimize its
distribution network. By analyzing the transportation costs and routes, the company
reduces its overall transport expenses by 15%, ensuring timely delivery of products
across the country.
Cost Reduction Strategies
Transportation models enable businesses to utilise various cost-saving strategies, such
as consolidating shipments, selecting cost-effective transportation modes, and
negotiating better rates with carriers.
Environmental Impact
Optimizing transportation routes and methods can help to reduce carbon emissions
and fuel consumption, which contributes to a company's sustainability goals.
Example: Environmental Benefits
One large Indian e-tail firm chose to introduce transportation optimization models to
minimise its impact on the environment. Building on the idea of reducing unnecessary
delivery and the combination of transportation, the company managed to cut down is

32
fuel consumption as well as carbon emissions, thus implementing an environmentally
friendly policy.
Customer Service Improvement
By timely and cost-effective delivery of products, transportation models enhance
customer satisfaction. Accurate delivery of the product and efficient use of logistics
operations lead to customer satisfaction and loyalty.
Example: Enhancing Customer Service
Transportation models were applied to chemist shops and hospitals by a
pharmaceutical firm based in India to ensure the effective delivery of drugs. This
optimisation of the supply chain meant that supply chain issues were more
trustworthy and consequently affected customer satisfaction and trust directly.
Strategic Decision-making
Transportation models produce meaningful information that supports decision-making
in strategic business management. Organizations are able to analyse possible
transportation modes, analyse costs, prospect efficient and effective transportation
modes, weigh the cost consequences, and plan accordingly for strategic objectives.
Example: Strategic Use of Transportation Models
A logistics company in India once adapted transport models to increase its coverage
area. The company did proper modelling of different scenarios to decide on opening
new distribution centres and choosing better transportation approaches that would
bring more benefits and cut costs for the company while broadening its service
coverage.

 Knowledge Check 2
State True or False.
1. Degeneracy in transportation problems occurs when the number of filled cells
is less than m + n − 1. (True)
2. The Stepping Stone Method is used to find the initial feasible solution in
transportation problems. (False)
3. An e-commerce company can use transportation models to reduce its carbon
footprint by optimizing delivery routes. (True)
4. The MODI method calculates the opportunity cost for each filled cell in the
transportation table. (False)

33
 Outcome-Based Activity 2
Using a sample transportation problem, identify if there is degeneration and
explain how you would handle it using the small value (ε) method.

3.6 Summary
 Transportation problems involve moving goods from multiple sources to multiple
destinations at the lowest cost. This requires balancing supply and demand
constraints using a cost matrix.
 These problems are formulated using variables representing quantities transported,
costs, supplies, and demands, aiming to minimize total transportation costs while
satisfying constraints.
 The North-West Corner Method begins allocation at the top-left corner of the cost
matrix, proceeding to the next cell until all supply and demand constraints are
met.
 The Least Cost Method selects the cell with the lowest transportation cost for
allocation, continuing until all supplies and demands are satisfied.
 Degeneracy occurs when the number of filled cells is fewer than m + n − 1,
potentially causing the solution method to cycle indefinitely.
 To handle degeneracy, a small value (ε) is introduced into vacant cells, ensuring
the number of allocations meets the required count without affecting the final
solution.
 The Stepping Stone Method involves evaluating empty cells for potential cost
improvements, constructing closed paths, and adjusting allocations if net cost
decreases.
 The MODI (Modified Distribution) Method systematically calculates opportunity
costs for empty cells, identifying and adjusting the most negative cost cells to
optimize the solution.
 Both methods aim to refine the initial feasible solution, ensuring the total
transportation cost is minimized while meeting all constraints.
 Transportation models are essential in supply chain management, enabling
businesses to minimize costs and improve efficiency in logistics and distribution.

34
 These models assist in production planning, inventory management, and aligning
production schedules with distribution plans to meet demand effectively.

3.7 Keywords
 Sources: Locations from which goods are shipped in a transportation problem.
 Destinations: Locations where goods are received in a transportation problem.
 Cost Matrix: A table representing the cost of transporting goods from each source
to each destination.
 Degeneracy: A situation in transportation problems where the number of filled
cells is less than m + n − 1.
 MODI Method: A method used to find the optimal solution by calculating the
opportunity costs for each empty cell in the transportation table.

3.8 Self-Assessment Questions

1. What are the main components of a transportation problem?
2. How is the initial allocation determined using the North-West Corner Method?
3. Explain the concept of degeneracy in transportation problems and how it can be
handled.
4. Describe the steps involved in the Stepping Stone Method for finding an optimal
solution.
5. What are the practical applications of transportation models in business?

3.9 References / Reference Reading

 Sharma, J.K. Operations Research: Theory and Applications. Macmillan India,
2022.
 Taha, Hamdy A. Operations Research: An Introduction. Pearson Education India,
2023.
 Ravindran, A., Don T. Phillips, and James J. Solberg. Operations Research:
Principles and Practice. Wiley India, 2021.
 Kanti Swarup, P. K. Gupta, and Man Mohan. Operations Research. Sultan Chand
& Sons, 2023.
 Hira, D.S., and P.K. Gupta. Operations Research. S. Chand Publishing, 2022.

35
 Unit 4: Assignment Problems

Learning Outcomes:
 Students will be able to define the structure of assignment problems.
 Students will be able to apply different methods to solve assignment problems.
 Students will be able to identify variations in assignment problems.
 Students will be able to utilize the Hungarian method for assignment problems
effectively.

Structure:
4.1 Structure of Assignment Problems
4.2 Methods to Solve Assignment Problems
 Knowledge Check 1
 Outcome-Based Activity 1
4.3 Variations in Assignment Problems
4.4 Hungarian Method for Assignment Problems
 Knowledge Check 2
 Outcome-Based Activity 2
4.5 Summary
4.6 Keywords
4.7 Self-Assessment Questions
4.8 References / Reference Reading

36
4.1 Structure of Assignment Problems
Assigning a set of resources to a set of tasks in a way that minimizes cost or time or
maximizes profit is the aim of assignment issues, a particular kind of combinatorial
optimization problem. This kind of issue frequently comes up in operational research
and has applications in a number of different domains, including logistics, scheduling,
and resource allocation.
Definition: An assignment problem can be defined as follows:
 There are n tasks and n agents.
 Each agent can be assigned to perform exactly one task.
 Each task must be assigned to exactly one agent.
 The cost associated with assigning an agent to a task is known and given in a
cost matrix.
Cost Matrix: The cost matrix is a square matrix C where Cij represents the cost of
assigning the i-th agent to the j-th task. The objective is to find a one-to-one matching
between agents and tasks that minimizes the total cost.
Example: Consider a simple example where there are three agents and three tasks.
The cost matrix might look like this:
Task 1 Task 2 Task 3
Agent 1 9 2 7
Agent 2 6 4 3
Agent 3 5 8 1

The goal is to find an assignment of agents to tasks that minimize the total cost.

4.2 Methods to Solve Assignment Problems

Several methods can be used to solve assignment problems. These methods vary in
complexity and efficiency. The most commonly used methods include:
 Enumerative Methods
Enumerative methods entail identifying all the possible solutions and choosing the
one that has minimized cost. This approach is more efficient and provides the best
solution as compared to all other methods. Still, it is almost impossible to solve
the large problem as the number of possible combinations is equal to the factorial
of a total number of items.

37
 Example: For n agents and n tasks, there are n! (n factorial) possible assignments.
For example, with 3 agents and 3 tasks, there are 3! = 6 possible assignments:
1. Agent 1 to Task 1, Agent 2 to Task 2, Agent 3 to Task 3
2. Agent 1 to Task 1, Agent 2 to Task 3, Agent 3 to Task 2
3. Agent 1 to Task 2, Agent 2 to Task 1, Agent 3 to Task 3
4. Agent 1 to Task 2, Agent 2 to Task 3, Agent 3 to Task 1
5. Agent 1 to Task 3, Agent 2 to Task 1, Agent 3 to Task 2
6. Agent 1 to Task 3, Agent 2 to Task 2, Agent 3 to Task 1
After calculating the total cost for every assignment, selecting the one with the
minimum cost can become computationally infeasible for larger nnn.
 Simplex Method
The simplex method can solve assignment problems by applying a procedure that
converts assignment problems into linear programming issues. This method is
useful for solving particular types of assignment problems but can sometimes be
impractical because of the transformation time.
Example: The primary goal of the assignment problem, which can be expressed
as a linear programming problem, is to minimize the total cost. The restrictions
guarantee that every agent is paired with a specific task and that every task is
paired with a particular agent. The optimal solution can be found by iterating
through possible solutions using the simplex algorithm.
Hungarian Method
The Kuhn-Munkres algorithm, sometimes referred to as the Hungarian technique,
is a combinatorial optimization algorithm specifically created to address
assignment problems. It is efficient and straightforward to implement. This
method transforms the original cost matrix into a series of steps to find the optimal
assignment.
Example: The Hungarian method is based on constructing a cost matrix by
subtracting row and column minima in order to produce a reduced matrix and
going through a sequence of steps, covering zeros and scaling until an assignment
is possible.

38
 Knowledge Check 1
Fill in the Blanks.
1. An assignment problem can be defined as one where each agent is assigned to
perform exactly _____ task. (one)
2. In the Hungarian method, the first step is to create the _____ matrix. (cost)
3. The _____ method involves listing all possible assignments and selecting the
one with the minimum cost. (enumerative)
4. For an assignment problem to be balanced, the number of agents and tasks
must be _____. (equal)

 Outcome-Based Activity 1
Create a simple cost matrix for three agents and three tasks, then perform the first
two steps of the Hungarian method on this matrix.

4.3 Variations in Assignment Problems

Assignment problems can be in various forms, and understanding these variations is
crucial for applying the correct technique for solving assignment problems. Some
common variations include:
Unbalanced Assignment Problem
In an unbalanced assignment problem, the number of agents and tasks are not equal.
This can be made by adding dummy agents or tasks with zero cost to balance the
problem.
Example: If there are 4 tasks and 3 agents, a dummy agent can be added with costs of
zero for each task, making it a balanced problem. The new cost matrix might look like
this:
Task 1 Task 2 Task 3 Task 4
Agent 1 9 2 7 0
Agent 2 6 4 3 0
Agent 3 5 8 1 0
Dummy 0 0 0 0

By adding the dummy row, the problem becomes a 4x4 balanced assignment problem;
after that, a balanced assignment problem can be solved using standard methods.

39
Maximization Assignment Problem
Sometimes, the main objective is to maximize the total profit instead of reducing the
the cost. This can be transformed into a minimization problem by subtracting each
profit value from a large constant and then solving the resulting minimization
problem.
Example: Consider a profit matrix:
Task 1 Task 2 Task 3
Agent 1 10 5 8
Agent 2 7 6 9
Agent 3 9 8 6
To transform this into a minimization problem, subtract each value from the
maximum profit (10 in this case):
Task 1 Task 2 Task 3
Agent 1 0 5 2
Agent 2 3 4 1
Agent 3 1 2 4
This new matrix can now be solved as a minimization problem.
Multiple Assignment Problem
In multiple assignment problems, each agent is allowed to be assigned to several
tasks, or each task can be completed by a number of agents. This introduces an extra
degree of freedom and might possibly need further techniques such as algorithms or
linear programming.
Example: If an agent can handle multiple tasks, then the problem may involve
capacity constraints. Consider an agent can handle up to two tasks. This would require
the formulation of the problem with additional constraints to ensure that no agent is
assigned more tasks than they can handle.
Generalized Assignment Problem
In generalized assignment problems, one or all of the tasks of an agent can be
assigned to multiple tasks with capacity constraints the amount of time that the agent
has. This kind of problem needs formulation and solution at a higher level of
mathematics than the basic one mentioned above.

40
Example: If agents have different capacities (e.g., hours available), the problem must
account for these capacities. Consider each agent has a maximum number of hours
they can work, and each task requires a certain number of hours. The goal is to assign
tasks to agents without exceeding their capacity while minimizing the total cost.

4.4 Hungarian Method for Assignment Problems

The Hungarian method is identified as one of the best and most commonly used
algorithms that work a lot in addressing assignment issues. The steps enabling it
consist of arriving at a simpler form of the issue and then solving it through an
iteration. The steps involved in the Hungarian method are as follows:
Step 1: Create the Cost Matrix
Create a cost matrix representing the assignment costs between tasks and agents.
Ensure that the matrix is square. If not, then make it square by adding dummy agents
or tasks.
Example: A 3x3 cost matrix might look like this:
Task 1 Task 2 Task 3
Agent 1 9 2 7
Agent 2 6 4 3
Agent 3 5 8 1

Step 2: Subtract Row Minimums

For each row in the cost matrix, subtract the smallest value in that row from all other
values in the same row. This step ensures that each row has at least one zero.
Example: Subtracting row minimums:
Task 1 Task 2 Task 3
Agent 1 7 0 5
Agent 2 3 1 0
Agent 3 4 7 0
Step 3: Subtract Column Minimums
Next, for each column in the modified matrix, subtract the smallest value in that
column from all other values in the same column. This step ensures that each column
has at least one zero.
Example: Subtracting column minimums:

41
 Task 1 Task 2 Task 3
Agent 1 4 0 5
Agent 2 0 1 0
Agent 3 1 7 0

Step 4: Cover All Zeros with Minimum Lines

Cover all zeros in the resulting matrix with a minimum number of horizontal and
vertical lines. The goal is to cover all zeros with the fewest possible lines.
Example: In the above matrix, the minimum number of lines required to cover all
zeros is two.
Step 5: Adjust the Matrix
An optimal assignment can be established if the number of rows (or columns) equals
the number of lines. If not, modify the matrix by adding the smallest uncovered value
to the components covered by two lines and subtracting it from all uncovered
elements.
Example: If the minimum number of lines is less than the number of rows, adjust the
matrix and repeat steps 3 and 4 until an optimal assignment is found. Suppose the
smallest uncovered value is 1. Subtract this value from all uncovered elements and
add it to the elements covered by two lines:
Task 1 Task 2 Task 3
Agent 1 3 0 4
Agent 2 0 2 0
Agent 3 0 6 0

Step 6: Make the Optimal Assignment

Once the minimum number of lines equals the number of rows or columns, make the
optimal assignment by choosing the zeros in such a way that no two assignments are
in the same row or column.
Example: In the adjusted matrix, select the zeros to assign agents to tasks optimally:
 Agent 1 to Task 2
 Agent 2 to Task 3
 Agent 3 to Task 1

42
This results in the optimal assignment with a total cost of:

 Knowledge Check 2
State True or False.
1. An unbalanced assignment problem occurs when the number of agents is
equal to the number of tasks. (False)
2. The Hungarian method can handle both minimization and maximization
assignment problems by transforming the cost matrix appropriately. (True)
3. In the Hungarian method, if the number of lines needed to cover all zeros in
the matrix equals the number of rows, an optimal assignment can be made.
(True)
4. A generalized assignment problem involves each agent being assigned exactly
one task without any capacity constraints. (False)

 Outcome-Based Activity 2
Using the Hungarian method, cover all zeros in a given 3x3 cost matrix with the
minimum number of lines and determine if an optimal assignment can be made.

4.6 Summary
 Assignment problems involve assigning nnn agents to nnn tasks such that each
agent performs one task and each task is assigned to one agent, aiming to
minimize cost or maximize profit.
 The problem is represented using a cost matrix where each element indicates the
cost of assigning a particular agent to a particular task.
 The goal is to find the optimal assignment that minimizes the total cost, ensuring a
one-to-one matching between agents and tasks.
 Enumerative methods list all possible assignments and select the one with the
minimum cost, but are computationally impractical for large problems due to
factorial growth.
 The simplex method transforms the assignment problem into a linear
programming problem, offering an efficient solution for specific problem types.

43
 The Hungarian method is a combinatorial optimization algorithm that iteratively
reduces the cost matrix to find the optimal assignment efficiently.
 Unbalanced assignment problems arise when the number of agents and tasks are
unequal. They can be converted to balance by adding dummy agents or tasks at
zero cost.
 Maximization assignment problems aim to maximize profit, and they can be
converted to minimization problems by subtracting each profit value from a large
constant.
 Generalized assignment problems involve assigning different tasks to agents
under capacity constraints, requiring advanced algorithms for solutions.
 The Hungarian method starts by creating a cost matrix, ensuring it is square by
adding dummy agents or tasks if necessary.
 It then reduces the matrix by subtracting row and column minimums, covering
zeros with the minimum number of lines, and adjusting the matrix iteratively.
 The process continues until the number of lines equals the number of rows or
columns, allowing for an optimal assignment by selecting zeros appropriately.

4.7 Keywords
 Cost Matrix: A square matrix where each element represents the cost of assigning
a particular agent to a particular task.
 Hungarian Method: An algorithm used to find the optimal assignment in
assignment problems by reducing the problem into a simpler form.
 Enumerative Methods: Techniques that involve listing all possible assignments
and selecting the optimal one.
 Unbalanced Assignment Problem: An assignment problem where the number of
agents and tasks are not equal.
 Generalized Assignment Problem: An assignment problem where agents can
handle multiple tasks within their capacity constraints.

4.8 Self-Assessment Questions

1 What is the structure of an assignment problem, and how is it typically
represented?
2 Describe the Hungarian method and its steps for solving assignment problems.

44
3 Explain the concept of an unbalanced assignment problem and how it can be
addressed.
4 Discuss the variations in assignment problems and their implications.
5 How can the Hungarian method be applied to a maximization assignment
problem?

4.9 References / Reference Reading

 Kapoor, V. K., and Sultania, S. C. Operations Research: Techniques for
Management. Sultan Chand & Sons, 2019.
 Sharma, J. K. Operations Research: Theory and Applications. 6th ed., Macmillan
Publishers India, 2020.
 Taha, Hamdy A. Operations Research: An Introduction. 10th ed., Pearson
Education India, 2017.
 R. Panneerselvam. Operations Research. 2nd ed., PHI Learning, 2010.
 Gupta, Prem Kumar, and Hira, D. S. Operations Research: An Introduction. S.
Chand Publishing, 2015.

45
 Unit 5: Network Analysis

Learning Outcomes:
 Students will be able to understand the principles of PERT and CPM.
 Students will be able to identify the stages in the application of PERT.
 Students will be able to analyze networking with PERT.
 Students will be able to compare the characteristics and applications of PERT and
CPM.
 Students will be able to evaluate real-world examples of PERT and CPM in
industry settings.

Structure:
5.1 Introduction to PERT/CPM
5.2 Background and Development of PERT/CPM
 Knowledge Check 1
 Outcome-Based Activity 1
5.3 Stages in Application of PERT
5.4 Networking Analysis with PERT
5.5 Comparison of PERT and CPM
 Knowledge Check 2
 Outcome-Based Activity 2
5.6 Summary
5.7 Keywords
5.8 Self-Assessment Questions
5.9 References / Reference Reading

46
5.1 Introduction to PERT/CPM
Scheduling is another fundamental activity in network analysis when it comes to the
planning and controlling of a project. There are two important methods in network
analysis, which are PERT and CPM. Among the network analysis techniques, the two
commonly used are the Program Evaluation and Review Technique (PERT) & Critical
Path Method (CPM).
PERT (Program Evaluation and Review Technique): Pert is a planning and control
tool in project management that helps identify and schedule the activities to be carried
out in a project. This is very useful for those projects where the amount of time
needed to complete different activities is not really clear. PERT was developed in the
1950s by the U. S. navy to assist in design choice and project management where
there is uncertainty within the timing of a single project.
CPM (Critical Path Method): CPM is a systematic project management approach
for process planning that distinguishes between essential and non-essential tasks with
the aim of avoiding time frame issues and process constriction. Du Pont and
Remington Rand invented it in the 1950s. CPM is most appropriate when there are
pre number of tasks, and each of them has a fixed time measurement. While PERT
uses statistical time estimates, CPM uses arbitrary time estimates and is appropriate
for projects where the time estimate is precise.

5.2 Background and Development of PERT/CPM

The origins of PERT and CPM can be traced back to the 1950s. They were developed
independently but have many similarities in their application and purpose.
Development of PERT:
 Origin: PERT, originally known as the Program Evaluation Review Technique,
was acquired by the United States Navy in 1957 for the Polaris missile submarine
project. The purpose of implementing this protocol was to coordinate the
challenging functions and make sure to facilitate the project schedule. Certain
roles in the Polaris project were highly exposed and complex, so they could not
be managed efficiently through traditional models, which included subcontractors
and thousands of tasks at different phases. Concerning which, PERT offered a
patterned approach to deal with the probabilities of risks and dependence of the
tasks.

47
 Purpose: PERT stands for Program Evaluation Review Technique, and it was
meant to handle complex activities and dependencies and guarantee the
accomplishment of project phases on schedule. It was used when the timing of
tasks involved in a project was unpredictable and could take a wide range of time
frames PERT stands for Program Evaluation Review Technique, and it was meant
to handle complex activities and dependencies as well as guarantee the
accomplishment of project phases on schedule. It was used when the timing of
tasks involved in a project was unpredictable and could take a wide range of time
frames.
 Features: PERT is probabilistic, which deals with uncertain activity times by
using three-time estimates: optimistic, most likely, and pessimistic. This approach
allows project managers to anticipate potential delays and plan accordingly.
Development of CPM:
 Origin: CPM was developed for the DuPont Corporation and Remington Rand in
1956 to assist in the management of large-scale projects. The first use was in the
construction industry when DuPont required an accurate approach to schedule and
monitor the repair of chemical compound facilities. Later on, the procedure was
applied to other sectors of the economy.
 Purpose: CPM was created to address the project management needs in the
construction and manufacturing industries. It focuses on the optimal allocation of
resources and the identification of critical tasks that must be completed on time to
avoid delays in the overall project.
 Features: CPM is deterministic, focusing on the time-cost trade-offs in project
management by using a single time estimate for each activity. This approach helps
project managers to optimize project schedules and manage costs effectively.

