VIVEKANANDHA

VIVEKANANDHA
ARTS AND SCIENCE COLLEGE FOR
WOMEN
ARTS AND SCIENCE COLLEGE
[An ISO 9001:2015 FOR WOMEN
Certified Institution]
(Affiliated to Periyar University, Salem andCertified
[An ISO 9001:2015 Accredited with 2(f) and 12 (B) status)
Institution]
Veerachipalayam
(Affiliated – 637 303,Salem
to Periyar University, Sankagiri, Salem Dt., with
and Accredited Tamil Nadu.
2(f) and 12 (B) status)
www.viaasrtt.ac.in
Veerachipalayam – 637 303, Sankagiri, Salem Dt., Tamil Nadu.
www.viaasrtt.ac.in

Mr.B.M.TAMILZHARASAN,M.Sc.,M.Phil.,
Mr.B.M.TAMILZHARASAN,M.Sc.,M.Phil.,
(Ph.D).,
(Ph.D).,
ASSISTANT PROFESSOR
ASSISTANT PROFESSOR
PG DEPARTMENT OF MATHEMATICS
PG DEPARTMENT OF MATHEMATICS
Notes of Lesson
Notes of Lesson
Resource Management Techniques
ResourceII Management Techniques
M.Sc Mathematics
II M.Sc Mathematics
2024-2025
2024-2025
(Even Semester)
 (Even Semester)

Resource Management Techniques

Linear Programming

1.1 Principal Components of Decision Problem

1.2 Modeling Phases
1.3 LP Formulation
1.4 Graphic Solution
1.5 Resource Allocation Problems
1.6 Simplex method
1.7 Artificial Starting Solution
1.8 Sensitivity Analysis
1.9 Algebraic Sensitivity Analysis

1.1 Principal Components of a Decision Problem

1. Definition of Decision Variables

Decision variables are the variables that are under the control of the decision-maker. They are the
variables that are adjusted to achieve the objective.

 Characteristics of decision variables:

o They are controllable.
o They are quantitative (can be measured).
o They are relevant to the decision problem.
 Examples of decision variables:
o Number of units produced.
o Amount of resource allocated.
o Price of a product.
o Production level.

2. Objective Function

The objective function is a mathematical expression that describes the objective of the decision
problem. It is a function of the decision variables.

 Characteristics of an objective function:

o It is a mathematical expression.
o It is a function of the decision variables.
o It represents the objective of the decision problem.
 Types of objective functions:
o Maximization: The goal is to maximize the objective function (e.g., maximize profit).
o Minimization: The goal is to minimize the objective function (e.g., minimize cost).
 Examples of objective functions:
o Maximize profit = Revenue - Cost.
o Minimize cost = Labor cost + Material cost.

3. Constraints

Constraints are limitations on the values of the decision variables. They represent the restrictions
on the decision variables.

 Types of constraints:
o Less-than-or-equal-to (≤) constraints.
o Greater-than-or-equal-to (≥) constraints.
o Equality (=) constraints.
 Examples of constraints:
o Limited resources (e.g., labor, materials).
o Minimum or maximum production levels.
o Safety regulations.
o Environmental regulations.

By understanding these three components, you can formulate a decision problem in Linear
Programming and solve it to find the optimal solution.
1.2 Modeling Phases
1. Problem Definition

 Identify the problem: Clearly articulate the problem or opportunity, including the key issues
and challenges.
 Define the goals: Determine what you want to achieve, such as maximizing profit, minimizing
cost, or optimizing resource allocation.
 Specify the objectives: Identify the key performance indicators (KPIs) that will measure
success, such as return on investment (ROI), productivity, or customer satisfaction.
 Define the scope: Determine the boundaries of the problem, including the stakeholders,
resources, and timelines involved.

2. Data Collection

 Gather relevant data: Collect data on the decision variables, objective function, and constraints,
including historical data, industry benchmarks, and market trends.
 Identify data sources: Determine where the data will come from, such as internal records,
external sources, or surveys.
 Ensure data accuracy: Verify the accuracy and reliability of the data, including checking for
errors, inconsistencies, and missing values.
 Organize the data: Structure the data in a way that facilitates analysis, such as creating
spreadsheets, databases, or data visualizations.

3. Model Formulation

 Define the decision variables: Identify the variables that will be adjusted to achieve the
objective, such as production levels, resource allocation, or pricing.
 Formulate the objective function: Develop a mathematical expression that represents the
objective, such as maximizing profit or minimizing cost.
 Specify the constraints: Identify the limitations on the decision variables, such as resource
availability, demand constraints, or regulatory requirements.
 Formulate the constraints: Develop mathematical expressions that represent the constraints,
such as linear equations or inequalities.

4. Model Solution

 Choose a solution method: Select a method to solve the LP model, such as the graphical
method, simplex method, or interior-point method.
 Solve the model: Use the chosen method to find the optimal solution, including the values of the
decision variables and the objective function.
 Verify the solution: Check the solution for feasibility and optimality, including checking for
errors, inconsistencies, or violations of constraints.