 Knowledge Check 1
Fill in the Blanks.
1. PERT uses a _______ approach, dealing with uncertain activity times by using
three-time estimates: optimistic, most likely, and pessimistic. (probabilistic)
2. The critical path is the _______ path through the network with the shortest
overall project completion time. (longest)
3. CPM is best suited for projects with _______ tasks and established durations.
(well-defined)
48
 Outcome-Based Activity 1
Draw a simple network diagram with at least five tasks and identify the critical
path.

5.3 Stages in Application of PERT

Applying PERT involves several key stages that ensure comprehensive project
planning and control:
1. Project Identification:
o Define the project scope and objectives: Clearly outline the goals of the project
and the deliverables expected upon completion. This step involves defining the
major phases and tasks required to achieve the project objectives.
o List all tasks and activities required to complete the project: Break down the
project into smaller, manageable tasks. Each task should have a clear objective
and outcome.
2. Task Sequencing:
o Identify the dependencies between tasks: Identify the sequence in which they
should undertake tasks if each of the activities has to be completed at various
stages. This implies that one activity may have to wait for another to be
completed, while some may be done in parallel.
o Develop a sequence of activities based on their dependencies: Set out what links
their work needs to be logically and all associated dependences taken into
consideration. This step is useful for plotting the activities and detecting where the
major process constraints are shown.
3. Time Estimation:
o For each task, estimate the time required using three-time estimates:
 Optimistic Time (O): The shortest time in which the task can be completed if
everything goes perfectly.
 Most Likely Time (M): It is the most probable time required to complete the task
under normal circumstances.
 Pessimistic Time (P): It is the longest time that the task might take if
complications arise.

49
4. Network Diagram:
o Draw a network diagram representing the sequence of tasks. Use nodes to
represent events and arrows to represent tasks. The network diagram usually gives
the flow of tasks and their dependencies.

o Calculate the expected time (TE) for each task: Use this formula to
calculate the expected time. This formula gives a weighted average, giving more
importance to the most likely time.
5. Critical Path Identification:
o Identify the critical path: The critical path is the longest path in the network with
the shortest overall project completion time. It includes all tasks that directly
impact the project completion date. Any delay in the tasks will delay the entire
project.
o Calculate the earliest start (ES) and finish (EF) times for each task: Determine the
earliest start and finish time for each task.
o Calculate the latest start (LS) and finish (LF) times for each task: By Performing a
backward pass, determine the latest start and latest finish time.
6. Monitoring and Control:
o Regularly update the PERT chart with actual progress: Continuously monitor the
progress of tasks and update the PERT chart to note down any changes.
o Identify and manage any delays or changes in the project schedule: Adjust the
project plan as needed to accommodate changes and ensure that the project stays
on track.

5.4 Networking Analysis with PERT

Networking analysis using PERT involves detailed examination and optimization of
the project schedule to ensure timely completion. The following are the steps involved
in performing PERT analysis:
1. Defining Activities and Milestones:
o In this step, the project is breakdown into smaller tasks and the criteria for
completion of each task.

50
2. Constructing the Network Diagram:
o Using all the data, draw the network diagram based on the defined activities and
their dependencies. This network diagram helps understand the flow of tasks and
identify potential bottlenecks.
3. Estimating Activity Durations:
o The three-point estimation technique is used to calculate the expected time for
each activity. This technique involves calculating the optimistic, most likely, and
pessimistic times and then using the formula to find the expected
time.
4. Calculating the Earliest and Latest Start and Finish Times:
o Forward Pass: Start with scheduling a forward pass through the network to find
out the earliest start (ES) and finish (EF) times for each task. This involves
performing the first task and fixing the latest times at which each of the following
tasks can begin and finish.
o Backward Pass: Shift the times by going back to find the latest start (LS) and
finish (LF) times for each task. This involves working from the last step in the
order and identifying the modes or latest times that the preceding steps may begin
and end without causing the project to be delayed.
5. Identifying the Critical Path:
o The critical path is evaluated by identifying the path with zero slack time
(difference between ES and LS, or EF and LF).
6. Analyzing the Critical Path:
o Monitoring of the activities on the critical path should be emphasized more since
any hindrance to these activities will automatically slow down the whole project.
This step requires project monitoring and evaluation to check whether the project
is on track and then make necessary modifications.
7. Review and Adjustments:
o Regularly review the progress of the project and make adjustments as necessary to
stay on track. This includes updating the PERT chart with actual progress and re-
evaluating the critical path if any changes occur.

51
5.5 Comparison of PERT and CPM
PERT and CPM have many similarities and distinct differences that make each
method suitable for different types of projects.
Similarities:
 Both methods use network diagrams to represent project tasks and their
dependencies.
 Both identify the critical path to highlight crucial tasks.
 Both are used for planning, scheduling, and controlling project activities.
 Both methods help identify potential delays and bottlenecks in the project
schedule.
Differences:
Feature PERT CPM
Focus Time management and meeting Time and cost management
deadlines
Nature Probabilistic (uses three time Deterministic (uses a single
estimates) time estimate)
Application Research and development Construction and
projects manufacturing projects
Activity Time Optimistic, most likely, and Single time estimate
Estimation pessimistic estimates
Flexibility More flexible, suitable for Less flexible, suitable for
uncertain tasks well-defined tasks

Examples:
 PERT: Used in projects like new product development, where task durations
are uncertain and can vary widely.
 CPM: Used in construction projects where tasks are well-defined, and
durations can be accurately estimated.
Real-World Industry Applications
1. Construction Industry:
o Example: Due to its accuracy, CPM is widely employed in the construction of
buildings, bridges, and roads. However, by determining the critical path, the

52
 project managers are able to maintain the timeline expected for the completion of
the project as well as incorporate only those activities that will be necessary for
the project’s completion within the expected budget.
o Application: In developing a high-rise building, CPM is useful in organizing
activities, which include construction of the substructure, Superstructure,
electrical installation and other overall activities such as finishing. It should also
be noted that each of these tasks has its own dependencies and time; you should
be aware of the potential threats in terms of time and increasing costs.
2. Software Development:
o Example: PERT is used in software development projects, mostly in situations
where uncertainty exists with the duration of tasks. It can assist in streamlining
large software development projects where the time to code, test & deploy may be
unpredictable.
o Application: When it comes to implementing a new software application
development, Pert assists in defining tasks, including requirement analysis,
designing, coding, testing, and deployment. Each of these tasks might take a
different time in hours based on elements such as resource accessibility and the
Technical Complexity Level.
3. Event Planning:
o Example: PERT and CPM techniques could be applied to any type of event
planning, including business conferences, weddings, festivals, etc. In this way,
based on the definition of the critical tasks, it will become possible to effectively
identify all critical activities that need to be performed for organizational success.
o Application: Thus, PERT and CPM techniques could cover all types of event
planning, which might include business conferences, weddings, festivals etc.
which, in accordance with the definition of critical tasks, will help to identify all
critical activities that should be completed for organizational efficiency..

5.6 Formulas and Calculations in PERT/CPM

PERT Formulas:

53
Example Calculation:
 Task A:
o Optimistic Time (O): 3 days
o Most Likely Time (M): 5 days
o Pessimistic Time (P): 9 days

o Expected Time (TE):

o Variance
o Standard Deviation (σ):

Diagrams in PERT/CPM
Network Diagram: A network diagram visually represents the sequence of tasks and
their dependencies. Here is a simple example:

Start
|
Task A
|
Task B
|
Finish

Gantt Chart: A Gantt chart is a visual representation of the project schedule, which
indicates the start and finish dates of each task. It is useful for tracking progress and
identifying delays.

| Task | Start | Finish | Duration |

|--------|-------|--------|----------|
| Task A | Day 1 | Day 5 | 4 days |
| Task B | Day 6 | Day 10 | 4 days |

Both PERT and CPM are essential in managing projects since each of the methods has
its particular pros in the right project. These techniques and knowledge help improve
and make processes easier when dealing with complex projects. Through the

54
formulation of networks to detail, PERT and CPM assisted project managers in their
organizational planning and programmes in order to guarantee the timely completion
of projects within acceptable standards of quality as well as costs.

 Knowledge Check 2
State True or False.
1. PERT uses a deterministic approach to handle task durations. (False)
2. The critical path is the longest path through the network that determines the
shortest project completion time. (True)
3. CPM is designed to manage projects with uncertain task durations. (False)
4. Both PERT and CPM are used for planning, scheduling, and controlling
project activities. (True)

 Outcome-Based Activity 2
List three real-world projects where PERT or CPM could be effectively applied
and explain why.

5.6 Summary
 PERT (Program Evaluation and Review Technique) is a planning and control tool
used to manage projects with uncertain task durations, employing three time
estimates for each task: optimistic, most likely, and pessimistic.
 CPM (Critical Path Method) is a project management technique focused on
optimizing project schedules by defining critical and non-critical tasks with
known durations, making them suitable for well-defined projects.
 PERT was developed by the U.S. Navy in 1957 to manage the Polaris missile
submarine project, helping to handle the complexities and uncertainties of
numerous tasks and subcontractors.
 CPM was created by the DuPont Corporation and Remington Rand in 1956, and
initially applied to construction projects to optimize time and cost management by
identifying critical tasks.
 The stages in applying PERT include defining the project scope, listing all tasks,
identifying task dependencies, and developing a sequence of activities based on
these dependencies.

55
 Time estimation involves using three time estimates for each task, followed by
constructing a network diagram, identifying the critical path, and continuously
monitoring and controlling the project schedule.
 Networking analysis with PERT involves defining activities and milestones,
constructing the network diagram, estimating activity durations using the three-
point estimation technique, and calculating the earliest and latest start and finish
times.
 The critical path is identified by performing forward and backwards passes
through the network diagram, focusing on tasks with zero slack time, and
regularly reviewing and adjusting the project plan to stay on track.
 PERT and CPM share similarities in using network diagrams and identifying the
critical path, but they differ in their approach: PERT is probabilistic with three-
time estimates, while CPM is deterministic with a single time estimate.
 PERT is more flexible and suited for research and development projects with
uncertain task durations, whereas CPM is suitable for construction and
manufacturing projects with well-defined tasks and durations.

5.7 Keywords
 PERT (Program Evaluation and Review Technique): A project management
tool used for planning and controlling tasks, particularly useful for projects with
uncertain task durations.
 CPM (Critical Path Method): A project management technique focused on
defining critical and non-critical tasks to prevent time-frame problems and
optimize project schedules.
 Critical Path: The longest sequence of tasks in a project that determines the
shortest possible completion time.
 Expected Time (TE): In PERT, the weighted average time is calculated using
optimistic, most likely, and pessimistic time estimates.
 Network Diagram: A visual representation of the sequence of tasks and their
dependencies in a project.

56
5.8 Self-Assessment Questions
1. What are the main differences between PERT and CPM in terms of their
application and focus?
2. Describe the steps involved in creating a network diagram for a project.
3. How do you calculate the expected time for a task in PERT? Provide a detailed
example.
4. Explain how the critical path is identified in a network diagram.
5. Discuss the advantages and disadvantages of using PERT in project management.

5.9 References / Reference Reading

 Goel, B.S. Project Management: PERT and CPM. New Age International, 2020.
 Bhave, Prasanna Chandra. Projects: Planning, Analysis, Selection, Financing,
Implementation, and Review. Tata McGraw-Hill, 2021.
 Dhillon, B.S., and Chandra Mohan. Project Management and Control: PERT,
CPM and S. Vikas Publishing House, 2022.
 Nicholas, John M., and Herman Steyn. Project Management for Engineering,
Business, and Technology. Routledge, 2020.
 Gopalakrishnan, P. Handbook of Project Management. Macmillan India, 2019.

57
 Unit 6: Critical Path Method (CPM)

Learning Outcomes:
 Students will be able to understand the fundamentals of the Critical Path Method
(CPM).
 Students will be able to identify the critical path in a project network.
 Students will be able to calculate the earliest expected and latest allowable times
for project activities.
 Students will be able to apply CPM techniques in project management scenarios.
 Students will be able to evaluate the effectiveness of CPM in optimizing project
schedules.

Structure:
6.1 Introduction to CPM
6.2 Determination of Critical Path
 Knowledge Check 1
 Outcome-Based Activity 1
6.3 Determination of Earliest Expected and Latest Allowable Times
6.4 Applications of CPM in Project Management
 Knowledge Check 2
 Outcome-Based Activity 2
6.5 Summary
6.6 Keywords
6.8 Self-Assessment Questions
6.9 References / Reference Reading

58
6.1 Introduction to CPM
Critical Path Method (CPM) is a project management technique which is employed to
identify activities whose early start or late finish influences the project completion
time. Established around 1958, CPM is extensively used today in different industries
as a fundamental tool in controlling project complexity. It plays a central role in
determining the activities that must be done within a given period to avoid any
complications to the entire project.
There are certain processes involved in CPM, which comprises the development of a
project network diagram that depicts the tasks of a project and the relation between
them. Boxes are used to depict tasks, while arrows depict the flow in the solution
process and how one task depends on another. Finding the longest path in the network
that requires the longest time to finish a task is the main goal of CPM. The network's
critical route indicates the shortest amount of time that can be spent finishing the
project.
6.2 Determination of Critical Path
The critical path represents the flow of activities that determine the longest time
possible in order to achieve the project’s completion. To determine the critical path,
follow these steps:
1. List All Activities: First of all, one should identify all the activities needed to
perform the entire project. A unique identifier should mark each of them as an
activity.
2. Identify Dependencies: Define the dependences between activities.
Determine which activities can only occur if they are carried out before others.
3. Draw the Network Diagram: Develop a simple network diagram based on
the established activities and their interrelations. Label activities using circles
or any geometric shapes called nodes while connecting different nodes using
lines known as arrows.
4. Estimate Activity Durations: Assign an estimated duration to every activity.
5. Perform Forward Pass Calculation: Compute the early start (ES) and the
early finish (EF) for each activity based on progressing forward though the
network from start to finish.
6. Perform Backward Pass Calculation: Calculate the latest start time for each
activity and the latest finish time for each activity, starting with the finish time
and the start time, respectively and working backwards through the network.

59
 7. Identify the Critical Path: Identify the critical path with zero slack (the
difference between LF and EF or LS and ES is zero).
Example
Consider a project with the following activities and durations:
Activity Duration (days) Predecessor
A 3 -
B 2 A
C 4 A
D 2 B
E 1 C, D

1. Forward Pass Calculation:

Activity ES EF
A 0 3
B 3 5
C 3 7
D 5 7
E 7 8

2. Backward Pass Calculation:

Activity LS LF
A 0 3
B 3 5
C 3 7
D 5 7
E 7 8
3. Identify the Critical Path: The critical path is A-C-E with a total duration of
8 days.

 Knowledge Check 1
Fill in the Blanks.
1. The Critical Path Method (CPM) was developed in the late _____ by DuPont.
(1950s)

60
 2. In CPM, the critical path is identified by the activities with zero _____. (slack)
3. The primary goal of CPM is to identify the _____ path through the network,
known as the critical path. (longest)
4. The earliest finish (EF) time for an activity is calculated as EF = ES + _____.
(duration)

 Outcome-Based Activity 1
Draw a simple network diagram for a project involving five activities and identify
the critical path.

6.3 Determination of Earliest Expected and Latest Allowable Times

Earliest Expected Time and Latest Allowable Time means calculating ES, EF, LS and
LF for each activity needed in a project. This computation enables the project
managers to know how best to schedule a particular project and, at the same time,
determine whether there are likely to be some delays.
Forward Pass Calculation
The forward pass calculation starts from the beginning of the project and moves to the
end, calculating the earliest times each activity can start and finish:
1. Earliest Start (ES): It is the time at which an activity begins.
2. Earliest Finish (EF): It is the time at which an activity can be completed,
calculated as EF = ES + Duration.
Backward Pass Calculation
The backward pass calculation starts from the project’s end. It moves towards the
beginning of the project, calculating the latest times each activity can start and finish
without delaying the project:
1. Latest Finish (LF): It is the time at which an activity can be completed
without delaying the project.
2. Latest Start (LS): It is the time at which an activity can begin without
delaying the project, calculated as LS = LF − Duration.
Example
Using the previous example, the forward and backward pass calculations have been
performed. Let’s break it down:
1. Forward Pass:

61
 Activity ES EF
A 0 3
B 3 5
C 3 7
D 5 7
E 7 8
2. Backward Pass:
Activity LS LF
A 0 3
B 3 5
C 3 7
D 5 7
E 7 8

6.4 Applications of CPM in Project Management

Here are the various ways in which the CPM technique is used widely in project
management: This is an important tool that enables project managers to keep track of
work schedules, timelines, costs and any other factors that may affect the completion
of a project. Here are some key applications:
1. Construction Projects
CPM is widely used in construction to plan and schedule large projects. If the critical
path is known, a project manager is able to optimally allocate resources, schedule
subcontractors, or complete tasks on time. For example, in the construction of a high-
rise building, activities such as laying the foundation, framing, and finishing are
closely linked and require coordination. It is effective in determination of the right
sequence for the execution of these tasks.
2. Manufacturing
In manufacturing, CPM assists in managing the manufacturing processes, managing
maintenance, and the creation of a new product. The critical path helps identify areas
where manufacturers might be experiencing some amount of delay and can help
reduce this kind of time wastage in production. For example, in a manufacturing line
manufacturing automobiles, crucial activities like engine fitting, painting, and final
checking follow the CPM in order to facilitate efficiency in production lines.

62
3. IT Projects
In information technology projects, CPM helps design the software, integrate it, and
put other structures of information technology in place. It is very useful in ensuring
that core activities such as coding, testing, and deployment of applications and
software are undertaken on time. For example, in the process of creating a new
software application, CPM assists in deciding when to code, test the application, and
train users in order to meet the date when the program must be released to the public.
4. Event Planning
Event planners apply CPM to facilitate the processes involved in arranging events,
including conferences, weddings, and business events. From determining the critical
path, it would be easier for planners to organize different activities, from booking
halls to ordering food and even inviting people to attend the event. For example,
during the organization of a corporate event, key activities such as selection of
location, arrangement of speakers, and mainstreaming of advertisements are strategic
and coordinated under CPM.
5. Research and Development
In R & D activities, CPM assists in managing the sequence of activities in
experimentations, development of prototypes and their testing. The critical path is
important since it helps the project managers know the activities that are important in
achieving the project objectives. For example, in pharmaceutical research, many
delicate tasks, such as clinical trials, gaining regulatory approval, and market
introduction, are controlled by CPM.

Real-World Example: Delhi Metro Project

The usage of CPM is clear in the Delhi Metro project, which is a mass rapid
transportation system implemented in India. Some of the challenges include the nature
of the construction work, the arrangements made with other parties to the project, and
the pressure of time constraints. Through applying CPM, the project managers were
able to determine which is the most crucial activity in the project, how to allocate
required resources and when different phases of constructing the metro should be
completed. This then led to the success of the metro system, which led to efficient
transport in the urban centre of Delhi.

Benefits of CPM
63
CPM offers several benefits to project managers and organisations, including:
1. Improved Scheduling: CPM provides a timeline for project tasks, helps
managers to schedule activities efficiently.
2. Resource Allocation: Resource allocation helps identify resource needs and
allocate them optimally to critical tasks.
3. Risk Management: After identifying the critical path, project managers focus
on high-risk tasks and implement various risk management strategies.
4. Project Control: CPM allows better monitoring and control of the progress of
the project, by ensuring tasks are completed on time.
5. Decision-making: It holds significant information that assists managers in
decision-making, for example, shifting agendas or altering resources to handle
such complications.
Limitations of CPM
While CPM is a powerful tool, it has certain limitations:
1. Complexity: CPM becomes very complex for large projects due to tasks and
dependencies.
2. Static Nature: In static nature, activity duration is assumed to be fixed.
3. Resource Constraints: CPM does not take into account resources that may be
needed for a project and hence have impacts on the sequencing.
4. Focus on Time: It’s important to understand that cost, quality, scope, and time
are all interconnected and that CPM principally centres on time only.

Overcoming Limitations
To overcome the limitations of CPM, project managers can use the following
strategies:
1. Regular Updates: Make sure to maintain the project schedule and make
necessary adjustments as the project progresses, such as changes in time
estimations and dependencies.
2. Resource Levelling: Resource levelling techniques should be used in the plan
to respond to resource issues effectively and allocate resources properly.
3. Incorporating Uncertainty: Apply probability techniques like PERT where
one can assess the uncertainties in each activity’s duration.

64
 4. Integrated Approach: Integrate CPM with other project management tools,
like cost and quality assurance, for optimal results.

 Knowledge Check 2
State True or False.
1. The forward pass calculation starts from the end of the project and moves to
the beginning. (False)
2. The critical path method helps identify the most crucial tasks that need to be
completed on time to avoid delays in the entire project. (True)
3. CPM is not useful in managing construction projects. (False)
4. One of the applications of CPM is in event planning, where it helps organize
and manage events efficiently. (True)

 Outcome-Based Activity 2
List five real-world projects where CPM can be applied and explain briefly why it
would be useful in each project.

6.6 Summary
 The Critical Path Method (CPM) is a project management tool designed to
identify the sequence of tasks that impact the project's overall duration. Developed
in the late 1950s by DuPont, CPM helps in managing complex projects by
highlighting crucial activities.
 A project network diagram consisting of nodes (tasks) and arrows (dependencies),
is used in CPM. The primary goal is to determine the longest path through the
network, known as the critical path, which dictates the shortest possible project
completion time.
 Determining the critical path involves listing all project activities, identifying
dependencies, creating a network diagram, and estimating the duration of each
activity. The forward and backward pass calculations help identify the earliest
start, earliest finish, latest start, and latest finish times for each task.
 In critical path analysis, it is crucial to define all the activities within a project,
establish relations between these activities, draw a network diagram, and assess
the time needed to accomplish each activity. Calculation in forward pass and

65
 backward pass helps determine the original start and end, as well as the latest start
and end for each particular activity.
 The steps involved in identifying the critical path include defining all activities
within the project, defining dependencies, drawing a network diagram, and then
defining the time required for every activity. Calculation in forward pass and
backward pass helps determine the original start and end, as well as the latest start
and end for each particular activity.
 The backward pass calculation starts from the project’s end and moves to the
beginning to determine the latest start (LS) and latest finish (LF) times. This
ensures that each activity is completed in time to avoid delaying the project.
 CPM is widely used in industries like construction, manufacturing, IT, and event
planning. It helps in resource allocation, scheduling, and monitoring project
progress to ensure timely completion and optimal use of resources.
 The method also helps in the management of risks by implementing a method
that shows which tasks are risky, and the management will be in a position to
focus more on risky activities. This serves to provide ways of solving problems
and putting measures that may prevent likely delays in the implementation of
strategies.

6.7 Keywords
 Critical Path Method (CPM): A project management tool used to determine the
sequence of tasks that directly affect the project completion time.
 Critical Path: The longest path through a project network diagram, determining
the shortest possible project duration.
 Earliest Start (ES): The earliest time an activity can begin without delaying the
project.
 Latest Finish (LF): The latest time an activity can be completed without delaying
the project.
 Slack: The amount of time that a task can be delayed without causing a delay to
the project completion date.

6.8 Self-Assessment Questions

66
1. What are the key steps involved in determining the critical path in a project
network?
2. How do forward-pass and backward pass calculations help identify the critical
path?
3. Discuss the applications of CPM in different industries with examples.
4. Explain the benefits of using CPM in project management.
5. Identify and explain the limitations of CPM.

6.9 References / Reference Reading

 Chandra, Prasanna. Projects: Planning, Analysis, Selection, Financing,
Implementation, and Review. 8th ed., McGraw Hill Education, 2019.
 Agrawal, Ravi K. Project Management: The Managerial Process. 6th ed., Tata
McGraw Hill Education, 2018.
 Nagarajan, K. Project Management. 8th ed., New Age International Publishers,
2020.
 Kerzner, Harold. Project Management: A Systems Approach to Planning,
Scheduling, and Controlling. 12th ed., Wiley, 2017.
 Pinto, Jeffrey K. Project Management: Achieving Competitive Advantage. 5th ed.,
Pearson Education, 2019.

67
 Unit 7: Inventory Control

Learning Outcomes:
 Students will be able to understand the classification and types of inventory
control.
 Students will be able to explain the Economic Order Quantity (EOQ) model and
its significance.
 Students will be able to describe various inventory control systems.
 Students will be able to perform ABC analysis for inventory management.
 Students will be able to discuss the advantages of the EOQ model in management.

Structure:
7.1 Classification of Inventory Control
7.2 EOQ (Economic Order Quantity) Model
7.3 Inventory Control Systems
 Knowledge Check 1
 Outcome-Based Activity 1
7.4 ABC Analysis
7.5 Advantages of EOQ Model in Management
7.6 Just-In-Time (JIT) Inventory Systems
 Knowledge Check 2
 Outcome-Based Activity 2
7.7 Summary
7.8 Keywords
7.9 Self-Assessment Questions
7.10 References / Reference Reading

68
7.1 Classification of Inventory Control
Inventory control is crucial for managing stock efficiently and ensuring that a
business has the right products in the right quantity at the right time. Inventory can be
classified based on various criteria, including:
Raw Materials
Raw materials are the basic inputs required for the production of goods. Proper
control ensures uninterrupted production and cost efficiency. For example, a car
manufacturing company needs to manage its inventory of steel, glass, and rubber to
ensure continuous production. If the raw materials are not adequately controlled, the
production line can halt, causing delays and increased costs.
Effective management of raw materials involves:
 Forecasting: Accurate demand forecasting helps in planning the procurement
of raw materials.
 Supplier Management: Developing strong relationships with reliable
suppliers ensures a steady supply of raw materials.
 Inventory Levels: Maintaining optimal levels of raw materials prevents
production delays and reduces holding costs.
Work-in-Progress (WIP)
WIP inventory includes items that are in the production process but not yet finished.
Effective control reduces production time and bottlenecks. For example, in a furniture
manufacturing company, partially assembled chairs and tables represent WIP
inventory. Controlling WIP helps identify and eliminate production bottlenecks,
ensuring a smooth flow of production.
Key strategies for managing WIP inventory include:
 Process Optimization: Streamlining production processes to reduce WIP
levels and improve efficiency.
 Bottleneck Identification: Identifying and addressing bottlenecks in the
production process to ensure a smooth flow of work.
 Lean Manufacturing: Implementing lean manufacturing principles to reduce
WIP and increase productivity.
Finished Goods
These are completed products ready for sale. Controlling finished goods inventory
helps in meeting customer demand without overproduction. For example, an
electronics retailer needs to manage its stock of smartphones and laptops to ensure

69
availability for customers without overstocking, which ties up capital and increases
storage costs.
Managing finished goods inventory involves:
 Demand Forecasting: Accurate forecasting helps in planning the production
and inventory levels of finished goods.
 Inventory Turnover: Monitoring inventory turnover rates to ensure products
are sold before they become obsolete.
 Customer Demand: Aligning inventory levels with customer demand to
avoid stockouts and overstock situations.
Maintenance, Repair, and Operations (MRO) Goods
MRO goods are items used in the production process but not part of the final product.
Proper control ensures operational efficiency and reduces downtime. For example,
lubricants, cleaning supplies, and tools in a manufacturing plant are MRO goods.
Managing MRO inventory ensures that production equipment is well-maintained and
operational, preventing costly breakdowns.
Effective MRO inventory management includes:
 Regular Audits: Conduct regular audits to ensure MRO items are available
and in good condition.
 Vendor Relationships: Establishing strong relationships with MRO suppliers
for timely replenishment.
 Stock Levels: Maintaining optimal stock levels to avoid production delays and
equipment downtime.

7.2 EOQ (Economic Order Quantity) Model

The Economic Order Quantity (EOQ) model is a fundamental principle in inventory
management. It determines the optimal order quantity that minimizes the total cost of
inventory, including ordering and holding costs.
Definition and Formula
The EOQ formula is:

70
 Knowledge Check 1
Fill in the Blanks.
1. Raw materials are the basic inputs required for the production of goods.
Proper control ensures ______________ production and cost efficiency.
(uninterrupted)
2. The EOQ formula helps businesses determine the most economical order
quantity by balancing __________ and holding costs. (ordering)
3. A _______________ inventory system continuously updates inventory records
with each transaction, providing real-time inventory data. (perpetual)
4. The Just-In-Time (JIT) system aims to reduce inventory levels by ordering
stock only when needed for ______________ or sales. (production)

 Outcome-Based Activity 1
List three examples of raw materials, work-in-progress items, and finished goods
from a company of your choice and discuss how they might be managed.

7.4 ABC Analysis

ABC analysis is an inventory control technique that divides items into three categories
based on their importance:
 Category A
Assets with the greatest worth common characteristics are that they account for a
large part of the inventory cost, but only a few. For example, in a pharmaceutical
company, the drugs that belong to Category A are specialized treatments for cancer,
and only limited access to them is needed.
Characteristics of Category A Items
 High Value: These category items represent a significant portion of the total
inventory value.
 Low Quantity: These category item make up a small percentage of the total
number of items in inventory.
 Critical for Operations: Category A items are crucial for the company's
operations and revenue generation.

71
Management Strategies for Category A Items
 Frequent Review: Category A shall also be checked frequently, and the
reorder points will be evaluated suitably.
 Tight Control: Restriction should be instituted in the handling of the
inventory by making sure the items are sensitive to storage and access.
 Supplier Relationships: Build trust with reliable suppliers to have consistent
access to high learners-value items.
 Category B
This is the second set of inventory that has a value that is higher than that of Category
C but lower than that of Category A products. Its value contributes to the total cost of
the inventory, except that it requires relatively moderate attention and control. From
the Hardware Store, tangible mid-range products such as power tools could fall under
this category.
Characteristics of Category B Items
 Moderate Value: These items represent a moderate portion of the total
inventory value.
 Moderate Quantity: They make up a significant percentage of the total
number of items in inventory.
 Important for Operations: Category B items are vital in the project but they
are not as prioritized as we do with Category A items.
Management Strategies for Category B Items
 Periodic Review: Check and monitor Category B inventory and reorder points
on a regular but less than Category A items..
 Moderate Control: Incorporate Safety Measures within cost Effective control
for Inventory Management to avoid incurring high costs in managing the
inventories..
 Supplier Collaboration: Maintain close relationships with suppliers of goods
and services used in the organization to order products early enough and meet
the right quality.
 Category C
Quantities include the smallest cost, which contains the majority of the inventory
items but contributes a small fraction of the costs. For example, a pen that is not worth

72
more than ten bucks might be in Category C, which means that managers would have
to make little or no effort to control it.
Characteristics of Category C Items
 Low Value: These items represent a small portion of the total inventory value.
 High Quantity: They make up the majority of the total number of items in
inventory.
 Low Impact on Operations: Category C items are non-essential items that
have no significance in operations and do not significantly affect the revenue-
generating capacity of the company.
Management Strategies for Category C Items
 Infrequent Review: The order frequency of Category C items should be less
frequent, with emphasis made on cost savings, including inventory levels and
re-order points.
 Minimal Control: Implement minimal control measures to reduce costs, such
as bulk ordering and less frequent stock checks.
 Supplier Coordination: Integrate with suppliers to get optimal pricing for
low volumes and frequent replenishments of ‘staple’ items.
Benefits of ABC Analysis
 Prioritization: It assists managers in allocating their efforts most effectively
since it considers valuable items.
 Optimization: Helps in optimizing stock levels and reducing holding costs.
 Decision-Making: Improves decision-making regarding inventory
investments and stock replenishment.
 Efficiency: Improves general supplemental inventory operations by sorting
out and classifying products based on their value.

7.5 Advantages of EOQ Model in Management

The EOQ model offers several advantages for inventory management:
Cost Reduction
Through the identification of the most convenient quantity to order, total inventory
costs, which include holding costs as well as ordering costs, can be lowered. For
example, using EOQ to find the reordering quantity that optimizes ordering and
carrying costs, a retail store will benefit greatly.

73
Improved Cash Flow
Through EOQ, the various products can be ordered in the right quantities, which helps
to avoid situations whereby the company is held back by too much inventory, hence
enhancing liquidity. For example, a manufacturing firm has to determine the optimum
level of inventory in order to meet its needs while not having to use money for other
necessities or investments through the EOQ model.
Enhanced Decision-Making
EOQ is a powerful tool that guides customers in choosing the right amount and
frequency of orders to provide to their suppliers, thus achieving better control over
resources. For example, a distribution company can apply the EOQ model to
determine the right order frequency in order to meet its inventory needs and avoid the
expenses of having too much or too little stock.
Inventory Optimization
It contributes to achieving the right stock environment by ensuring that no firm is
faced with stock out or stuck in situations. For example, an organization such as a
pharmaceutical company needs to ensure that medicines are replenished regularly but
does not want to overstock because this could lead to the expiry of drugs.
Simplification of Inventory Management
The EOQ model takes up the challenge of simplification by seeking to provide a
simple formula for the determination of the optimal order quantity. This simplifies the
choices of inventories, thus enabling managers to address other factors when
managing the business.
Reduction of Stockouts and Overstocks
The EOQ model helps a business avoid excess stockouts and overstocks that may be
fatal to the company’s profits. Excess inventory, which results in stockouts, may
result in lost sales, which may cause dissatisfaction among the customers. In contrast,
overstock keeps so much cash tied up that it increases the holding cost. The EOQ
model effectively provides a balance between the value of inventory level in an
organization.
Improved Supplier Relationships
The EOQ model helps a business avoid excess of stockouts and overstocks that may
be fatal to the company’s profits. Excess inventory which results to stockouts may
result to lost sales besides which may cause dissatisfaction to the customers whereas
overstocks keeps so much cash tied up this increases the holding cost. The EOQ

74
model effectively provides a balance between the value of inventory level in an
organization..
Adaptability to Changes
Static assumptions are used in the EOQ model; however, they can be modified and
used to involve dynamic aspects by taking into consideration changes in demand,
costs of the order, and the holding cost. By constantly updating the inputs fed into the
EOQ equation, transport cost is maintained at the lowest required level depending on
the new market conditions.

7.6 Just-In-Time (JIT) Inventory Systems

A just-in-time system is a technique that pulls raw materials from suppliers in direct
relation to production requirements. The purpose of this process is to increase
efficiency overall and control excess inventory use by only accepting deliveries as
necessary.
Principles of JIT
 Eliminate Waste: JIT concentrate on reducing excess inventory, which can
lead to waste in terms of storage costs, obsolescence, and spoilage.
 Improve Quality: By sorting out defects immediately and after sorting
defects, encouraging continuous improvement, JIT enhances product quality
and reduces rework and scrap.
 Enhance Efficiency: Streamlining production processes and reducing setup
times results in more efficient operations and lower costs.
Benefits of JIT
 Reduces Inventory Holding Costs: By minimizing the cost of inventory at
any given time, JIT significantly reduces storage costs.
 Decreases Waste: JIT leads to less waste in terms of materials, time, and
effort, overall contributing to cost savings.
 Improves Quality: Continuous improvement and quick feedback on defects
result in higher-quality of the products.
 Increases Flexibility: JIT allows organisations to respond immediately to
changes in customer demand and market conditions.
Challenges of JIT
 Accurate Demand Forecasting: JIT needs precise demand forecasting to
ensure that materials arrive just in time for production.

75
  Supplier Reliability: JIT depend on suppliers who can deliver quality
materials on time.
 Supply Chain Disruptions: Any disruption in the supply chain management
system, such as natural disasters or transportation issues, can significantly
affect JIT operations.
Implementing JIT
Implementing JIT requires careful planning and coordination across the supply chain.
Key steps include:
 Supplier Collaboration: Forming strong partnerships with suppliers to ensure
delivery of the materials on time.
 Process Streamlining: Optimisation of production processes to reduce setup
times and increase efficiency.
 Employee Training: providing Training to employees on JIT principles and
practices results in smooth implementation of JIT.
 Technology Integration: Applying technology to enable the swift response of
questions from suppliers and or responding to their questions and ensuring
inventory advisories are met.
Real-World Examples of JIT
 Toyota: Toyota is one of the primary adopters of JIT, wherein it has
implemented this system in order to control inventory fees as well as optimize
production. Due to the effectiveness of the use of JIT by the company, many
other organizations have also copied the strategy.
 Dell: In the same way, Dell adopts the concept that inventories of most
computer components are obtained as needed, making it possible to deliver
value by promptly offering customized products without holding large
inventories.

 Knowledge Check 2
State True or False.
1. ABC analysis divides inventory items into three categories based on their
importance. [True]
2. The EOQ model assumes variable demand and lead times. [False]
3. Just-In-Time (JIT) inventory systems rely on precise demand forecasting and
dependable suppliers. [True]
76
 4. Vendor-Managed Inventory (VMI) increases the administrative workload for
the customer. [False]

 Outcome-Based Activity 2
Conduct an ABC analysis on a list of 20 inventory items from your home or a
small business, categorizing them into A, B, and C items based on their value and
importance.

7.6 Summary
 Inventory management is the administration of inventory levels and the flow of
goods to ensure the availability of the right stock of products. Raw material, work-
in-progress inventory, finished goods inventory, and material management
resources are the categories that can be considered. The management of every
category is imperative to enhancing production and overall operations.
 Some of the benefits of inventory control include reduced costs through avoidance
of production holds due to the expiration of raw materials and enhancement of
supply chain processes by maintaining the inventory at the required optimum
levels. Classification thus proves great value by directing attention to where
resources should be devoted, especially in relation to inventory.
 The EOQ model aims to find the right order quantity that helps minimise the total
inventory costs, which comprise ordering and holding. This position ensures it is
able to regulate the frequency of orders with the holding of inventory, ensuring
costs are incurred efficiently.
 It is possible to get the optimal order quantity by using the EOQ formula so that
businesses can avoid having more inventory than required and work on their cash
flow. The model also assists in controlling the inventory turnover rate in order to
avoid holding too much stock that may not be sold or buying too little that may
not be adequate to meet the needs of the business, etc.
 Continual inventory systems update the inventory control records with each
transaction and provide actual data. Some of these systems improve inventory
accuracy in addition to helping with the demand for accurate forecasts.
 Periodic inventory involves taking stock periodically so that, at some point in
time, an accurate value of the inventory is determined. While constructing the

77
 configurations might be easier, they may result in discrepancies between counts
and a stock-out or overstock issue.
 Categorizing inventory into three groups (A, B, C) based on their usage in value
and importance of organization ABC analysis assists in managing inventory
efficiently. Class A items are those that are valuable or sensitive and should be
monitored and reviewed more often.”
 Class B items have moderate importance and require moderate supervision and
security; Class C items are unimportant but numerous and require little or no
security measures. It is crucial since it directs the efforts to stock essential items
with the aim of causing the greatest positive impact.
 The EOQ model plays the role of minimizing total inventory costs through the
establishment of the most appropriate order quantity that cuts across ordering and
holding costs. This optimization enhances cash flow and makes and controls for
the appropriate use of resources.
 Regarding coordinative benefits, the following is articulated: It makes it easy to
decide on the optimal inventory since it provides guidelines on what actions to
take. The model also greatly reduces the problem of stockouts/overstocks and
provides adequate inventory in the store.
 JIT systems ensure that raw materials needed to be used are ordered and matched
with the production cycle hence avoiding a lot of extra expenses on holding
inventory. This approach positively helps the company increase operational
efficiency in a sense that reduces the amount of idle inventory in the production
process as well as the quality of the products that are being produced.
 For JIT to be affected, it has to involve accurate forecasts on the demand for the
commodity and the availability of reliable suppliers. It boosts adaptability to
customers’ needs and firms ‘events through the effective execution of change-over
decisions.

7.7 Keywords
 EOQ (Economic Order Quantity): A formula used to determine the optimal
order quantity that minimizes the total inventory costs, including ordering and
holding costs.

78
 JIT (Just-In-Time): An inventory management system that aligns raw-material
orders from suppliers directly with production schedules to reduce holding costs
and waste.
 ABC Analysis: A categorization technique that divides inventory items into three
categories (A, B, and C) based on their importance, value, and frequency of use.
 Perpetual Inventory System: A system that continuously updates inventory
records with each transaction, providing real-time data on stock levels.
 WIP (Work-In-Progress): Items that are in the process of being manufactured
but are not yet completed products.

7.8 Self-Assessment Questions

1. What are the key differences between a perpetual inventory system and a periodic
inventory system?
2. How does the EOQ model help in minimizing inventory costs?
3. What are the main principles of the Just-In-Time (JIT) inventory system?
4. Describe the benefits and challenges of implementing a Vendor-Managed
Inventory (VMI) system.
5. How does ABC analysis help in inventory management?

7.9 References / Reference Reading

 Chopra, Sunil, and Peter Meindl. Supply Chain Management: Strategy, Planning,
and Operation. 7th ed., Pearson, 2019.
 Saxena, Pallavi. Inventory Management: A Practical Approach. 2nd ed., Himalaya
Publishing House, 2020.
 Sharma, S.C. Inventory Control and Management. S. Chand & Company, 2021.
 Bowersox, Donald J., and David J. Closs. Logistical Management: The Integrated
Supply Chain Process. McGraw-Hill Education, 2019.
 Mahadevan, B. Operations Management: Theory and Practice. 3rd ed., Pearson
Education India, 2021.

79
 Unit 8: Advanced Linear Programming Techniques

Learning Outcomes:
 Students will be able to understand the advanced techniques of the Simplex
Method.
 Students will be able to analyse the sensitivity of linear programming solutions.
 Students will be able to conduct post-optimality analysis to evaluate the impact of
changes.
 Students will be able to apply the Dual Simplex Method to solve linear
programming problems.

Structure:
8.1 Introduction to Advanced Linear Programming Techniques
8.2 Advanced Simplex Method Techniques
8.3 Sensitivity Analysis in Linear Programming
 Knowledge Check 1
 Outcome-Based Activity 1
8.4 Post-Optimality Analysis
8.5 Dual Simplex Method
 Knowledge Check 2
 Outcome-Based Activity 2
8.6 Summary
8.7 Keywords
8.8 Self-Assessment Questions
8.9 References / Reference Reading

80
8.1. Introduction to Advanced Linear Programming Techniques
Linear programming efficiency is crucial in operational research and management
science; it helps in the correct distribution of scarce resources to various activities. For
good results in solving linear programming problems, the Simplex Method has proved
to be very effective, but sometimes more advanced techniques are needed to apply in
complex cases. This chapter discusses these techniques, among others, such as the
advanced Simplex Method techniques, the sensitivity analysis, the post-optimality
analysis and the Dual Simplex Method.

8.2. Advanced Simplex Method Techniques

The simplex method is the widely used technique of solving linear programming
problems by George Dantzig in the year 1947. However, it becomes necessary to
apply some modifications to the standard Simplex Method as the problems become
large and complex. These modern approaches have been developed in an effort to
increase computational gains as well as tackle limitations in the problems in question
more effectively.
Revised Simplex Method
The Revised Simplex Method is enhanced from the standard Simplex Method, and it
maintains the features of the standard method. It applies operations on matrices to
ease the computations. This helps to minimize the number of computations to be done
particularly when dealing with large problems.
Definition
The beauty of the Revised Simplex Method is that it constantly preserves a reduced
tableau while only having the simplified parts of a complete tableau. This method is
efficient in operation since fewer calculations are carried out using this method as
opposed to the larger matrix.
Process
1. Initialization: In this step, an initial feasible solution is represented in a
compact form.
2. Iteration: Utilise matrix operations to update the basis, the solution, and the
objective function value.
3. Optimality Check: check whether the current solution is optimal. If not,
proceed to the next iteration.