5. Model Interpretation

 Analyze the results: Examine the optimal solution and understand what it means, including the
implications for the organization, stakeholders, and resources.
 Make decisions: Use the results to make informed decisions, including determining the best
course of action, allocating resources, or setting priorities.
 Refine the model: Revise the model as needed to reflect changes or new information, including
updating data, revising assumptions, or incorporating new constraints.
 Monitor and evaluate: Continuously monitor and evaluate the performance of the model,
including tracking key performance indicators (KPIs), identifying areas for improvement, and
refining the model as needed.

1.3 LP Formulation:
Step 1: Define the Decision Variables

 Decision variables are the variables that are under the control of the decision-maker.
 They are the variables that are adjusted to achieve the objective.
 Example: number of units produced (x), amount of resource allocated (y).

Step 2: Define the Objective Function

 The objective function is a mathematical expression that describes the objective of the decision
problem.
 It is a function of the decision variables.
 Example: Maximize profit = 10x + 20y.

Step 3: Define the Constraints

 Constraints are limitations on the values of the decision variables.

 They represent the restrictions on the decision variables.
 Example: Limited resources (2x + 3y ≤ 100), minimum production level (x ≥ 20).

Step 4: Formulate the LP Model

 Use mathematical equations and inequalities to formulate the LP model.

 Include the objective function, decision variables, and constraints.
 Example:

Maximize: 10x + 20y

Subject to:
2x + 3y ≤ 100
x ≥ 20
x, y ≥ 0
Key Elements of an LP Model

 Decision variables: x, y
 Objective function: Maximize 10x + 20y
 Constraints: 2x + 3y ≤ 100, x ≥ 20
 Non-negativity constraints: x, y ≥ 0

By following these steps, you can formulate a basic LP model that can be solved using various
methods, including graphical methods, simplex methods, or interior-point methods.
1.4 Graphic Solution
Step 1: Graphical Representation

 Plot the constraints on a graph, using the x-axis and y-axis to represent the decision variables.
 Draw the constraint lines, using solid lines for ≤ and ≥ constraints, and dashed lines for =
constraints.

Step 2: Identify the Feasible Region

 Shade the area that satisfies all the constraints, including the non-negativity constraints.
 The feasible region is the area where all the constraints are satisfied.
 Step 3: Plot the Objective Function

 Plot the objective function line on the graph, using a dashed line.
 Move the objective function line parallel to itself, to find the optimal solution.

Step 4: Find the Optimal Solution

 The optimal solution is the point where the objective function line intersects the feasible region.
 If the objective function line intersects the feasible region at a vertex, that vertex is the optimal
solution.
 If the objective function line intersects the feasible region along an edge, any point along that
edge is an optimal solution.

Example

Suppose we have the following LP problem:

Maximize: 2x + 3y
Subject to:
x + 2y ≤ 4
x+y≥2
x, y ≥ 0
To solve this problem graphically:

1. Plot the constraints on a graph.

2. Identify the feasible region.
3. Plot the objective function line.
4. Move the objective function line parallel to itself to find the optimal solution.

The optimal solution is the point where the objective function line intersects the feasible region.
Advantages of Graphic Solution

 Easy to visualize and understand the problem.

 Can be used to solve small-scale LP problems.
 Helps to identify the feasible region and optimal solution.

Limitations of Graphic Solution

 Can only be used for problems with two decision variables.

 Not suitable for large-scale LP problems.
 Can be time-consuming and tedious to plot the constraints and objective function.
 1.5 Resource Allocation Problems
Definition

Resource Allocation Problems involve allocating limited resources to meet specific demands or
objectives. The goal is to optimize the allocation of resources to achieve the best possible
outcome.
Characteristics

 Limited resources: Resources are scarce and limited.

 Multiple demands: There are multiple demands or objectives that need to be met.
 Optimization: The goal is to optimize the allocation of resources.

Examples

1. Production Planning: A manufacturing company has limited production capacity and

needs to allocate it among different products to meet demand.
2. Resource Allocation in Projects: A project manager has limited resources (e.g.,
personnel, equipment) and needs to allocate them among different tasks to meet project
deadlines.
3. Budget Allocation: A company has a limited budget and needs to allocate it among
different departments or projects.
4. Supply Chain Optimization: A company needs to allocate limited transportation
resources to meet demand from different customers.

LP Formulation

To formulate a Resource Allocation Problem as an LP model, you need to:

1. Define the decision variables (e.g., amount of resource allocated to each demand).
2. Formulate the objective function (e.g., maximize total value or minimize total cost).
3. Specify the constraints (e.g., limited resources, demand requirements).

Example LP Model

Suppose a company has 100 units of resource to allocate among three demands: A, B, and C. The
objective is to maximize the total value.
Maximize: 10x + 20y + 30z
Subject to:
x + y + z ≤ 100
x ≥ 20
y ≥ 30
x, y, z ≥ 0
In this example, x, y, and z represent the amount of resource allocated to demands A, B, and C,
respectively. The objective function maximizes the total value, and the constraints ensure that the
limited resource is allocated efficiently.