81
The idea of the Revised Simplex method is to not store the entire tableau at once,
which can take a lot of time and memory for larger problems. However, it maintains
the basis matrix and its inverse and updates these matrices at each iteration. It also has
the added advantage of ensuring significantly less memory is used to store the tableau
compared to what might have been used otherwise.
Bounded Variable Simplex Method
The Bounded Variable Simplex Method utilise linear programming problems with
variables that have explicit upper and lower bounds.
Definition
This method is an extension of the Simplex Method, which incorporates bounds on
the decision of variables directly into the iterative process.
Process
1. Initialization: The process starts with obtaining a feasible solution with
respect to the bounds of the variables.
2. Iteration: Find a better solution ensuring that the variable bounds are not
violated.
3. Optimality Check: Verify the optimal solution. If not, adjust the variables
within their bounds and repeat the process.
The Bounded Variable Simplex Method is an extension of the Simplex Method with
changes to the Simplex tableau to accommodate upper and lower bounds on the
variables of interest. This method is particularly useful in real-life situations where,
due to certain constraints in the problem, either in terms of resources or in any other
way, there are certain constraints on the decision variables.
Upper Bound Technique
The method of the Upper Bound Technique is used for solving the problem when
variables in linear programming are restricted by upper bounds.
Definition
Upper bound technic involves modifying the Simplex tableau to account for upper
bounds, enabling efficient handling of constraints.
Process
1. Initialization: Obtain an initial feasible solution.
2. Iteration: Find the number by modifying the tableau to include upper bounds
and perform pivot operations.

82
 3. Optimality Check: Check whether the current solution meets the optimal
conditions by considering the upper bounds.
The Upper Bound Technique is another related technique of optimization which
performs similar to the Bounded Variable complexity method with emphasis on the
upper bounds. This method ensures that the solution process achieves the maximum
or does not exceed the upper limits within the variables, which is very important in
most real life problems due to constraints such as resources or capacities.

8.3. Sensitivity Analysis in Linear Programming

Sensitivity analysis studies are a means of understanding how adjustments to the
coefficients of the articles in a linear programming model influence the initial optimal
solution. Such analysis is heavily required for decision-making to know the stability
of the solutions that are provided and any future changes that may occur in the
problem’s context.
Importance of Sensitivity Analysis
Sensitivity analysis plays an important part in the evaluation of how changes in
objective function coefficients, right-hand side (RHS) values of constraints and
almost all other parameters are likely to alter the optimal solution. It helps in
indicating how sensitive the solution is to changes to the input data.
 Key Concepts
Range of Optimality
The range of optimality specifies the degree to which an objective function coefficient
can vary for the current optimal solution to become non-optimal. This range is useful
for evaluating the stability of the optimal solution with respect to changing the
criterial function.
Range of Feasibility
The range of feasibility shows how much a right-hand side (RHS) constraint can shift
from its current position before the current solution becomes infeasible. It assists in
giving a perspective of the amount of flexibility available in the solution based on the
variations in constraints.

83
 Conducting Sensitivity Analysis
Objective Function Coefficients
 Determine Sensitivity: Determine the change in the objective function
coefficients which impact the optimal solution.
 Calculate Ranges: find the range in which the current solution remains
optimal.
To find the sensitivity of the objective function coefficients, consider the following
steps:
1. Identify the current optimal solution.
2. Perturb the coefficients: determine the change in the coefficients one at a
time.
3. Evaluate the impact: Determine whether the current solution is still the
optimal one from the current set of changed coefficients.
4. Determine the range: Determine suitable bounds on the coefficients so that
deviation from the optimum solution is negligible.
RHS Constraints
 Determine Sensitivity: Determine how variations in the RHS values of
constraints affect the feasibility of the current solution.
 Calculate Ranges: Develop the range within which the current solution
remains feasible.
To determine the sensitivity of the RHS constraints, follow these steps:
1. Identify the current feasible solution.
2. Perturb the RHS values: Change the RHS values one at a time.
3. Evaluate the impact: Check whether the current solution remains feasible or
not with the new RHS values.
4. Determine the range: Develop the range within which the RHS values can
vary without making the solution infeasible.

 Knowledge Check 1
Fill in the Blanks.
1. The Revised Simplex Method focuses on maintaining and updating a reduced
tableau, containing only the necessary parts of the full tableau. This method
improves efficiency by operating on a __________ matrix. (smaller)

84
 2. The Bounded Variable Simplex Method is an extension of the Simplex
Method that incorporates bounds on the __________ directly into the iterative
process. (decision variables)
3. The Upper Bound Technique modifies the Simplex tableau to include
__________ bounds and perform pivot operations. (upper)
4. In the Revised Simplex Method, the initialization step involves starting with
an initial __________ solution. (feasible)

 Outcome-Based Activity 1
Identify a real-life problem where the Bounded Variable Simplex Method can be
applied and describe the variables and their bounds.

8.4. Post-Optimality Analysis

The post-optimality analysis enables testing of the actual optimal solution whenever
there is a modification in the parameters of the linear programming option. This
undertaking evaluates sensitivity analysis to determine how the optimal solution is
affected by changes in various parameters.
Definition
Sensitivity analysis, or post-optimality analysis, is the assessment of the value of the
optimal solution of a linear programming problem after changes in the parameters of
the problem. It assists in perceiving the effects of these changes on the solution and
seeing if the solution is still reasonable and advisable.
Types of Changes
Changes in Objective Function Coefficients
 Impact on Solution: Evaluate how changes in the coefficients of the objective
function affect the optimal solution.
 Adjustment Strategies: Establish strategies to adjust the solution in response
to these changes.
When objective function coefficients alter, they will affect the optimality of the
solution. For example, where the cost of a raw material changes, this will require
achieving a new outlay, and the objective function requires alteration to reflect the
change before the optimal solution is recalculated to identify a new optimal
production mix.

85
Changes in RHS Constraints
 Impact on Feasibility: Determine how modifications in the RHS values affect
the feasibility of the solution.
 Adjustment Strategies: Formulate strategies to maintain feasibility with new
RHS values.
Alterations in RHS constraints can influence the optimality of other solutions, which
may also not be feasible. For example, when a resource is scarce, the RHS value of
the constraint that has this resource as a decision variable must be cut down, and the
solution must be examined to determine whether the current solution is feasible.
Practical Applications
An essential application of post-optimality analysis is in industries to verify that a
solution stays optimally acceptable even under conditions other than the optimal ones.
For example, it may be employed to redesign the production schedule of a
manufacturing firm to reflect alterations in the price or supply of raw materials. In the
field of supply chain management, post-optimality analysis allows for alterations to
transportation costs or delivery schedules, meaning that the supply chain is optimized
for a continued, efficient operation despite alterations to the cost of transportation.

8.5 Dual Simplex Method

The other technique used is the Dual Simplex Method, which is to be used in cases
where the solution is infeasible but is the optimum solution for the dual problem. This
belongs to the class of algorithms that are based on the principal that a feasible
solution can be made optimal as one progresses towards infeasibility.
Definition
The Dual Simplex Method continuously optimise an infeasible solution in an attempt
to reach an optimal feasible solution while remaining optimal for a given dual
problem. It is used most effectively when constraints have been added to an already
optimal solution, and there is a need to reoptimize and not begin from the start.
 Process
Initialization
Begin with an initial basic solution that is optimal for the dual problem but is integer
infeasible. The above solution can be arrived at by making some changes in the
solution of the original problem by changing those constraints which are not very
flexible.
86
Iteration
1. Pivot Operations: Perform the pivot operations to improve feasibility.
2. Update Basis: Adjust the basis to maintain optimality for the dual problem.
The Feasibility of the Solutions has an important place in the Dual Simplex Method
because the method underscores the modification of the basis and pivot operations.
All of them bring the solution nearer to feasibility and keep the duality gap of the dual
problem at its lowest.
Optimality Check
Check whether the current solution lies within the feasible region for the primal issue.
If not, a new iteration must be set wherein feasibility is met to achieve the objective of
the process. Optimality of the solution: If the solution is implementable in the context
of the primal problem in the modified formulation, it can also be implemented when
the original problem is considered.
 Applications
The Dual Simplex Method is most advantageous in practical situations where new
constraints are added to an optimal solution: that means we restart from that basically
optimal solution and just optimize afresh. For example, in project scheduling, if new
constraints arise from changes in project conditions, the Dual Simplex Method
enables one to re-schedule the project from a new mark instead of starting all over
with the first optimum solution.

 Knowledge Check 2
State True or False.
1. Post-optimality analysis involves evaluating the optimal solution after changes
have occurred in the parameters of the linear programming problem. (True)
2. Changes in the RHS constraints can affect the feasibility of the current
solution. (True)
3. The Dual Simplex Method iteratively improves a feasible solution until it
becomes optimal for the dual problem. (False)
4. Post-optimality analysis is not useful for understanding the stability of the
optimal solution. (False)

87
 Outcome-Based Activity 2
Explain the steps you would take to conduct post-optimality analysis for a given
linear programming problem.

8.6 Summary
 The Revised Simplex Method enhances computational efficiency by applying
matrix operations or manipulating and working only with a portion of the full
tableau, which is called the reduced tableau. This way of programming is more
efficient as it consumes less time and memory and thus can be used in cases of
very large problems.
 The Bounded Variable Simplex Method involves upper and lower bounds on the
decision variables into the iterations directly to guarantee that limits are not
violated by the variables during the steps of the solution process.
 Sensitivity analysis analyses how altering the parameters of a linear programming
problem – for example, the coefficients of the objective function of the RHS –
impacts the optimality of the solution. This analysis helps explicate how the
solution is safeguarded and the shifts that must be expected as part of the solution
strategy.
 Optimality range and feasibility range have become fundamental elements in
sensitivity analysis. They define how far the parameters may deviate from their
current values before the current solution can no longer be optimal or a solution to
the problem may be impossible for a different set of parameters, respectively.
 Sensitivity analysis is an analytical technique where the best solution reached by
the Linear programming model is analyzed after changes in problem parameters,
such as changes in objective function coefficients or RHS constraints. This
process is useful in relation to how these changes affect the solution.
 This analysis is used in various industries to enhance the cross-checking of
solutions to ensure that they are useful across different contexts and environments
so that companies can adjust their strategies and practices when needed due to
shifts in cost, capabilities, or market conditions.
 The Dual Simplex Method is applied when the non-basified solutions violate the
condition of feasibility while being optimal for the dual model. It again and again

88
 restores feasibility in the context of the primal problem while it assures optimality
in the context of the dual problem through pivot operations.
 The advantages of this method are most evident when new conditions to be met
are introduced to an optimum solution, in which case a new optimum must be
found without repeating much of the preceding computational work but rather
with modifications that can be made quickly and easily, allowing adaptation of an
organisational facility layout to the specifics of its intended usage in a timely and
efficient manner.

8.7 Keywords
 Simplex Method: A widely used algorithm for solving linear programming
problems by iterating through feasible solutions to find the optimal one.
 Revised Simplex Method: An efficient version of the Simplex Method that uses
matrix operations to streamline calculations.
 Bounded Variable Simplex Method: An extension of the Simplex Method that
includes bounds on the decision variables in the iterative process.
 Sensitivity Analysis: The study of how changes in the parameters of a linear
programming problem affect the optimal solution.
 Dual Simplex Method: A variant of the Simplex Method is used when the initial
solution is infeasible but optimal for the dual problem.

8.8 Self-Assessment Questions

1. What are the advantages of the Revised Simplex Method over the standard
Simplex Method?
2. How does the Bounded Variable Simplex Method handle variable constraints?
3. Explain the process and significance of sensitivity analysis in linear programming.
4. What is post-optimality analysis, and why is it important?
5. Describe the steps involved in the Dual Simplex Method.

8.9 References / Reference Reading

 Sharma, J. K. Operations Research: Theory and Applications. 6th ed., Macmillan
India, 2020.

89
 Taha, Hamdy A. Operations Research: An Introduction. 10th ed., Pearson
Education, 2017.
 Panneerselvam, R. Operations Research. 2nd ed., PHI Learning Pvt. Ltd., 2012.
 Gupta, Prem Kumar, and D. S. Hira. Operations Research. S. Chand & Company
Ltd., 2014.
 Kanti Swarup, P.K. Gupta, and Man Mohan. Operations Research. 15th ed., Sultan
Chand & Sons, 2018.

90
 Unit 9: Advanced Network Analysis Techniques

Learning Outcomes:
 Students will be able to understand advanced PERT techniques.
 Students will be able to analyse crash time and cost in project management.
 Students will be able to apply resource allocation and levelling strategies.
 Students will be able to implement advanced network analysis techniques in
complex projects.

Structure:
9.1 Advanced PERT Techniques
9.2 Crash Time and Cost Analysis
 Knowledge Check 1
 Outcome-Based Activity 1
9.3 Resource Allocation and Leveling
9.4 Applications of Advanced Network Analysis in Complex Projects
 Knowledge Check 2
 Outcome-Based Activity 2
9.5 Summary
9.6 Keywords
9.7 Self-Assessment Questions
9.8 References / Reference Reading

91
9.1 Advanced PERT Techniques
A statistical technique for project management, the Program Evaluation and Review
Technique (PERT), is intended to assess and illustrate the tasks necessary to finish a
project. When it comes to projects where it is difficult to anticipate the amount of time
required for particular tasks, PERT becomes more appropriate.
PERT employs three-time estimates:
 Optimistic Time (O): In the event that everything goes as smoothly as
possible, it is the least amount of time needed to finish a task.
 Most Likely Time (M): It is the best estimate of the time required to complete
a task, assuming everything proceeds as normal.
 Pessimistic Time (P): It is the maximum possible time required to complete a
task, assuming everything goes wrong.
The expected time (TE) for each task can be calculated using the formula:

This formula assists when it comes to providing an estimate of the time that would be
required to complete a task with due consideration to variability and uncertainties in
the process. This will offer a better average of the most likely time with more impact
and concern for the extremes.
Example: Let us call the software development task and find the optimistic time of 10
days the most likely time of 15 days, and the pessimistic time of 25 days. The
expected time would be calculated as:

Advanced PERT Techniques

Sophisticated PERT methodologies build upon the simple PERT concepts used for
PM and are useful for application in more complex cases. These techniques include:
 Multiple Precedence Relationships: Dealing with activities that involve
many previous and subsequent activities, thus the network structure is
complex. This makes it feasible and accurate for schedules to be adjusted as
the working processes progress.
 Probabilistic PERT: Including a probabilistic part and statistical distributions
to describe the systematic errors in the time estimates more accurately. This
can call for using monte carlo simulations to estimate the likelihood of the
project being done within a specific period.

92
  PERT with Resource Constraints: Adding constraints of resource
availability to the PERT analysis in order to correspond to resource
capabilities and capacity. This tends to go hand in hand with the time factor
but also takes extra care in assessing the kind of resources that are employed
as well as how they are utilized.
Example: In a construction project, some activities are likely to be closely linked or
overlap while others depend on resources such as skilled labour or machinery. At this
stage, the more refined PERT techniques can be applied in managing interlinked or
overlapping activities in order to identify the best time to obtain the required
resources and maximize the use of the limited resources. For example, suppose in a
construction company where certain items like a crane are critical to implementing
various projects in different places. The advanced PERT chart efficiently allows its
scheduling to avert any holds up.
Advantages and Limitations of PERT
Advantages:
 Helps in identifying the critical path and critical tasks: PERT focus on the
most critical tasks that need to be completed on time to avoid delaying the
project.
 Provides a more realistic project timeline by considering uncertainties:
With particular reference to how PERT addresses the problem of variability in
the time required to accomplish a given task, it is probably fair to conclude
that PERT is, indeed, the more realistic way of charting the project schedule.
 Facilitates better project planning and control: It can thus be seen that
PERT’s logical detail helps the project manager’s planning and control of the
project since this brings out which parts of the project are critical and hence
require more attention.
Limitations:
 Can be complex and time-consuming to implement: PERT can be
demanding because the time estimate collection process and the creation of the
PERT network are iterative activities.
 Relies heavily on accurate time estimates: However, if the time estimates
are not accurate, then this whole PERT analysis may not be entirely reflective
of the actual situation.

93
  It may not be suitable for all types of projects, especially those with highly
deterministic tasks: PERT is most useful for projects with uncertain task
durations and may not add much value to projects where tasks have fixed,
known durations.

9.2 Crash Time and Cost Analysis

Understanding Crash Time
Crash time refers to the shortest possible time in which a task or project can be
completed by deploying additional resources. It involves expediting certain activities
to reduce the overall project duration. Crashing a project often involves working
overtime, hiring additional personnel, or using more expensive equipment.
Cost-Time Trade-Off
The cost-time trade-off analysis is essential in determining the most cost-effective
way to reduce the project duration. The key concepts include:
 Normal Cost (NC): The cost associated with completing a task under normal
circumstances.
 Crash Cost (CC): The cost associated with completing a task in the shortest
possible time.
 Normal Time (NT): The usual time required to complete a task.
 Crash Time (CT): The shortest possible time to complete a task.
The cost slope, which indicates the cost per unit of time saved, is calculated as:

By properly examining the cost slope, the project manager is able to determine which
of the project’s activities should be crashed to the desired project duration at minimal
added cost.
Example: In software development, there is a critical module needed; to use extra
people to carry out their development faster, one needs to pay more, but the project
will be done in less time.

Steps in Crash Analysis

1. Identify the Critical Path: Understand that there is a set of tasks defining the
project time and identify it. The critical path is identified as the longest path

94
 within the project network, and each task on this network cannot be completed
beyond the authorized timeline in order to delay the project.
2. Calculate Cost Slopes: When determining the cost slope of each task on the
critical path with respect to other tasks, do the following: They help in getting
to know the most economic activities to crash.
3. Select Tasks to Crash: Select those tasks that have the lowest cost slopes to
crash initially before proceeding with other stages. This allows the project
duration to be minimized comfortably at the lower total incremental cost.
4. Update the Project Plan: The efficiency of the crashed tasks should guide the
alteration of the project schedule and costs. Making this involves resorting to
alterations in the timetables and costs.
5. Re-evaluate the Critical Path: Make sure that the critical path has been
successfully identified and updated; if necessary, go through another iteration
of the detailed procedure for selecting this path. The critical path may shift
when performing a crashing exercise on activities, as was shown in our
example.
Practical Considerations
 Resource Availability: Be certain to supplement the resources used and that
their supplementary use can be optimized. This checks on the availability of
all resources presuming for crashing the tasks, including personnel, equipment
and other materials.
 Impact on Quality: It is essential to take into account several scenarios
related to potential changes in the quality of work when it is decided to
accelerate some of the tasks performed. It is imperative to note that pressing
the undertakings can result in increased errors and lower quality of work.
 Stakeholder Approval: Gather the approval of the stake holders regarding
other additional costs that were likely to be incurred due to crashing. This is
important so as to justify the action of crashing to the stakeholders whenever
one intends to undertake it.

 Knowledge Check 1
Fill in the Blanks.
1. Crash time refers to the shortest possible time in which a task or project can be
completed by deploying additional __________. (resources)
95
 2. Advanced PERT techniques include Multiple Precedence Relationships,
Probabilistic PERT, and PERT with __________ constraints. (resource)

 Outcome-Based Activity 1
Calculate the expected time (TE) for a given task with optimistic, most likely, and
pessimistic time estimates.

9.3 Resource Allocation and Leveling

Introduction to Resource Allocation
Resource management refers to the process where resources that are available to be
used in completing tasks in a project are allocated to those tasks. It seeks to determine
how resources used on any particular project can be best utilised to achieve set goals.
Proper usage of resources can be useful in avoiding times when there are inadequate
resources to support a project’s progress or even when resources are scarce and are
being prioritized in their usage.
 Techniques of Resource Allocation
Resource Loading: Resource loading involves the allocation of those resources to
tasks that are most appropriate for that particular project at that specific time. It is
useful in determining activity periods that may receive heavy traffic and when there
may be constraints in terms of resources needed. As project managers are able to
visually look at the resource demand in relation to the timeline, they will be able to
deal with resource conflict issues, even before becoming a significant problem.
Example: In a marketing campaign, resource loading can guarantee that advertising
and graphic designers, writers and social media officers come in handy to develop and
release promotional tools.
Resource Leveling: Resource smoothing, also referred to as resource balancing, is
the process whereby the project timescale is engineered to ensure that of resources.
Some are pushed hard while others are made to relax. It seeks to avoid situations that
may lead to resource competition as well as facilitate a well-coordinated and effective
process. It often means that certain activities are scheduled to be postponed because it
would be undesirable for resources to be committed to many unrelated projects.

96
Example: Resource levelling in a construction project is a technique that helps to
allocate the work in the project’s time span in a way that will assist in using the same
resource on time when it is needed, mostly for use on a number of different tasks.
Steps in Resource Leveling
1. Identify Resource Constraints: Define resource type, its availability and
limitations. This is in regard to the number of resources and the capacity to
utilize them.
2. Analyze Task Dependencies: Different tasks require different types of
resources with varying characteristics and features. This helps pinpoint those
tasks that can be shelved, postponed, or otherwise delayed as per the general
project plan.
3. Adjust the Schedule: Replan the project duration if the assigned resources
have a conflicting schedule so that the resources used and the required plots
present an even combination. This may require re-scheduling of tasks and
twists in the project schedule to control resource utilization.
4. Monitor and Control: This means that constant resource control and
modification of any resources when deemed necessary should be conducted in
order to meet the project requirements. Resource levelling is always a time-
bound analysis that must be performed continuously throughout the course of
the given project.
Benefits and Challenges
Benefits:
 Prevents resource overloading and burnout: In other words, resource
levelling helps in the equitable distribution and management of resources in
that it prevents the over working of resources and fatigue within the team.
 Ensures optimal utilization of resources: Proper utilization of resources will
free up time that would otherwise be spent unnecessarily and minimize
expenditure on resources that may not generate any returns.
 Enhances project stability and reduces delays: By trying to even out
resource usage, resource levelling can assist in the avoidance of congestion
and schedule delay, leading to increased balance.
Challenges:

97
  Can lead to extended project durations: Duration may be prolonged if
adjustments are made to the schedule, moreover, for the purpose of optimizing
resource consumption.
 Requires detailed planning and continuous monitoring: Resource levelling
is a process that involves timely adjustment of resource consumption so as to
better manage the schedule to meet its completion demands.
 May necessitate changes in task priorities and dependencies: Resource
levelling may have implications that may suggest shifts in priority as well as
in the relationship between tasks.

9.4 Applications of Advanced Network Analysis in Complex Projects

Understanding Complex Projects
Defining the nature of complexity in elaborate projects, one should mention its vast
scope, the number of dependencies between the tasks that are implemented, and still,
the rather high level of unpredictability. Examples include infrastructural
development, major projects such as implementing enterprise resource planning
systems, and business research and development, among others. Such activities
usually span across the organization and its environment, require a lot of time for
preparation, and entail crucial measures to avoid or mitigate risks.

Importance of Advanced Network Analysis

Advanced network analysis techniques are crucial in managing complex projects as
they help in:
 Identifying Critical Paths: Making certain that the organisational calendar is
well aligned in relation to the projects. The critical path refers to the chain of
activities that prevents a project from being completed in less time than the
identified path, and, as such, its management is vital for ensuring completion
within the required timespan.
 Managing Uncertainties: Analyzing the ability to handle differences in tasks,
time, and resources. Even in the context of aspects that deal with uncertainties,
such as time constraints, there are more features that can be utilized, such as
probabilistic PERT, enabling the minimization of the risk of a delay.
 Optimizing Resource Utilization: Monitoring the points where future
resources should be allocated through the implementation of the project. Such

98
 techniques as resource levelling work to ensure that there is a proper
distribution of resources in relation to time so that certain activities do not
overtake other activities.
Example: In a metro rail construction project, the consequent management of
material and human logistics could benefit from the enhanced network analysis to
coordinate various contractors and suppliers together with various government
agencies to achieve the set project time line. With A PERT techniques and resource
levelling applied, project schedules, together with resource utilisation, can be
optimised in order to minimise schedule disruptions and escalating costs.

 Knowledge Check 2
State True or False.
1. Resource levelling often involves delaying non-critical tasks to ensure that
resources are not over-allocated. (True)
2. Complex projects are characterized by their small scale, few interdependent
tasks, and low degree of uncertainty. (False)
3. Advanced network analysis techniques are not useful in managing
uncertainties in project schedules. (False)
4. Future trends in network analysis include the integration of AI and real-time
data analytics. (True)

 Outcome-Based Activity 2
Identify a project you are familiar with and list the resources required. Perform a
simple resource levelling by adjusting the project schedule.

9.5 Summary
 Advanced PERT techniques incorporate multiple precedence relationships and
probabilistic approaches to handle uncertainties in project timelines more
effectively. These methods provide a more accurate and flexible project schedule.
 PERT, with resource constraints, integrates resource availability into the analysis,
ensuring that resources are allocated efficiently and potential bottlenecks are
avoided, optimizing overall project management.

99
 Crash time involves reducing project duration by deploying additional resources,
often at an increased cost. The cost-time trade-off is analyzed using the cost slope,
which helps identify the most cost-effective tasks to crash.
 The essential elapses in the crash procedure include the determination of the
critical path, computation of cost slope s, selection of available tasks to crash,
revision of the project plan and the final evaluation of the new critical path.
 Resource management makes sure that the available resources are properly
deployed in project tasks so that there is no scarcity and projects are run with the
most appropriate efficiency. Methods such as resource loading are used to display
the timeline of resource applications.
 Resource levelling alters the project schedule with a view of equalizing the
resource demanded with that available in a project so that there will not be a
situation where there are many tasks demanding the same resource and leads to
creating disturbance to the flow of work; this is done by either stretching the
duration of a particular activity or combining two or more activities so that they
can fit in the available resource rather than having many activities competing for
the same resource in the same.
 Complex networks also require sophisticated methods for controlling when it
comes to the project plan, avoiding potential bottlenecks or uncertainties, not to
mention the efficient use of resources. They help to manage timelines effectively
by achieving set goals within the target project time and properly utilizing
resources.
 Newer forms of network analysis in the future include using AI and machine
learning to develop prognostic information for network analysis, continual
surveillance to detect actionable information and improve decision-making, and
highly developed technological integration to foster teamwork and coordination to
increase the efficiency of projects.

9.6 Keywords
 PERT (Program Evaluation and Review Technique): A project management
tool used to plan, schedule, and control complex tasks by considering
uncertainties in task durations.

100
 Crash Time: The shortest possible time to complete a task by using additional
resources, often at an increased cost.
 Cost Slope: A measure of the cost increase per unit of time saved when a task is
crashed.
 Resource Allocation: The process of assigning available resources to tasks to
optimize their use.
 Resource Leveling: Adjusting the project schedule to balance resource demand
with availability, preventing overallocation and ensuring a smooth workflow.

9.7 Self-Assessment Questions

1. Explain the three-time estimates used in PERT and their significance.
2. Describe the steps involved in crash time and cost analysis.
3. What are the benefits and challenges of resource levelling in project management?
4. How can advanced network analysis techniques help in managing complex
projects?
5. Discuss the future trends in network analysis and their potential impact on project
management.

9.8 References / Reference Reading

 Kerzner, Harold. Project Management: A Systems Approach to Planning,
Scheduling, and Controlling. 12th ed., Wiley, 2017.
 Choudhury, S. Project Management. Tata McGraw-Hill Education, 2013.
 Prasanna Chandra. Projects: Planning, Analysis, Selection, Financing,
Implementation, and Review. 9th ed., McGraw Hill Education (India) Private
Limited, 2019.
 Panneerselvam, R., and P. Senthilkumar. Project Management. PHI Learning Pvt.
Ltd., 2019.
 Shtub, Avraham, Jonathan F. Bard, and Shlomo Globerson. Project Management:
Processes, Methodologies, and Economics. 3rd ed., Pearson Education India,
2015.

101
 Unit 10: Dynamic Programming

Learning Outcomes:
 Students will be able to explain the concept of dynamic programming.
 Students will be able to identify the various applications of dynamic
programming.
 Students will be able to apply dynamic programming techniques to solve
problems.
 Students will be able to analyse real-world scenarios where dynamic programming
is used.

Structure:
10.1 Introduction to Dynamic Programming
10.2 Applications of Dynamic Programming
 Knowledge Check 1
 Outcome-Based Activity 1
10.3 Solution Techniques
10.4 Real-World Applications of Dynamic Programming
 Knowledge Check 2
 Outcome-Based Activity 2
10.5 Summary
10.6 Keywords
10.7 Self-Assessment Questions
10.8 References / Reference Reading

102
10.1 Introduction to Dynamic Programming
Dynamic Programming (DP) is a method used in mathematics and computer science
to solve problems by breaking them down into simpler subproblems. It is particularly
useful for optimization problems where the goal is to find the best solution among
many possible options. The key idea is to store the results of subproblems to avoid
redundant computations, thus improving efficiency.
Problems with overlapping subproblems and optimal substructures can be solved
using dynamic programming. While optimal substructure indicates that the optimal
solution to a problem can be built from the optimal solutions to its subproblems,
overlapping subproblems refer to subproblems that are solved more than one time.
Definitions:
 Overlapping Subproblems: When a recursive algorithm revisits the same
problem repeatedly, these subproblems are shelp to overlap.
 Optimal Substructure: A problem has an optimal substructure if an optimal
solution can be constructed from optimal solutions of its subproblems.

10.2 Applications of Dynamic Programming

Dynamic Programming is widely used in various fields due to its efficiency and
effectiveness. Some common applications include:
Optimization Problems
Optimization problems aim to find the best solution from a set of feasible solutions.
Dynamic Programming is particularly effective for such problems because it reduces
the computation time by storing intermediate results. Examples include the shortest
path problem, knapsack problem, and resource allocation problems.
Example: Shortest Path Problem In the shortest path problem, we aim to find the
shortest path between two nodes in a graph. Dynamic Programming can be used in
algorithms like Floyd-Warshall to find the shortest paths between all pairs of nodes
efficiently.
Combinatorial Problems
Combinatorial problems involve finding the best combination of elements from a set
that satisfies certain criteria. Dynamic programming helps explore all possible
combinations systematically while avoiding redundant calculations. Examples include
the travelling salesman problem and bin packing problem.

103
Example: Travelling Salesman Problem (TSP) In TSP, a salesman must visit a set
of cities exactly once and return to the starting point, with the goal of minimizing the
total travel distance. Dynamic Programming can be used to solve TSP by breaking it
down into subproblems and storing the results of these subproblems.
Sequence Alignment
Sequence alignment is a critical problem in bioinformatics, where we aim to align
sequences of DNA, RNA, or proteins to identify regions of similarity. Dynamic
Programming is used to solve this problem efficiently, as seen in algorithms like
Needleman-Wunsch and Smith-Waterman.
Example: Needleman-Wunsch Algorithm This algorithm aligns two sequences by
breaking down the alignment problem into smaller subproblems and solving each one
to find the optimal alignment.

 Knowledge Check 1
Fill in the Blanks.
1. Dynamic Programming can be applied to problems that exhibit the properties
of _______ subproblems and optimal substructure. (overseeing)
2. The shortest path problem in graphs can be efficiently solved using algorithms
like ________ which employs dynamic programming techniques. (Floyd-
Warshall)
3. In sequence alignment, dynamic programming is used in the ________-
Wunsch algorithm to find the optimal alignment of sequences. (Needleman)
4. Combinatorial problems, like the travelling salesman problem, can be
effectively solved using dynamic programming by breaking them into
________. (subproblems)

 Outcome-Based Activity 1
Research and list three real-world problems in your field of study where dynamic
programming could be applied, and explain why it is suitable for these problems.

104
10.3 Solution Techniques
Dynamic Programming can be implemented using two main approaches: top-down
and bottom-up. Both approaches aim to solve the same problem, but they differ in
their methodology.
Top-Down Approach (Memoization)
In the top-down approach, we solve the problem recursively and store the results of
subproblems in a table (or memo). Whenever we need the result of a subproblem, we
first check if it has already been computed. If yes, we use the stored result; otherwise,
we compute it and store the result for future use.
Steps in the Top-Down Approach:
1. Identify the subproblems.
2. Recursively solve the subproblems.
3. Store the results of subproblems.
4. Use stored results to avoid redundant computations.
Example: Fibonacci Sequence The Fibonacci sequence is a classic example of
dynamic programming. By storing the results of previously computed Fibonacci
numbers, we avoid redundant calculations, thus reducing the time complexity from
exponential to linear.
Bottom-Up Approach (Tabulation)
In the bottom-up approach, we solve the problem iteratively by solving all the
subproblems first and using their results to solve the larger problem. This approach
typically involves filling up a table, starting from the simplest subproblems and
building up to the solution of the original problem.
Steps in the Bottom-Up Approach:
1. Identify the subproblems.
2. Order the subproblems.
3. Solve every subproblem iteratively and save the results.
4. Use the results of solved subproblems to build up to the final solution.

10.4 Real-World Applications of Dynamic Programming

Several real-life users access dynamic programming to find solutions across different
fields. In this section, we review some of the major domains of human activity in
which DP has a critical function.
Finance and Investment
105
In finance, dynamic programming is used for portfolio selection, options valuation,
and risk assessment. DP assists in the formulation of better investment decisions by
reducing the level of difficulty in the overall financial problems and turning them into
more solvable sub-problems.
Example: Portfolio Optimization Dynamic Programming is used in decision-
making, especially in investments. It helps identify the best holdings for a portfolio in
order to achieve the highest returns at minimum risks. The optimal solution by
evaluating various combinations of assets is made possible through DP, hence
offering an efficient outcome to the problem.
Operations Research
Many problems that can be encountered in logistics, supply chains, and production
scheduling use the dynamic programming approach in operations research. Overall,
DP increases operational efficiency through proper planning and distribution of
resources with regard to their availability and pertinent requirements.
Example: Production Scheduling In production scheduling, dynamic programming
serves the purpose of deciding on which activities should be done ahead of others to
make the production cost low and within the stipulated time. Therefore, splitting the
scheduling problem into sub-problems will be useful, and DP will be useful in coming
up with the solution.
Machine Learning and Artificial Intelligence
In machine learning and artificial intelligence, dynamic programming was used for
the training of algorithms and sequence prediction training, and it was also used in
reinforcement learning. Serving as a learning process optimizer, DP helps training
models in machine learning to achieve the intended performance.
Example: Reinforcement Learning Consequently, dynamic programming is
beneficial in solving the reinforcement learning problem in order to determine the
policy that will give the highest cumulative reward to an agent. DP establishes an
efficient method of finding out the actions that the agent needs to opt for by solving
the Bellman equations.
Robotics and Path Planning
In robotics planning and control, Dynamic Programming is sometimes used for path
planning. Using DP approaches, the overall path planning problem is divided into a
number of smaller sub-problems to identify the best possible path for a robot to
perform a certain task in a given environment.

106
Example: Path Planning for Robots In path planning, dynamic programming also
proves useful in that it makes it possible to determine the shortest way through which
a robot can get to a certain location, all the while avoiding congested areas. Since path
planning can be solved iteratively through subproblems, the DP strategy provides an
optimization solution to this problem.

 Knowledge Check 2
State True or False.
1. The top-down approach in dynamic programming is also known as
memoization. (True)
2. Dynamic Programming is rarely used in finance and investment for portfolio
optimization. (False)
3. The bottom-up approach solves the problem iteratively by solving all the
subproblems first. (True)
4. Dynamic programming is not applicable to machine learning and artificial
intelligence. (False)

 Outcome-Based Activity 2
Identify a problem in robotics or path planning and outline the steps to solve it
using the bottom-up approach of dynamic programming.

10.5 Summary
 In a larger context, Dynamic Programming (DP) is a methodology for solving
problems based on dividing them into subproblems and storing the answers to
these subproblems to be reused multiple times or to avoid solving them in the
future. It is more useful where the given function is to be optimized and stems
from concepts such as overlapping subproblems and optimal substructure.
 DP can be implemented using two main approaches: The methods by which they
are solved are also two, that is, the top-down approach, also known as
memoization, and the bottom-up approach, also known as tabulation. The top-
down approach tackles problems of a higher level through recursion, saving the
result of previously solved sub-problems in the process, while the bottom-up
approach works though successive iterations of solving the simplest sub-problems.

107
 The Dynamic Programming concept is frequently used when solving optimization
problems like the shortest path problem, the knapsack problem, and a variety of
resource allocation problems. In it, they found out that it was able to save a lot of
computation time because it was unnecessary to continually compute the result
where the result can be stored and retrieved when required.
 In areas such as bioinformatics and many combinatorial problems such as
Sequence Alignment and travelling salesman, there is no way you can proceed
with them if DP is not used. The genetic algorithm covers all the solutions and
systematically tries to optimize the resources and time needed for the search.
 The approach of memoization, termed the top-down approach, involves solving
problems via recursion and then retaining the solutions to some of the sub-
problems in the table so that they can be easily retrieved when solving similar
problems. Another better way is that this method can be applied when having
subproblems that have solved problems with the Fibonacci sequence many times.
 The second one is a bottom-up approach, which we know as tabulation; this
involves filling a table with solutions to problems, starting with the simplest ones
to the most complex ones. It is applied in problems such as the longest common
consequence, where solutions to subproblems are incorporated in a way that
brings up the overall solution.
 In finance and investment, DP is applied in decisions on diversification, option
valuation, and assessment of risk. DP also assists in transforming big financial
issues into adjunct problems that can help in decisions on the best ways of
investing and risks to avoid.
 DP is applied in machine learning, artificial intelligence and robotic applications
for algorithm training, sequence forecasting, and path finding. It enhances
learning ops and offers gains in the performance of models to enhance path
planning and navigation in robots.

10.6 Keywords
 Overlapping Subproblems: When a problem can be broken down into smaller
subproblems that are reused multiple times.
 Optimal Substructure: A property where the optimal solution to a problem can
be constructed from the optimal solutions of its subproblems.

108
 Memoization: A top-down dynamic programming technique where results of
expensive function calls are cached and reused.
 Tabulation: A bottom-up dynamic programming technique where results are
stored in a table and built up to solve the overall problem.
 Portfolio Optimization: A process in finance where dynamic programming is
used to allocate assets in a portfolio to maximize returns and minimize risks.

10.7 Self-Assessment Questions

1. What are the key characteristics that make a problem suitable for dynamic
programming?
2. Explain the differences between the top-down and bottom-up approaches in
dynamic programming with examples.
3. How is dynamic programming applied to the shortest path problem in graphs?
4. Describe an application of dynamic programming in bioinformatics.
5. How does dynamic programming improve the efficiency of solving combinatorial
problems?

10.8 References / Reference Reading

 "Introduction to Algorithms" by Thomas H. Cormen, Charles E. Leiserson,
Ronald L. Rivest, and Clifford Stein. 4th ed., MIT Press, 2022.
 "Dynamic Programming and Optimal Control" by Dimitri P. Bertsekas. 4th ed.,
Athena Scientific, 2019.
 "Algorithms" by Robert Sedgewick and Kevin Wayne. 4th ed., Pearson, 2019.
 "Operations Research: An Introduction" by Taha H. A., 10th ed., Pearson India
Education Services, 2017.
 "Introduction to the Theory of Computation" by Michael Sipser. 3rd ed., Cengage
Learning, 2013.

109
 Unit 11: Game Theory

Learning Outcomes:
 Students will be able to define game theory and its key concepts.
 Students will be able to identify different types of games using game theory.
 Students will be able to apply solution concepts in game theory to practical
problems.
 Students will be able to analyse the application of game theory in various
management scenarios.
 Students will be able to explain the Nash Equilibrium and its significance in
strategic decision-making.

Structure:
11.1 Introduction to Game Theory
11.2 Types of Games
 Knowledge Check 1
 Outcome-Based Activity 1
11.3 Solution Concepts in Game Theory
11.4 Applications of Game Theory in Management
11.5 Nash Equilibrium and its Applications
 Knowledge Check 2
 Outcome-Based Activity 2
11.6 Summary
11.7 Keywords
11.8 Self-Assessment Questions
11.9 References / Reference Reading

110
11.1 Introduction to Game Theory
Game theory is a mathematical framework designed for understanding strategic
interactions among rational decision-makers. The core idea of game theory is to study
how individuals or groups make decisions that will, in turn, influence the actions of
others. It provides tools for predicting outcomes in situations where participants have
conflicting interests and their choices depend on the actions of others.
Definition of Game Theory
Game theory is the study of mathematical models of conflict and cooperation between
intelligent, rational decision-makers.
Historical Background
Game theory initially evolved in the early part of the 20th century when
mathematician John von Neumann and economist Oskar Morgenstern developed the
theme. Possible formalization of such a supposition was provided by their works,
specifically their book “Theory of Games and Economic Behaviour” published in
1944. John Nash had later on build on game theory through the introduction of Nash
Equilibrium, for which he captured the Nobel Prize in Economic Sciences.
Importance of Game Theory
The main practical areas in which game theory can be applied are economics, politics,
psychology, and management. It is useful in analyzing rivalry behaviours and may be
applied in strategy development for business, bidding, voting, and other areas.
Basic Concepts
 Players: The decision-makers in the game.
 Strategies: The plans of action available to the players.
 Payoffs: The outcomes received by the players as a result of the strategies
chosen.
 Games: The interactions between players, which can be cooperative or
competitive.

11.2 Types of Games

One classification system used in game theory divides them according to the number
of players, the sum total of the results, the sequence of actions, and the level of
information. It is, therefore, desirable to know these types to be able to apply the
proper model in a given set of circumstances.

111
Cooperative and Non-Cooperative Games
 Cooperative Games: Players can form binding agreements and coalitions.
The focus is on how groups of players can collaborate to achieve the best
outcomes.
 Non-Cooperative Games: No binding agreements are allowed, and each
player acts independently to maximize their own payoff.
Symmetric and Asymmetric Games
 Symmetric Games: In this type of game, the payoffs depend on the strategies
chosen, not on who is playing them. If two players swap their strategies, their
payoffs remain unchanged.
 Asymmetric Games: In this type of game, he payoffs depend on both the
strategies chosen and the players themselves. Players have different roles or
information, leading to different payoffs for the same strategies.
Zero-Sum and Non-Zero-Sum Games
 Zero-Sum Games: One player's gain is exactly equal to another player's loss.
The total payoff for all players is zero.
 Non-Zero-Sum Games: The total payoff for all players can be more or less
than zero, allowing for the possibility of mutual gain or mutual loss.
Simultaneous and Sequential Games
 Simultaneous Games: Players make their decisions at the same time, without
knowledge of the others' choices.
 Sequential Games: Players make decisions one after another, with each
player aware of the previous actions.
Perfect and Imperfect Information Games
 Perfect Information Games: All players have complete information about
the previous moves and the game structure.
 Imperfect Information Games: Players lack information about some aspects
of the game, such as the previous actions of others or certain payoffs.

 Knowledge Check 1
Fill in the Blanks.
1. Game theory is a mathematical framework designed for understanding
strategic interactions among ________. (rational decision-makers)

112
 2. The book "Theory of Games and Economic Behaviour" was authored by John
von Neumann and ________. (Oskar Morgenstern)
3. In ________ games, players can form binding agreements and coalitions.
(cooperative)
4. A zero-sum game is one in which one player's gain is exactly equal to another
player's ________. (loss)

 Outcome-Based Activity 1
Discuss with your peers and list three real-life examples of non-cooperative
games.

11.3 Solution Concepts in Game Theory

Solution concepts are methods used to predict the outcomes of games and to suggest
optimal strategies for the players. Several solution concepts help determine the best
course of action in strategic situations.
Dominant Strategy
A dominant strategy is a strategy that always provides a better outcome for a player,
no matter what the other players do. If one strategy is dominant, it is often the rational
choice for the player.
Nash Equilibrium
A Nash Equilibrium occurs when no player can improve their payoff by unilaterally
changing their strategy, given that the other players' strategies remain unchanged. It is
a stable state where players have no incentive to deviate.
Pareto Efficiency
A situation is Pareto efficient if no player can be made better off without making at
least one other player worse off. It represents an optimal allocation of resources where
improving one player's payoff means reducing another's.
Mixed Strategy Equilibrium
In some games, players may adopt mixed strategies, where they randomize over
possible moves. A mixed strategy equilibrium is a set of strategies where each player's
strategy is optimal, given the mixed strategies of the others.

113
Subgame Perfect Equilibrium
In sequential games, a subgame perfect equilibrium is a refinement of Nash
Equilibrium that requires players to play optimally at every stage of the game. It
eliminates non-credible threats and ensures consistency throughout the game.
Bayesian Equilibrium
In games with incomplete information, a Bayesian Equilibrium accounts for the
uncertainty players have about each other's types or payoffs. To the players,
probabilities turn into beliefs about the rest that are unknown to establish the best
possible ways of playing games.

11.4 Applications of Game Theory in Management

Game theory provides valuable insights and tools for addressing various strategic
issues in management. It helps managers to understand competitive dynamics,
negotiate better deals, and make informed decisions.
Business Negotiations
It also helps in the analysis of the negotiation processes where a number of parties
have conflicting objectives or interests but can ideally cooperate. Methods such as the
Nash Bargaining Solution give directions on how two parties can come to a mutually
suitable agreement.
Competitive Strategies
It is a tool that firms employ to understand and predict probable actions from their
competitors’ end while coming up with strategies that may give them an edge over
their competitors. There are different industrial organization models that seek to
explain how firms compete in terms of quantities and prices, including the Cournot
competition model and the Bertrand competition model.
Auctions and Bidding
Auctions are significant activities in the existence of game theory, and it is crucial to
plan and participate in auctions. It assists bidders in learning how to bid strategically
while helping the auctioneer create structures that will offer the most profits. For
example, the Vickrey auction guarantees the break-through deception completion by
the participants.
Market Entry and Exit
This is applied to finding out the best strategy to choose, either to venture into a
market or to pull out from it. Bargaining strategies look into a position where threats

114
can be made credible to potential entrants, while exit strategies take into account the
likely reactions of competitors and the market.
Pricing Strategies
Game theory is resourceful in the process of arriving at the appropriate price level that
firms must set in order to optimize their profits in the face of competitors’ likely
actions. Some promising research directions include price wars, strategies of
collusion, and price discrimination.
Supply Chain Management
Mutual choice theory helps in supply chain coordination whereby inventory policies,
supply choices, and contractual terms are enhanced. It assists the firms in bringing
together the bonus and ensuring counterpart cooperation in the chain of supply.
Voting and Collective Decision-Making
In organizations, game theory is applied to evaluate voting models and make
systematic group choices. It is useful in proposing sound and equitable systems and
procedures for voting and for ascertaining the nature of a coalition.
Corporate Governance
The theories of games offer solutions to work with different interests and
contradictions of shareholders, managers, and other participants. It assists in the
formulation and evaluation of incentives as well as the control systems in order to
enhance the conjunction of personnel interest and corporate governance.

11.5 Nash Equilibrium and its Applications

Game theory has a significant principle called Nash equilibrium, which was believed
to have been discovered by John Nash. It describes a situation where there is no
player who can gain by changing his/her strategy, knowing the anticipated strategy of
the rest of the players. This topic is applicable in most disciplines as it encompasses
the operation and functioning of the human body.
Understanding Nash Equilibrium
A Nash Equilibrium is defined as a condition where, for each participant, their
strategy represents the best possible choice achievable in light of the other
participant’s strategies. It is a situation where each individual on the field is striving to
perform at an optimum level regarding the plans of the opposing team.

115
Mathematical Representation
For a game with players i = 1, 2, …, n let si be the strategy chosen by player i. The set
of strategies (s1, s2, …, sn) forms a Nash Equilibrium if:
ui(si, s−i) ≥ ui (si′, s−i) in the strategy set of player i, where ui is the payoff function for
player i, and s−i denotes the strategies chosen by all players except i.
Applications in Economics
In law, the legal realists expounded on how legal processes arrived at promised
fruition in a commercial setting by applying game theory’s Nash Equilibrium, where
supply meets demand. It occurs in the models of oligopoly, with the specialization in
the price and quantity of firms and the analysis of public goods and externalities.
Applications in Politics
If you specialise in politics, then you’ll get to know that Nash Equilibrium utilizes
quantitative analysis for voting patterns, the formation of coalitions, and legislative
negotiations. It assists in explaining how executive and legislative branch participants
plan and operate in order to secure the best possible vote outcomes.
Applications in Business Strategy
Companies apply Nash Equilibrium in managerial decisions, especially on matters
involving price fixing, production, advertising, and introducing new products in the
market. Competitive strategy is beneficial for firms because it enables them to
anticipate what competitors will do and to develop good strategies, knowing what a
competitor is likely to do next.
Applications in Sports
Nash Equilibrium is applied in sports to analyze the pertinent approaches to conflict
appearing in competitive sports games. Coaches and players use it in strategy
formulation on how they could record a win given the available chances and the
corresponding strategies of the counterpart.
Applications in Evolutionary Biology
In evolutionary biology, Nash Equilibrium is helpful in understanding what kinds of
reaction patterns are selected by nature and taken up by species to avoid being
eliminated and to reproduce. It is mentioned in various models of animal behaviour
prediction, which assumes that animals fight for food and reproduction.

 Knowledge Check 2
State True or False.
116
 1. Nash Equilibrium occurs when no player can improve their payoff by
unilaterally changing their strategy. (True)
2. In cooperative games, no binding agreements are allowed, and each player
acts independently. (False)
3. The Prisoner's Dilemma is a classic example illustrating the concept of Pareto
efficiency. (False)
4. Game theory helps in analysing negotiation scenarios where parties have
conflicting interests but can benefit from cooperation. (True)

 Outcome-Based Activity 2
Analyze a business case study where Nash Equilibrium could be applied to
understand competitive strategies.

11.6 Summary
 Game theory is a mathematical framework for analysing strategic interactions
among rational decision-makers. It helps in predicting outcomes in scenarios
where participants' choices affect one another.
 Originating with John von Neumann and Oskar Morgenstern, game theory has
significant applications in economics, politics, psychology, and management,
helping in the formulation of competitive strategies and understanding cooperative
behaviours.
 Games can be categorized into cooperative and non-cooperative, symmetric and
asymmetric, zero-sum and non-zero-sum, simultaneous and sequential, and games
of perfect and imperfect information.
 Each type of game has unique characteristics that influence how players
strategize, such as the ability to form alliances in cooperative games or the lack of
complete information in imperfect information games.
 Solution concepts like dominant strategy, Nash Equilibrium, Pareto efficiency,
mixed strategy equilibrium, subgame perfect equilibrium, and Bayesian
equilibrium help in predicting optimal strategies and outcomes in games.
 These concepts provide frameworks for understanding strategic stability and
rational decision-making, ensuring that players adopt strategies that are best
responses to each other.

117
 It helps in business disputes, strategic management, contests, entry and exit
strategies, pricing approach, sourcing and supply chain, voting, and board of
directors.
 Using the concept of game theory in managerial decision-making enables the
managers to gain insight into the game, the strategies of their rivals, as well as the
best way to approach certain business transactions, increasing the potential
effectiveness of the results that are likely to be obtained.
 Nash Equilibrium can be described as a core concept that implies that no one of
the players can improve the value of his pay-off unilaterally by abandoning the
previous strategy that guarantees the stability of a situation in strategic relations.
 A great number of approaches are applied in economics, politics, business
strategies, sports, and even evolutionary biology, helping to recognize how
rational individuals in multiple competitions act and make it possible to predict
stable and stable results in various spheres.

11.7 Keywords
 Game Theory: The study of mathematical models of conflict and cooperation
between intelligent, rational decision-makers.
 Nash Equilibrium: A situation where no player can benefit from changing their
strategy while the other players' strategies remain unchanged.
 Zero-Sum Game: A type of game where one player's gain is exactly equal to
another player's loss.
 Dominant Strategy: A strategy that always provides a better outcome for a
player, no matter what the other players do.
 Pareto Efficiency: A situation where no player can be made better off without
making at least one other player worse off.

11.8 Self-Assessment Questions

1. Define game theory and explain its importance in strategic decision-making.
2. Differentiate between cooperative and non-cooperative games with examples.
3. What is Nash Equilibrium, and how is it significant in strategic interactions?
4. Explain the concept of a zero-sum game with a real-life example.
5. How does game theory apply to business negotiations and competitive strategies?

118
11.9 References / Reference Reading
 Basu, Kaushik. Lectures in Industrial Organization Theory. Oxford University
Press, 2020.
 Dixit, Avinash, and Susan Skeath. Games of Strategy. 4th ed., W.W. Norton &
Company, 2015.
 Pandey, I. M. Financial Management. 12th ed., Vikas Publishing House, 2019.
 Raghuram, G., and Rekha Jain. Principles of Infrastructure: An Introduction to
Critical Infrastructure Analysis and Decision-Making. 2nd ed., Tata McGraw-Hill
Education, 2018.
 Tirole, Jean. The Theory of Industrial Organization. MIT Press, 2019.

119
 Unit 12: Queuing Theory

Learning Outcomes:
 Students will be able to identify the fundamental principles of queuing theory.
 Students will be able to describe the characteristics and components of queuing
systems.
 Students will be able to analyse various models and applications of queuing
theory.
 Students will be able to evaluate performance measures in queuing systems.
 Students will be able to apply queuing theory concepts to real-world service
industry scenarios.

Structure:
12.1 Introduction to Queuing Theory
12.2 Characteristics of Queuing Systems
 Knowledge Check 1
 Outcome-Based Activity 1
12.3 Models and Applications of Queuing Theory
12.4 Performance Measures in Queuing Systems
12.5 Applications of Queuing Theory in Service Industries
 Knowledge Check 2
 Outcome-Based Activity 2
12.6 Summary
12.7 Keywords
12.8 Self-Assessment Questions
12.9 References / Reference Reading

120
12.1 Introduction to Queuing Theory
Queuing theory is defined as the study of waiting lines or queues using mathematical
principles as well. This theory is important in areas as diverse as operation research,
telecommunication, traffic, engineering and services. The major goal of queuing
theory is the performance evaluation of queues with the aim of estimating the queue
size and time taken therein in designing better systems of waiting.
Definition and Importance
Queuing theory is defined as systems in which customers or clients arrive, join
queues, wait for services, and finally leave after receiving the services. It offers means
and ways to design and control the processes and flows, as well as the usage of the
resources to guarantee such situations that allow avoiding the waiting time as much as
possible. One cannot underestimate the importance of queuing theory, especially
when it comes to satisfying the customer’s demands and optimizing company
operations.
Historical Background
Although the queuing theory has its roots in the twentieth century, researchers have
linked its beginning with the works of Danish mathematician Agner Krarup Erlang,
who focused on telephone networks. Erlang’s body of work can be believed to have
initiated the modern queuing theory, which at present comprises models that are
versatile and used in different organizations.

12.2 Characteristics of Queuing Systems

Systems of queuing have a number of defining attributes which are used to describe
the nature of the queue and may affect its behaviour and functioning.
Elements of a Queuing System
1. Arrival Process: This explains how customers get to the line or queue.
Depending on the exact definition of probabilistic reward, it can be
deterministic or stochastic, where the later is more realistic in real-world
situations.
2. Service Mechanism: This relates to how customers are served. It involves the
number of servers and the service rate.
3. Queue Discipline: This explains the order in which customers are served,
such as First-Come-First-Served (FCFS), Last-Come-First-Served (LCFS), or
priority-based.

121
 4. System Capacity: This explain the maximum number of customers that can
be in the system at any time.
5. Number of Servers: This shows how many service channels are available to
serve customers.
Arrival Process
The arrive process is typically shown using a probability distribution. Poisson
distribution is used most often when it is necessary to model random arrival time.
This distribution is defined by a rate parameter λ (lambda), which may metaphorically
be described as the number of arrivals per period of time.
Service Mechanism
The service mechanism is often modelled using the exponential distribution,
particularly when service times are memoryless. The rate parameter, µ (mu),
represents the average number of customers that can be served per time unit.
Queue Discipline
Queue discipline determines the order of service. The most common disciplines
include:
 First-Come-First-Served (FCFS): Customers are served in the order they
arrive.
 Last-Come-First-Served (LCFS): The most recent arrivals are served first.
 Priority Queues: Customers with higher priority are served before those with
lower priority.

 Knowledge Check 1
Fill in the Blanks.
1. Queuing theory involves the study of systems in which customers arrive, wait
for service, and ______ after receiving the service. (arrive)
2. The ______ model is the simplest queuing model with a single server and
exponential service times. (M/M/1)
3. In queuing systems, the arrival process is often modelled using the ______
distribution. (Poisson)
4. Queue discipline determines the ______ of service. ( order)

 Outcome-Based Activity 1

122
 Draw a simple diagram of a basic queuing system, labelling the arrival process,
queue, service mechanism, and departure process.

12.3 Models and Applications of Queuing Theory

Queuing theory encompasses various models, each suitable for different types of
queuing situations. These models help analyse and design efficient queuing systems.
Common Queuing Models
1. M/M/1 Model: This is the simplest model, with a single server, Poisson
arrival process, and exponential service times. It is widely used due to its
simplicity and applicability in many real-world scenarios.
2. M/M/c Model: This model extends the M/M/1 model to multiple servers (c
servers). It is useful for systems like call centres or hospital emergency rooms
where multiple service channels are available.
3. M/G/1 Model: This model allows for general service time distributions,
making it more flexible than the M/M/1 model.
4. G/G/1 Model: This is the most general model, allowing for any arrival and
service time distributions. It is used in complex and highly variable systems.
Applications in Various Industries
Queuing theory finds applications in numerous industries, including:
1. Telecommunications: Managing call traffic, network congestion, and data
packets.
2. Healthcare: Scheduling surgeries, managing emergency room queues,
optimising patient flow in hospitals,
3. Transportation: Traffic light timings, toll booths, and airport security checks.
4. Retail: Managing checkout lines, customer service desks, and inventory
replenishment.

12.4 Performance Measures in Queuing Systems

Performance measures are crucial for evaluating and improving queuing systems.
These measures help to understand the efficiency and effectiveness of the system.
Key Performance Metrics
1. Average Queue Length (Lq): The average number of customers in the queue.
2. Average Time in Queue (Wq): The average time a customer spends waiting
in the queue.

123
 3. Average Number of Customers in the System (L): The average number of
customers in the entire system (queue + service).
4. Average Time in the System (W): The average time a customer spends in the
system from arrival to departure.
5. Server Utilisation (ρ): The proportion of time the server is busy. It is
calculated as λ/µ for a single server system.
Calculating Performance Metrics
For the M/M/1 model, the performance metrics can be calculated using the following
formulas:

Example Calculation
Consider a bank with a single teller (M/M/1 model), where customers arrive at an
average rate of 10 per hour (λ = 10) and the teller can serve 12 customers per hour (µ
= 12).

12.5 Applications of Queuing Theory in Service Industries

One of the issues typical to service industries concerns the customer waiting time and
the use of the implementation resources. The queuing theory, therefore, offers great
contributions and possible remedies to these issues.
Queuing Theory in Retail
In retail, queuing theory is useful in managing the checkout procedures and
configuring the organization of the checkouts. Through studying the rates of customer
arrivals and the time it takes to serve each customer at the store, its managers can find

124
the right staffing density of cashiers that will help to reduce such a wait time of
customers as well as improve the quality of the stores.
Queuing Theory in Telecommunications
Telecommunications operators employ queuing theory in the management of traffic
flow and receipt of data within the telecommunication network. Solving the problem
of call arrivals and the service execution process enables the organization to build a
network that controls the flow of traffic during the highest usage of its services.
Queuing Theory in Transportation
Several essays describe how queuing theory applies to passages that airports have to
handle, namely the security lines and boarding. When information regarding the
passenger arrival pattern and the time taken to serve the passengers is obtained, it
becomes easy for the airport to plan its resources properly and effectively.

 Knowledge Check 2
State True or False.
1. The M/G/1 model allows for general service time distributions. (True)
2. In the M/M/c model, 'c' stands for the number of customers. (False)
3. Average time in the system (W) includes both waiting time in the queue and
service time. (True)
4. In telecommunications, queuing theory is rarely used to manage network
traffic. (False)

 Outcome-Based Activity 2
Identify a real-world example of a queuing system (e.g., a bank, supermarket, or
hospital) and describe its main elements (arrival process, service mechanism,
queue discipline, etc.).

12.6 Summary
 Queuing theory studies waiting lines, focusing on predicting queue lengths and
waiting times to design efficient systems. Originating from Agner Krarup Erlang's
work on telephone networks, it has broad applications in various fields like
operations research and telecommunications.

125
 The theory is vital for improving operational efficiency and customer satisfaction
by managing waiting times and resource utilisation in different sectors, including
healthcare, transportation, and retail.
 Key elements of a queuing system include the arrival process, service mechanism,
queue discipline, system capacity, and number of servers. These elements
collectively influence the system's performance and efficiency.
 These two basic forms of distribution, particularly the arrival and service formulas
based on the Poisson and exponential distribution, respectively, are central to the
study of queuing systems and their behaviour.
 The simple queuing models are M/M/1, M/M/c, M/G/1, and G/G/1 for accuracy
and detailed analysis of the queue. These models enable one to plan for and assess
proper queuing systems to be incorporated into different uses.
 In various fields such as telecommunications, health, transport, and stores,
queuing theory is used to better plan techniques, decrease the time consumers
must wait, and enhance the general performance of services.
 An indication of how the system is performing can be availed by some measures,
which include the average number of customers in queue (Lq), average waiting
time per customer (Wq), average number of customers in the system (L), average
time a customer spends in the system (W) and server utilization ratio (ρ).
 These are. Utilisation metrics give details of system efficiency and the use of the
available resources so that the business can come up with ways how to make its
clients happy and carry out its operations efficiently.
 In retail applications, queuing theory is useful in determining the proficiency and
staffing of check out points in order to reduce customer waiting time. The
achievement of the appropriate amount of resources to deliver the service is made
possible by comparing the rate of arrival of customers and service time.
 In communication technology, the queuing theory is used by telecommunications
industries to manage traffic; in transportation technology, it is applied in
manoeuvres such as airports and toll booths with the intention of enhancing the
quality of services as well as productivity.

12.7 Keywords

126
 Arrival Process: The pattern or distribution of times between consecutive arrivals
of customers to the queue.
 Service Mechanism: The method or process through which customers are served
in the system, including the number of servers and service rate.
 Queue Discipline: The rule or order by which customers are selected for service
from the queue (e.g., FCFS, LCFS).
 Poisson Distribution: A statistical distribution often used to model the random
arrival of customers in queuing theory.
 Utilisation (ρ): The proportion of time the server is busy, indicating how
effectively the service resources are being used.

12.8 Self-Assessment Questions

1. What are the key elements of a queuing system, and how do they impact its
performance?
2. Describe the M/M/1 queuing model and provide an example of its application in a
real-world scenario.
3. Explain the importance of queue discipline and list different types of queue
disciplines used in queuing systems.
4. How can performance measures such as average queue length and server
utilisation be calculated in an M/M/1 system?
5. Discuss the applications of queuing theory in the telecommunications industry.

12.9 References / Reference Reading

 Gross, Donald, John F. Shortle, James M. Thompson, and Carl M. Harris.
Fundamentals of Queueing Theory. 5th ed., Wiley, 2018.
 Bhat, U. Narayan. An Introduction to Queueing Theory: Modeling and Analysis in
Applications. 2nd ed., Birkhäuser, 2015.
 Medhi, J. Stochastic Models in Queueing Theory. Academic Press, 2002.
 Gokhale, Shriram S. Queueing Theory and Telecommunications: Networks and
Applications. Wiley India Pvt. Ltd, 2011.
 Sharma, J. K. Operations Research: Theory and Applications. 5th ed., Macmillan
India, 2013.

127
 Unit 13: Simulation Modeling

Learning Outcomes:
 Students will be able to understand the basic principles of simulation.
 Students will be able to apply Monte Carlo simulation to various problems.
 Students will be able to analyse the applications of simulation in operations
research.
 Students will be able to evaluate the advantages and limitations of simulation.
 Students will be able to understand the concepts and application of discrete event
simulation.

Structure:
13.1 Introduction to Simulation
13.2 Monte Carlo Simulation
 Knowledge Check 1
 Outcome-Based Activity 1
13.3 Applications of Simulation in Operations Research
13.4 Advantages and Limitations of Simulation
13.5 Discrete Event Simulation
 Knowledge Check 2
 Outcome-Based Activity 2
13.6 Summary
13.7 Keywords
13.8 Self-Assessment Questions
13.9 References / Reference Reading

128
13.1 Introduction to Simulation
Simulation is a necessary play area where functions of the actual system and its
behaviour are replicated for examination. It can be described as the reverting of a real
system into an artificially simplified one, which can be experimented on and be used
for analyzing results without practice contact. It is employed in operations research,
engineering, economics, and management as a problem-solving and decision-making
technique.
Definition of Simulation
Simulation is the modelling of a real process and its functioning throughout a
chronological period of time. In particular, it involves creating artificial data that
models the behaviour of an actual system in different conditions in order to make
further conclusions about how the system will perform.
Types of Simulation
1. Deterministic Simulation: This type of simulation has no stochastic factors
like in other forms of simulation studies. It is noteworthy that the final
outcomes are defined based on the initial conditions and parameters selected
by the user.
2. Stochastic Simulation: Includes such aspects that imply specific random
characteristics and probabilistic ones. These outcomes are stochastic, which
implies that the outcomes are not specifically certain and are instead
dependent on random inputs.
Steps in Simulation
1. Problem Definition: Identify the problem and the objectives of the
simulation.
2. Model Formulation: Develop a model that accurately represents the system.
3. Data Collection: Gather the data required for the model.
4. Model Translation: Convert the conceptual model into a computer-based
model.
5. Verification and Validation: Ensure the model is accurate and behaves as
expected.
6. Experimentation: Conduct experiments using the model to study different
scenarios.
7. Analysis: Analyse the results of the simulation to make informed decisions.

129
 8. Documentation and Reporting: Document the model and the findings for
future reference and decision-making.

13.2 Monte Carlo Simulation

Monte Carlo simulation is a quantitative approach that provides information on the
variability of the risk and the uncertainty in the models of prediction and forecasting.
It’s a process in which several simulations have been conducted to achieve a
statistical distribution of results.
Definition of Monte Carlo Simulation
Montre Carlo simulation is an approach that involves random sampling and statistical
simulations to approximate mathematical functions and the functioning of real
systems. It assists in explaining variation, as well as the randomness and risk of
occurrences.
Steps in Monte Carlo Simulation
1. Define the Problem: Clearly state the problem and determine the objectives
of the simulation.
2. Build the Model: Develop a mathematical model that represents the system.
3. Specify Probabilities: Identify the probability distributions of the input
variables.
4. Generate Random Inputs: Use random number generators to produce
random values for the input variables.
5. Run Simulations: Perform multiple iterations of the model using the random
inputs.
6. Analyse Results: Collect and analyse the results to estimate the probabilities
of different outcomes.
Applications of Monte Carlo Simulation
1. Financial Forecasting: Estimating future financial performance under
different market conditions.
2. Risk Analysis: Assessing the risk associated with investment portfolios.
3. Project Management: Evaluating the impact of uncertainty on project
schedules and costs.
4. Manufacturing: Optimising production processes and managing inventory
levels.

130
 Knowledge Check 1
Fill in the Blanks.
1. Simulation is the imitation of the operation of a real-world process or system
over _____. (time)
2. Deterministic simulation does not involve any _____ variables. (random) -
3. In Monte Carlo simulation, random sampling is used to estimate _____
functions. (mathematical)
4. One of the steps in Monte Carlo simulation is to specify the _____ of the input
variables. (probabilities)

 Outcome-Based Activity 1
Identify a real-world problem that could be solved using Monte Carlo simulation
and outline the steps you would take to build and run the simulation.

13.3 Applications of Simulation in Operations Research

There is no doubt that simulation is an essential element of operative research since it
brings light to specific systems and assists in enhancing decisions. It is applied
everywhere, including the supply chain, production, transportation, and medical
fields.
Supply Chain Management
The use of simulation helps in maximizing the supply chain because it is used in
modelling parts such as procurement, production, inventory, and distribution. It
enables an organisation to try out a number of solutions and then find the one that best
suits the organisation and is within the organisation’s capability as far as cost is
concerned.
Manufacturing
In manufacturing, simulation is used to create, model, enhance, and optimize
manufacturing processes. It becomes easier to measure cycle times, detect issues with
bottlenecks, and enhance resource utilization. Using certain systematic methods, the
manufacturers can improve the performance and efficiency of their operations.
Transportation
Transportation simulation is employed to analyze the efficiency of transport systems
or modes of transport individually and in combination. It plays a role in organizing

131
and enhancing traffic flow and in avoiding or minimizing traffic jams and transport
work.
Healthcare
In health care, simulation is done with the goal of coming up with the best approach
to the flow of patients, resources available, and overall services to be provided. It
assists in the general layout of the healthcare unit, especially the waiting time of
patients and the quality of the services to be provided.

Example: Simulation in Inventory Management

A retail company requires the right inventory management solution that can help
reduce costs while incorporating necessary products. Using the concept of inventory
policies, systems, and demand variations, the company is able to find meaningful
reorder points and order quantities, which in return minimize holding and stockout
costs.

13.4 Advantages and Limitations of Simulation

Simulation offers several advantages but also has certain limitations that need to be
considered.
Advantages of Simulation
1. Flexibility: Simulation can model complex systems with multiple interacting
components.
2. Experimentation: Allows for conducting experiments without disrupting the
real system.
3. Risk-Free: Enables the study of dangerous or expensive scenarios in a safe
environment.
4. Insightful: Provides insights into system behaviour and performance under
different conditions.
5. Cost-Effective: Reduces the need for costly physical prototypes and testing.
Limitations of Simulation
1. Model Accuracy: The accuracy of the simulation depends on the quality of
the model and data used.
2. Complexity: Building and validating complex models can be time-consuming
and resource-intensive.

132
 3. Interpretation: Analysing simulation results requires expertise and can be
subjective.
4. Computational Resources: High-fidelity simulations may require significant
computational power.
5. Assumptions: Simulations are based on assumptions that may not always hold
true in real-world scenarios.

13.5 Discrete Event Simulation

The discrete event simulation, or DES, is a broad kind of simulation that models a
system as a continuous sequence of operations or events. Each event happens at a
given time and is an instantiation of the transition, causing a state change in the
system.
Definition of Discrete Event Simulation
Discrete event simulation relates to the simulation of a system or equipment by
mapping behaviour as a series of particular events at distinct points in time. Turning
activates a state change in the system, and the simulation of the events records these
states in real time.
Components of Discrete Event Simulation
1. Entities: Objects or items that move through the system and are affected by
events (e.g., customers, products).
2. Events: Actions or occurrences that change the state of the system (e.g.,
arrival of a customer, completion of a service).
3. State Variables: Variables that describe the current state of the system (e.g.,
number of customers in a queue).
4. Resources: Elements that provide services to entities (e.g., servers, machines).
5. Queues: Holding areas where entities wait for resources or events (e.g.,
waiting lines).
Steps in Discrete Event Simulation
1. Model Development: Define the entities, events, and state variables of the
system.
2. Event Scheduling: Schedule events based on their occurrence times.
3. Event Execution: Execute events and update the state of the system.
4. Data Collection: Collect data on system performance during the simulation.
5. Analysis: Analyse the results to draw conclusions and make decisions.

133
Applications of Discrete Event Simulation
1. Manufacturing: Modelling production lines and assembly processes to
improve efficiency and reduce downtime.
2. Healthcare: Simulating patient flow in hospitals to optimise resource
allocation and reduce waiting times.
3. Transportation: Evaluating traffic management strategies and public
transportation systems.
4. Logistics: Optimising warehouse operations and distribution networks.

 Knowledge Check 2
State True or False.
1. Simulation can model complex systems with multiple interacting components.
(True)
2. One limitation of simulation is that it cannot handle probabilistic elements.
(False)
3. Discrete event simulation tracks changes in the system over time based on
specific events. (True)
4. The accuracy of simulation models does not depend on the quality of data
used. (False)

 Outcome-Based Activity 2
Create a simple discrete event simulation model for a queue at a bank and describe
the events and state variables involved.

13.6 Summary
 Simulation is the imitation of the operation of a real-world process or system over
time, allowing the study of complex systems without direct interaction.
 It involves steps such as problem definition, model formulation, data collection,
model translation, verification, experimentation, and analysis.
 Monte Carlo simulation uses random sampling and statistical modelling to
estimate mathematical functions and mimic complex systems under uncertainty.
 It is widely applied in financial forecasting, risk analysis, project management,
and manufacturing to understand variability and make informed decisions.

134
 Through simulation, it is possible to assess the effectiveness and efficiency of the
supply chain, manufacturing strategies, roles of transport, and future healthcare
operations and strategies.
 With the help of the simulations and the development of different situations,
individuals and companies can make superior and faster decisions, not only
increasing productivity but also decreasing organizational expenses.
 There are some benefits of simulation; they include flexibility since the simulation
model does not have to be altered each time there is a slight change in a particular
system; risk-free experiments since the actual system does not get affected; cost-
effective since it is cheaper in the long run; and it can also help provide overviews
of system behaviour under various circumstances.
 Challenges include the requirement of appropriate and sound models and data, as
well as the model construction process, which can be very costly and time-
consuming. Lastly, the use of models may be quite subjective.
 Based on this notion of event, discrete event simulation describes the actions of a
system in terms of events at certain points in time, which result in changes in
states.

13.7 Keywords
 Simulation: A technique used to imitate the operation of a real-world process or
system over time.
 Monte Carlo Simulation: A statistical method that uses random sampling to
estimate mathematical functions and mimic complex systems.
 Discrete Event Simulation: A simulation where the operation of a system is
represented as a chronological sequence of events.
 Probabilities: The likelihood of different outcomes in a simulation, essential for
Monte Carlo simulations.
 Entities: Objects or items that move through a simulation system and are affected
by events.

13.8 Self-Assessment Questions

1. What are the key steps involved in building a simulation model?
2. How does Monte Carlo simulation help in risk analysis and decision-making?

135
3. Discuss the applications of simulation in supply chain management.
4. What are the advantages and limitations of using simulation for complex system
analysis?
5. Explain the components and steps involved in discrete event simulation.

13.9 References / Reference Reading

 Narsimhan, S. L., and M. G. Subramanian. Simulation Modelling and Analysis.
Tata McGraw-Hill Education, 2021.
 Gupta, Prem Kumar, and D. S. Hira. Operations Research. S. Chand Publishing,
2020.
 Banks, Jerry, et al. Discrete-Event System Simulation. Pearson, 2019.
 Law, Averill M., and W. David Kelton. Simulation Modeling and Analysis.
McGraw-Hill, 2014.
 Sharma, J. K. Operations Research: Theory and Applications. Macmillan
Publishers India, 2017.

136
 Unit 14: Integer Programming

Learning Outcomes:
 Students will be able to define key concepts in integer programming.
 Students will be able to Fformulate integer programming problems from real-
world scenarios.
 Students will be able to solve integer programming problems using various
methods.
 Students will be able to apply integer programming to diverse practical
applications.
 Students will be able to implement the branch and bound method to solve integer
programming problems.

Structure:
14.1 Introduction to Integer Programming
14.2 Formulation of Integer Programming Problems
 Knowledge Check 1
 Outcome-Based Activity 1
14.3 Solution Methods for Integer Programming
14.4 Applications of Integer Programming
14.5 Branch and Bound Method
 Knowledge Check 2
 Outcome-Based Activity 2
14.6 Summary
14.7 Keywords
14.8 Self-Assessment Questions
14.9 References / Reference Reading

137
14.1 Introduction to Integer Programming
On the other hand, Integer programming (IP) is a branch of mathematical
programming that involves constraining some or all decision variables to assume
integral values. It is generally employed as an effective tool in many disciplines,
including Operations Research, Management Science, and Industrial Engineering. As
opposed to LP model, which allows decision variables to have continuous values,
integer programming offers more realistic solutions in scenarios where particular
variables are numeric and singly-valued, indicating the quantities of certain products,
machines, or people.
Definition: Integer programming as the name suggests, is a type of linear
optimization problem where some or all of the variables are restricted to integration
values.. It can be classified into three main types:
1. Pure Integer Programming: All decision variables are integers.
2. Mixed-Integer Programming (MIP): Some decision variables are integers, and
others are continuous.
3. Binary Integer Programming: Decision variables are binary (0 or 1).
Key Features:
 Decision Variables: Variables that decision-makers will use to decide the
values of.
 Objective Function: The function to be maximized or minimized.
 Constraints: Equations or inequalities that restrict the values of the decision
variables.

14.2 Formulation of Integer Programming Problems

Formulating an integer programming problem involves defining the decision
variables, objective function, and constraints. The process typically follows these
steps:
1. Identify the Decision Variables: Determine what variables will be controlled.
2. Define the Objective Function: Establish a clear objective, such as
minimizing costs or maximizing profits.
3. Set Up the Constraints: Identify the limitations or requirements that must be
satisfied.

138
Example:
Consider a manufacturing company that wants to decide how many units of two
products to produce. Let x1 be the number of units of product 1 and x2 be the number
of units of product 2.
Objective Function: Maximize profit: Z = 3x1 + 5x2
Constraints:
1. Production time constraint: 2x1 + 3x2 ≤ 50
2. Material availability constraint: x1 + x2 ≤20
3. Non-negativity constraint: x1, x2 ≥ 0
4. Integer constraint: x1, x2 are integers.

 Knowledge Check 1
Fill in the Blanks.
1. Integer programming is a branch of mathematical programming where some
or all of the variables are restricted to take ________ values. ( integer)
2. In mixed-integer programming, some decision variables are ________ while
others are continuous. (integers)
3. The first step in formulating an integer programming problem is to identify the
________. (decision variables)
4. The objective function in an integer programming problem is the function to
be ________. (maximized)

 Outcome-Based Activity 1
Create a simple integer programming problem with two decision variables and
write down its objective function and constraints.

14.3 Solution Methods for Integer Programming

Solving integer programming problems can be more complex than linear
programming due to the integrality constraints. Several methods are used to find
solutions:

139
Exact Methods
1. Branch and Bound: A tree-based method where the problem is divided into
smaller subproblems. Each subproblem is solved, and bounds are used to eliminate
suboptimal solutions.
Steps:
 Branching: Divide the problem into subproblems.
 Bounding: Calculate upper and lower bounds for each subproblem.
 Pruning: Eliminate subproblems that do not lead to optimal solutions.
2. Cutting Planes: A method that iteratively adds constraints (cuts) to the problem to
remove non-integer solutions from the feasible region.
Steps:
 Solve the LP relaxation of the IP.
 Check for the integrality of the solution.
 If the solution is not integer, add a cut to the feasible region.
 Repeat until an integer solution is found.
Heuristic and Metaheuristic Methods
1. Greedy Algorithm: A simple, intuitive algorithm that builds a solution step-by-
step, always choosing the next step that offers the most immediate benefit.
2. Genetic Algorithms: Inspired by natural selection, these algorithms use
crossover, mutation, and selection to evolve solutions over generations.
3. Simulated Annealing: A probabilistic technique that explores the solution space
and accepts worse solutions with a decreasing probability of escaping local
optima.

14.4 Applications of Integer Programming

Integer programming has a wide range of applications in various industries. Here are
some notable examples:
1. Supply Chain Management: Optimizing the production, transportation, and
distribution of goods. For example, determining the number of products to
manufacture and ship to different locations to minimize costs and meet demand.
2. Scheduling: Planning and scheduling in industries like manufacturing, airlines,
and project management. For example, assigning tasks to machines or employees
to maximize efficiency or minimize total completion time.

140
3. Capital Budgeting: Selecting investment projects under budget constraints.
Companies use IP to choose the most profitable projects while staying within
budget limits.
4. Facility Location: Deciding the optimal locations for warehouses, factories, or
service centres to minimize costs or maximize service coverage.

14.5 Branch and Bound Method

The branch and bound method is one of the most widely used techniques for solving
integer programming problems. It systematically explores the solution space by
dividing it into smaller subproblems (branching) and calculating bounds for each
subproblem to eliminate non-promising areas (bounding).
Steps in Branch and Bound:
1. Initial Problem: Start with the original problem and solve its LP relaxation
(i.e., ignore the integrality constraints).
2. Branching: Select a variable with a non-integer value in the LP solution and
create two subproblems by setting the variable to its lower and upper integer
bounds.
3. Bounding: Solve the LP relaxations of the subproblems to find bounds. If a
subproblem's bound is worse than the current best solution, discard that
subproblem.
4. Pruning: Eliminate subproblems that do not improve the current best solution
or that cannot produce feasible solutions.
5. Iteration: Repeat the process for the remaining subproblems until all have
been either solved or pruned.
Example:
Consider the earlier manufacturing example with the objective function and
constraints. The branch and bound method would involve:
 Solving the LP relaxation.
 Identifying non-integer variables in the solution.
 Branching on these variables to create subproblems.
 Bounding and pruning subproblems based on the bounds obtained.

141
Diagrams and Formulas
Linear Programming Relaxation (LPR):

Branch and Bound Tree Example:

Initial Problem (LPR)
/ \
x1=2 x1=3
/ \ / \
x2=5 x2=6 x2=4 x2=5
(Source: Created for illustration purposes)

 Knowledge Check 2
State True or False.
1. Branch and bound is a heuristic method for solving integer programming
problems. (False)
2. The cutting planes method adds constraints to the problem to remove non-
integer solutions from the feasible region. (True)
3. The greedy algorithm is an exact method used in integer programming. (False)
4. Integer programming can be used to optimize supply chain management
problems. (True)

 Outcome-Based Activity 2
Explain how the branch and bound method can be used to solve an integer
programming problem of your choice.

14.6 Summary
 Integer programming is a type of mathematical optimization where some or all
decision variables must be integers. It is essential for problems where variables
represent discrete items like products, machines, or people.

142
 The main types of integer programming are pure integer programming, mixed-
integer programming (MIP), and binary integer programming, each with specific
applications based on the nature of the decision variables.
 Formulating an integer programming problem involves identifying decision
variables, defining the objective function, and setting up constraints. This process
ensures the problem accurately represents the real-world scenario.
 Concrete techniques for integer programming problems include the branch and
bound and the cutting planes techniques that involve exploration of the entire
feasible solution region.
 Greedy and meta-greedy methods, including greedy algorithms, genetic
algorithms, and Simulated annealing, offer sub-optimal solutions and are useful
when exact models cannot be solved due to the exponential time.
 Integer programming is widely applicable in the supply chain to address
production, transportation, and distribution issues and meet the necessary
demands at the lowest cost possible.
 Other uses are in the management of tasks in the manufacturing industry, the
choice of investment projects in capital asset pricing, and the decision on location
for the establishment of warehouses or service centres.
 This method involves solving the LP relaxation of the IP and then branching on
the discrete variables with non-integers in their changing domain, thus subsequent
refining of the subproblem until an integer solution is successfully obtained.

14.7 Keywords
 Integer Programming (IP): A type of optimization problem where some or all
variables are constrained to be integers.
 Branch and Bound: A method used to solve integer programming problems by
dividing them into smaller subproblems.
 Cutting Planes: A technique that adds linear constraints to remove non-integer
solutions from the feasible region of an IP problem.
 Mixed-Integer Programming (MIP): An optimization problem where some
variables are integers and others are continuous.
 Objective Function: The function in an optimization problem that is to be
maximized or minimized.

143
14.8 Self-Assessment Questions
1. Define integer programming and explain its key differences from linear
programming.
2. Describe the process of formulating an integer programming problem.
3. Explain the branch and bound method and outline its key steps.
4. Discuss the applications of integer programming in real-world scenarios.
5. How does the cutting planes method work in solving integer programming
problems?

14.9 References / Reference Reading

 Kapoor, V.K. Operations Research: Quantitative Techniques for Management.
Sultan Chand & Sons, 2021.
 Sharma, J.K. Operations Research: Theory and Applications. Macmillan
Publishers India, 2020.
 Taha, Hamdy A. Operations Research: An Introduction. 10th ed., Pearson
Education, 2017.
 Rardin, Ronald L. Optimization in Operations Research. Pearson Education India,
2017.
 Ravindran, A., Don T. Phillips, and James J. Solberg. Operations Research:
Principles and Practice. 2nd ed., Wiley India, 2020.

144
 Unit 15: Non-Linear Programming

Learning Outcomes:
 Students will be able to define the fundamental concepts of non-linear
programming.
 Students will be able to formulate non-linear programming problems.
 Students will be able to apply various solution techniques for non-linear
programming problems.
 Students will be able to analyse real-life case studies involving non-linear
programming.
 Students will be able to distinguish between convex and non-convex optimization.

Structure:
15.1 Introduction to Non-Linear Programming
15.2 Formulation and Applications of Non-Linear Programming
 Knowledge Check 1
 Outcome-Based Activity 1
15.3 Solution Techniques for Non-Linear Programming
15.4 Case Studies in Non-Linear Programming
15.5 Convex and Non-Convex Optimization
 Knowledge Check 2
 Outcome-Based Activity 2
15.6 Summary
15.7 Keywords
15.8 Self-Assessment Questions
15.9 References / Reference Reading

145
15.1 Introduction to Non-Linear Programming
Non-linear programming (NLP) is an applicable approach to decision-making where
the constraints are expressed with non-linear functions. Recall that while linear
programming involves the objective function and the constraints are represented as
linear equations/inequalities, non-linear programming contains non-linear elements.
Definition and Basic Concepts
Nonlinear programming involves finding the optimal value of an objective function –
this can be to maximise or minimise with regard to constraints. Multiple, non-linear,
objective functions or constraints can make a problem complicated because these
expressions can model various complex situations that can occur in numerous
disciplines.
Importance of Non-Linear Programming
Nonlinear programming is important because of the many problems in the fields of
engineering, economics, finance, management, and many others. They enable the
development of realistic models, facilitating an optimal representation of improved
systems and processes for proper decision-making.
Differences Between Linear and Non-Linear Programming
Linear programming focuses on issues in which objective functions and constraints
are linear, while non-linear programming addresses issues of non-linearity. These
differences have a colossal bearing on the level of difficulty and the approaches
needed to address them.

15.2 Formulation and Applications of Non-Linear Programming

Formulating Non-Linear Programming Problems
A problem formulation means coming up with an objective function and constraints to
the problem. The objective can be profit, cost, or any other measure of performance
that has to be optimized in the process of making decisions. The constraints define the
problem, the boundaries and parameters within which the recurrence must be solved,
e. g. time constraints, lack of resources.
Example
Consider a manufacturing company that needs to maximize its profit (P) based on the
production levels of two products, X and Y. The profit function might be non-linear
due to economies of scale or other factors:

146
 P = 50X + 40Y − 0.5X2 − 0.3Y2
Subject to constraints:
X + Y ≤ 100
2X + Y ≥ 50
X, Y ≥ 0
Applications of Non-Linear Programming
Non-linear programming is used in various fields to solve complex optimization
problems:
1. Engineering: For designing optimal structures, machinery, and systems.
2. Economics: To determine the most efficient allocation of resources.
3. Finance: For portfolio optimization and risk management.
4. Management: To optimize production schedules, supply chains, and resource
allocation.

Real-World Example
In the finance sector, non-linear programming is used to optimize investment
portfolios. Investors aim to maximize returns while minimizing risk, which involves a
non-linear relationship between the returns and the risk associated with different
assets.

 Knowledge Check 1
Fill in the Blanks.
1. Non-linear programming involves optimizing an objective function subject to
__________. (Variables)
2. One of the key differences between linear and non-linear programming is that
non-linear programming involves at least one __________ component.
(Non-linear)
3. Non-linear programming is crucial for solving real-world problems in various
fields such as __________ and finance. (Engineering)
4. The formulation of a non-linear programming problem includes defining an
objective function and __________. (Constraints)

147
 Outcome-Based Activity 1
Write down a real-life example of a non-linear programming problem you might
encounter in your field of study or future career.

15.3 Solution Techniques for Non-Linear Programming

Graphical Method
The graphical method can be used for simple NLP problems with two variables. This
involves plotting the objective function and constraints on a graph to find the feasible
region and the optimal solution.
Analytical Methods
1. Lagrange Multipliers: Used to find the local maxima and minima of a
function subject to equality constraints.
2. Kuhn-Tucker Conditions: Generalizes the method of Lagrange multipliers
for inequality constraints.
Example
To maximize f (x,y) = xy subject to x2 + y2 ≤ 1, we use the Lagrange multipliers: L(x,
y, λ) = xy + λ (1 − x2 − y2)
Solving the partial derivatives and setting them to zero gives the optimal values of
xxx and yyy.
Numerical Methods
1. Gradient Descent: Iteratively moves towards the minimum of a function
using its gradient.
2. Newton's Method: Uses the second-order Taylor series expansion to find the
minimum or maximum of a function.
3. Interior-Point Methods: Suitable for large-scale non-linear programming
problems, finding solutions by traversing the interior of the feasible region.
Software for Non-Linear Programming
Various software tools are available for solving NLP problems, such as MATLAB, R,
Python (SciPy library), and specialized optimization software like GAMS and AMPL.

148
15.4 Case Studies in Non-Linear Programming
Case Study 1: Optimizing Production in a Manufacturing Plant
A manufacturing plant aims to maximize its output while minimizing costs. The
objective function is non-linear due to varying production efficiencies and costs at
different production levels.
Problem Formulation
Objective Function:
Z = 100X + 150Y −2X2 − 3Y2
Constraints:
4X + 6Y ≤ 240
X,Y ≥ 0
Solution
Using numerical methods like gradient descent or software tools, the optimal
production levels of products X and Y are determined to maximize profit.
Case Study 2: Portfolio Optimization
An investment firm wants to maximize returns while managing risk. The relationship
between different assets and the portfolio return is non-linear due to market dynamics.
Problem Formulation
Objective Function:

Constraints:

Solution
By applying the Kuhn-Tucker conditions and using software tools, the optimal
weights w1 and w2 for the assets are found to achieve the desired balance of return and
risk.
Case Study 3: Optimal Pricing Strategy
A company wants to determine the optimal pricing strategy for its products to
maximize revenue, considering non-linear demand functions.
Problem Formulation
Objective Function:

149
Where D1 and D2 are non-linear demand functions for products 1 and 2.
Solution
Using numerical methods and market data, the optimal prices p1p_1p1 and p2p_2p2
are calculated to maximize revenue.

15.5 Convex and Non-Convex Optimization

Definition of Convex Optimization
Convex optimization deals with problems where the objective function is convex, and
the feasible region formed by the constraints is also convex. In convex optimization,
any local minimum is a global minimum, making these problems easier to solve.
Characteristics of Convex Functions
1. First Derivative Test: A function f(x) is convex if f′′(x) ≥ 0 for all xxx.
2. Second Derivative Test: For a multivariable function f(x), it is convex if its
Hessian matrix is positive semi-definite.
Convex Optimization Techniques
1. Linear Programming: A special case of convex optimization.
2. Quadratic Programming: Involves a quadratic objective function and linear
constraints.
3. Semidefinite Programming: Optimizes a linear objective function subject to
the constraint that a symmetric matrix is semidefinite.
Example
Minimize f(x) = x2 + y2 subject to x + y ≥ 1. This is a convex optimization problem
since the objective function and feasible region are convex.
Non-Convex Optimization
Non-convex optimization involves objective functions or constraints that are not
convex, leading to multiple local minima and maxima. These problems are more
complex and require advanced techniques for finding global optima.
Techniques for Non-Convex Optimization
1. Simulated Annealing: A probabilistic technique that searches for the global
optimum by exploring the solution space.
2. Genetic Algorithms: Mimics natural selection to find optimal solutions.
3. Branch and Bound: Systematically divides the problem into smaller
subproblems to find the global optimum.

150
Example
Maximize f(x) = sin(x) over [0,2π]. This is a non-convex problem with multiple local
maxima.

 Knowledge Check 2
State True or False.
1. Gradient descent is a numerical method used for solving non-linear
programming problems. (True)
2. Lagrange multipliers are used to find the global maximum of a function
subject to equality constraints. (False)
3. Convex optimization problems are easier to solve because any local minimum
is also a global minimum. (True)
4. Non-convex optimization problems always have a unique solution.
(False)

 Outcome-Based Activity 2
Identify a case study or real-world application of non-linear programming in your
local industry or community.

15.6 Summary
 Non-linear programming (NLP) involves optimizing an objective function that
includes non-linear components, which adds complexity compared to linear
programming. It is crucial in real-world applications for more accurate modelling.
 NLP is widely embedded in disciplines where the dependency between variables
is nonlinear, such as engineering sciences, economics, finance, and the
management field.
 NLP problems are to be defined with a non-linear term/objective function and
constraints that describe/part of the problem. An example is a manufacturing
company making decisions with regard to profits influenced by rates of
production.
 Applications of NLP are diverse and evident in engineering disciplines for design
purposes, in finance for asset selection, and in management for distribution of
resources; thus, the significance of these tools is great.

151
 Most researchers prefer to use specific software tools to solve NLP problems
because of the complexity and massive datasets involved; some of these tools
include MATLAB, R, Python, GAMS, and AMPL.
 The study also presents a case of using NLP in the context of a manufacturing
plant as it demonstrates how it is possible to achieve maximum production levels
to maximum profits through consideration of non-linear cost functions and
constraints on resources.
 Convex optimization is committed in-range problems in which the target function
and the feasible set are convex so that any local optimum is a global optimum,
making these problems easier to solve.
 Non-convex optimization has the condition that functions do not necessarily have
linear relationships and therefore the objective function can have a number of
local minima and local maxima which means that optimization algorithms like
simulated annealing and genetic algorithms are useful in identifying the global
optima..

15.7 Keywords
 Non-Linear Programming (NLP): A type of optimization where the objective
function or constraints are non-linear.
 Objective Function: A mathematical expression that defines the goal of an
optimization problem, such as maximizing profit or minimizing cost.
 Constraints: Limitations or requirements in an optimization problem that the
solution must satisfy.
 Convex Optimization: A subclass of optimization problems where the objective
function is convex, making the problem easier to solve.
 Gradient Descent: A numerical method used to find the minimum of a function
by iteratively moving in the direction of the steepest descent.

15.8 Self-Assessment Questions

1. What is non-linear programming, and how does it differ from linear
programming?
2. Describe the process of formulating a non-linear programming problem.

152
3. Explain the graphical method for solving simple non-linear programming
problems.
4. What are Lagrange multipliers, and how are they used in non-linear
programming?
5. Discuss a case study where non-linear programming was used to solve a complex
problem.

15.9 References / Reference Reading

 Kambo, Narayan C. "Mathematical Programming Techniques." Tata McGraw-
Hill, 2021.
 Sharma, J. K. "Operations Research: Theory and Applications." Macmillan India,
2022.
 Pant, Madhu. "Introduction to Optimization Techniques." Laxmi Publications,
2023.
 Rao, Singiresu S. "Engineering Optimization: Theory and Practice." John Wiley &
Sons, 2020.
 Bazaraa, Mokhtar S., Hanif D. Sherali, and C. M. Shetty. "Nonlinear
Programming: Theory and Algorithms." Wiley-Interscience, 2021.

153
 Unit 16: Decision Analysis and Multi-Criteria Decision-making

Learning Outcomes:
 Students will be able to explain the fundamentals of decision analysis.
 Students will be able to construct decision trees and payoff tables for various
decision-making scenarios.
 Students will be able to apply multi-criteria decision-making (MCDM) techniques
to real-world problems.
 Students will be able to evaluate the applications of MCDM in operations
research.
 Students will be able to analyse the processes involved in the Analytical Hierarchy
Process (AHP) and Technique for order Preference by Similarity to Ideal Solution
(TOPSIS).

Structure:
16.1 Introduction to Decision Analysis
16.2 Decision Trees and Payoff Tables
 Knowledge Check 1
 Outcome-Based Activity 1
16.3 Multi-Criteria Decision-making (MCDM) Techniques
16.4 Applications of MCDM in Operations Research
16.5 Analytical Hierarchy Process (AHP) and Technique for Order Preference by
Similarity to Ideal Solution (TOPSIS)
 Knowledge Check 2
 Outcome-Based Activity 2
16.6 Summary
16.7 Keywords
16.8 Self-Assessment Questions
16.9 References / Reference Reading

154
16.1 Introduction to Decision Analysis
Decision analysis is a systematic, quantitative, and visual approach to addressing and
evaluating important choices confronted by businesses. It involves breaking down a
decision into its fundamental components, such as objectives, alternatives, and
consequences. The goal is to provide a clear framework to help decision-makers
evaluate their options and choose the best course of action based on their preferences
and the possible outcomes.
Definition and Importance
Decision analysis can be described as a field that contains the concepts, principles,
procedures, and knowledge that are pertinent and vital in the management of
decisions. For businesses, it is valuable because it has the potential to be instrumental
in enhancing the quality of business decisions while reducing the likelihood of
adverse outcomes and increasing awareness of the various impacts of different
options.
Key Components of Decision Analysis
 Objectives: What are we trying to achieve?
 Alternatives: What options do we have?
 Consequences: What are the potential outcomes of each alternative?
 Preferences: How do we value the different consequences?
Steps in Decision Analysis
1. Identify the Decision Problem: Define the decision to be made and the
context.
2. Determine Objectives: Establish what you want to achieve.
3. Generate Alternatives: List possible courses of action.
4. Evaluate Consequences: Assess the outcomes of each alternative.
5. Make the Decision: Choose the best alternative based on the evaluation.
Real-World Example
Consider a company deciding whether to launch a new product. The decision analysis
process would involve:
 Identifying the goal (e.g., increasing market share).
 Considering alternatives (e.g., launching the product, modifying it, or not
launching).
 Evaluating the potential market response and financial impact of each
alternative.

155
  Making a decision based on the analysis.

16.2 Decision Trees and Payoff Tables

Decision trees and payoff tables are methods employed in decision analysis and
applied to identify decisions and their potential repercussions. It also assists in the
identification of all pros and cons of the various options available.
 Decision Trees
A decision tree is a tree-like diagram in which decision nodes represent the decision
points and chance nodes represent the chance events where the probabilities of the
events can also be incorporated. It assists in building up formulation processes and
making the right decisions in this case.
Components of a Decision Tree
 Decision Nodes: Represent decisions to be made (usually depicted by
squares).
 Chance Nodes: Represent uncertain outcomes (usually depicted by circles).
 Branches: Show the options available at each node.
 Leaves: Represent the final outcomes of the decision paths.
Constructing a Decision Tree
1. Define the decision points.
2. List possible actions and outcomes for each decision point.
3. Calculate the probabilities and values of outcomes.
4. Evaluate the expected values to make the final decision.
Example
A company is considering whether to expand its operations to a new region. The
decision tree would include:
 The initial decision to expand or not.
 Potential market conditions (favourable or unfavourable).
 Financial outcomes based on market conditions.

 Payoff Tables
A payoff table lists the possible payoffs for different actions under various states of
nature. It is a tabular representation of the outcomes of different decisions and helps
in comparing them.

156
Components of a Payoff Table
 Actions: Different strategies or decisions.
 States of Nature: Different scenarios that might occur.
 Payoffs: The results of each action under each state of nature.
Constructing a Payoff Table
1. Identify the possible actions.
2. List the states of nature.
3. Determine the payoffs for each combination of action and state of nature.
Example
A retailer deciding on the quantity of stock to order might use a payoff table to
compare the profits under different demand scenarios (high, medium, low).

 Knowledge Check 1
Fill in the Blanks.
1. Decision analysis involves breaking down a decision into its fundamental
components, such as objectives, _________, and consequences. (alternatives)
2. In decision trees, ________ nodes represent decisions to be made and are
usually depicted by squares. (decision)
3. A payoff table lists the possible _________ for different actions under various
states of nature. (payoffs)
4. The goal of decision analysis is to provide a clear framework to help decision-
makers evaluate their options and choose the best _________. (alternative)

 Outcome-Based Activity 1
Create a simple decision tree for a personal decision, such as choosing a holiday
destination. Identify the decision points, possible actions, and outcomes.

16.3 Multi-Criteria Decision-making (MCDM) Techniques

Multi-Attribute Decision-Making (MADM) is a decision-making process in which
two or more attributes that may be in conflict are considered in the process of making
the decision. They are useful in assessing the relative merits or feasibility of different
courses of action considering several factors.
Introduction to MCDM

157
MCDM techniques are relevant in a situation where the decision-making process
involves several factors that need to be measured. These techniques help to organize a
decision-making process to make selections based on the values that define the goals
of the decision-maker.
Common MCDM Techniques
 Weighted Sum Model (WSM): Involves assigning weights to criteria and
summing the weighted scores of each alternative.
 Weighted Product Model (WPM): Similar to WSM but uses multiplication
instead of addition.
 Analytical Hierarchy Process (AHP): Breaks down a decision problem into
a hierarchy of more easily comprehended sub-problems.
 Technique for Order Preference by Similarity to Ideal Solution (TOPSIS):
Identifies solutions from a finite set of alternatives based on their distance to
an ideal solution.
Steps in MCDM
1. Define the problem and criteria.
2. Identify the alternatives.
3. Assign weights to criteria based on their importance.
4. Evaluate each alternative against the criteria.
5. Aggregate the evaluations to determine the best alternative.
Example
A company selecting a new supplier might use MCDM techniques to evaluate options
based on criteria such as cost, quality, and delivery time.

16.4 Applications of MCDM in Operations Research

Optimization and simulation techniques such as MCDM are common in operations
research to cater for diverse decision-making difficulties in different markets.

 Supply Chain Management

In supply chain management, MCDM can help identify the required suppliers, select
the available transportation modes for products, and manage inventory.
Supplier Selection
Companies can also use MCDM to make decisions on which supplier has a lower
cost, better quality products, and more reliable.
158
Example
An example of applying AHP in a manufacturing company is evaluating the supplier
portfolio using decision criteria such as price, quality, and delivery.
 Project Management
MCDM is used in project management to prioritise projects, allocate resources, and
manage risks.
Resource Allocation
MCDM techniques help managers identify where and how to allocate resources for
subprojects in different projects to support the goal of getting the highest favourable
outcome.
Example
A construction company might use TOPSIS to prioritise projects based on criteria
such as profitability, risk, and resource requirements.
 Environmental Management
MCDM is useful in decision-making as it pertains to specific environmental
management decisions, such as choosing between two or more pollution control
technologies or assessing sustainability projects.
Sustainability Initiatives
Businesses can apply MCDM to measure the sustainability outcomes of various
programs and operations in terms of environment, economy and society.
Example
A company might use the Weighted Sum Model to evaluate different waste
management practices based on cost, environmental impact, and regulatory
compliance.

16.5 Analytical Hierarchy Process (AHP) and Technique for Order Preference by
Similarity to Ideal Solution (TOPSIS)
AHP and TOPSIS are two popular MCDM techniques that help in making complex
decisions involving multiple criteria.
 Analytical Hierarchy Process (AHP)
AHP is a decision-making tool in which multiple factors that form a problem can be
systematically grouped and analysed. It involves the process of decomposing the

159
decision problem into a set of more manageable sub-problems, and one is capable of
addressing each of them.
Steps in AHP
1. Define the problem and the goal.
2. Structure the hierarchy from the top (goal) through intermediate levels
(criteria) to the bottom level (alternatives).
3. Construct pairwise comparison matrices for each level.
4. Calculate the weights of the criteria and the scores of alternatives.
5. Synthesize these results to determine the best alternative.
Example
A company choosing a location for a new plant might use AHP to compare sites based
on factors like cost, accessibility, and availability of skilled labour.
 Technique for Order Preference by Similarity to Ideal Solution (TOPSIS)
TOPSIS is a method for ranking alternatives based on their distance from an ideal
solution. The ideal solution is the one that has the best level for all criteria, whereas
the negative-ideal solution has the worst level for all criteria.
Steps in TOPSIS
1. Construct a decision matrix.
2. Normalize the decision matrix.
3. Construct the weighted normalized decision matrix.
4. Determine the ideal and negative-ideal solutions.
5. Calculate the separation measures for each alternative.
6. Calculate the relative closeness to the ideal solution.
7. Rank the alternatives based on their relative closeness.

Example
A company evaluating different software packages might use TOPSIS to rank the
options based on criteria such as cost, functionality, and user-friendliness.
 Comparison of AHP and TOPSIS
While both AHP and TOPSIS are used for multi-criteria decision-making, they have
different strengths:
 AHP: Suitable for decisions involving complex hierarchies and where
pairwise comparisons are meaningful.

160
  TOPSIS: Ideal for ranking alternatives based on their distance from the ideal
solution.
Real-World Applications
 AHP: Used in resource allocation, vendor selection, and strategic planning.
 TOPSIS: Applied in manufacturing, healthcare, and environmental
management.

 Knowledge Check 2
State True or False.
1. The Analytical Hierarchy Process (AHP) breaks down a decision problem into
a hierarchy of more easily comprehended sub-problems. (True)
2. The Weighted Sum Model (WSM) uses multiplication instead of addition for
evaluating alternatives. (False)
3. The ideal solution in TOPSIS has the worst level for all criteria. (False)
4. Multi-Criteria Decision-making (MCDM) techniques are crucial when
decisions involve balancing various factors. (True)

 Outcome-Based Activity 2
Identify and list three criteria that you would use to evaluate a new smartphone
purchase. Rank these criteria in order of importance.

16.6 Summary
 Decision analysis involves systematically evaluating and comparing different
choices to enhance decision quality. It is crucial for minimizing risks and
understanding the potential outcomes of various options.
 Key components include defining objectives, identifying alternatives, assessing
consequences, and understanding preferences. These elements form the basis of a
structured approach to decision-making.
 Decision trees are a logical flowchart like a graphical or model approach to show
the decisions and their likely outcomes, whether it be certain or speculative, and
the cost factors of resources that may be involved. They help in decision
architecture and the review of individual outcomes as well as architectures.

161
 MCDM techniques help in sorting out options, their preference, or ranking in
terms of various factors, which is crucial when much is at stake and the decision
involves compromising this or that factor. However, some of the commonly used
approaches include Weighted Sum Model, AHP, and TOPSIS.
 With respect to supply chain management, MCDM is very useful in determining
the best suppliers, identifying the appropriate means of transport within supply
chain and efficiently managing an inventory through consideration and ranking of
multiple attributes.
 In project management, MCDM serves the purpose of choosing priorities and
aggregators, managing risks, and distributing resources throughout projects in the
easiest way possible.
 AHP is a rational methodology that deconstructs complex decisions into
constituent components, enabling a finer grappling with the challenge.

16.7 Keywords
 Decision Analysis: A systematic approach to making decisions using quantitative
and visual methods to evaluate different options.
 Decision Tree: A graphical representation of decisions and their possible
outcomes, including chance events, resource costs, and utilities.
 Payoff Table: A tabular representation that lists the possible outcomes for
different actions under various states of nature.
 Multi-Criteria Decision-making (MCDM): Techniques used to evaluate and
prioritize different options based on multiple criteria.
 Analytical Hierarchy Process (AHP): A structured technique for organizing and
analysing complex decisions, breaking them down into a hierarchy of more
comprehensible sub-problems.

16.8 Self-Assessment Questions

1. What are the key components of decision analysis and why are they important in
business management?
2. Explain the process of constructing a decision tree and its significance in decision-
making.

162
3. How does a payoff table help in comparing different decision options? Provide an
example.
4. Discuss the steps involved in Multi-Criteria Decision-making (MCDM) and its
applications.
5. Compare and contrast the Analytical Hierarchy Process (AHP) and Technique for
Order Preference by Similarity to Ideal Solution (TOPSIS).

16.9 References / Reference Reading

 Saaty, Thomas L. Decision-making for Leaders: The Analytic Hierarchy Process
for Decisions in a Complex World. Pittsburgh: RWS Publications, 2001.
 Ishizaka, Alessio, and Philippe Nemery. Multi-Criteria Decision Analysis:
Methods and Software. Chichester: Wiley, 2013.
 Rao, R. Venkata. Decision-making in the Manufacturing Environment: Using
Graph Theory and Fuzzy Multiple Attribute Decision-making Methods. London:
Springer, 2007.
 Mishra, Rajesh. Operations Research and Decision-Making Techniques. Mumbai:
Himalaya Publishing House, 2019.
 Tiwari, R.N. Operations Research: Methods and Problems. New Delhi: CBS
Publishers & Distributors, 2020.

163