1.6 The Simplex Method

Overview

The Simplex Method is an iterative algorithm that finds the optimal solution to an LP problem by
moving from one basic feasible solution to another, improving the objective function value at
each step.
Key Steps

1. Initialization: Convert the LP problem into standard form, and find an initial basic
feasible solution.
2. Pivot Operation: Choose a pivot element, and perform a pivot operation to move from
one basic feasible solution to another.
3. Iteration: Repeat the pivot operation until the optimal solution is found.
4. Optimality Test: Check if the current basic feasible solution is optimal. If it is, stop.
Otherwise, continue with the next iteration.

Simplex Method Example

Suppose we have the following LP problem:

Maximize: 2x + 3y
Subject to:
x + 2y ≤ 4
x+y≥2
x, y ≥ 0
To solve this problem using the Simplex Method:

1. Initialization: Convert the problem into standard form, and find an initial basic feasible
solution.
2. Pivot Operation: Choose a pivot element, and perform a pivot operation to move from
one basic feasible solution to another.
3. Iteration: Repeat the pivot operation until the optimal solution is found.
Advantages of the Simplex Method

1. Efficient: The Simplex Method is relatively fast and efficient, especially for small- to
medium-sized LP problems.
2. Easy to Implement: The Simplex Method is relatively easy to implement, especially
when using specialized software or libraries.
3. Flexible: The Simplex Method can be used to solve a wide range of LP problems,
including those with multiple objectives and constraints.

Limitations of the Simplex Method

1. Computational Complexity: The Simplex Method can be computationally expensive for

large LP problems, especially those with many variables and constraints.
2. Degeneracy: The Simplex Method can encounter degeneracy issues, where the algorithm
gets stuck in an infinite loop or fails to converge.
3. Numerical Instability: The Simplex Method can be sensitive to numerical instability,
especially when dealing with ill-conditioned matrices or large numbers.

1.7 Artificial Starting Solution

Why is an Artificial Starting Solution needed?

The Simplex algorithm requires an initial basic feasible solution to start the iterative process.
However, in many cases, finding an initial basic feasible solution can be challenging.
What is an Artificial Starting Solution?

An Artificial Starting Solution is a technique used to create an initial basic feasible solution by
introducing artificial variables into the problem.
How is an Artificial Starting Solution created?

To create an Artificial Starting Solution:

1. Introduce artificial variables: Add artificial variables to the problem, one for each
constraint.
2. Create a new objective function: Create a new objective function that minimizes the
sum of the artificial variables.
3. Solve the new problem: Solve the new problem using the Simplex algorithm.
4. Eliminate the artificial variables: Once an initial basic feasible solution is found,
eliminate the artificial variables from the problem.

Example

Suppose we have the following LP problem:

Maximize: 2x + 3y
Subject to:
x + 2y ≤ 4
x+y≥2
x, y ≥ 0
To create an Artificial Starting Solution:

1. Introduce artificial variables: Add artificial variables A1 and A2 to the problem.

2. Create a new objective function: Create a new objective function that minimizes the
sum of A1 and A2.
3. Solve the new problem: Solve the new problem using the Simplex algorithm.
4. Eliminate the artificial variables: Once an initial basic feasible solution is found,
eliminate A1 and A2 from the problem.

Advantages of Artificial Starting Solution

1. Guarantees an initial basic feasible solution: The Artificial Starting Solution

guarantees an initial basic feasible solution, which is essential for starting the Simplex
algorithm.
2. Easy to implement: The Artificial Starting Solution is relatively easy to implement,
especially when using specialized software or libraries.
3. Flexibility: The Artificial Starting Solution can be used with various LP problems,
including those with multiple objectives and constraints.

1.8 Sensitivity Analysis

What is Sensitivity Analysis?

Sensitivity Analysis is a technique used to analyze how changes in input parameters affect the
optimal solution of an LP problem.
Why is Sensitivity Analysis important?

Sensitivity Analysis is essential because:

1. Uncertainty: Input parameters are often uncertain or subject to change.

2. Robustness: Sensitivity Analysis helps to understand the robustness of the optimal
solution.
3. Decision-making: Sensitivity Analysis provides valuable insights for decision-makers.

Types of Sensitivity Analysis

1. Parameter Sensitivity Analysis: Analyzes how changes in input parameters affect the
optimal solution.
2. Scenario Analysis: Analyzes how different scenarios or outcomes affect the optimal
solution.
 3. Tolerance Analysis: Analyzes how much a parameter can change before the optimal
solution changes.

How to perform Sensitivity Analysis?

1. Identify the input parameters: Identify the input parameters that are subject to change
or uncertainty.
2. Analyze the effect of changes: Analyze how changes in the input parameters affect the
optimal solution.
3. Interpret the results: Interpret the results of the Sensitivity Analysis to understand the
robustness of the optimal solution.

Example

Suppose we have the following LP problem:

Maximize: 2x + 3y
Subject to:
x + 2y ≤ 4
x+y≥2
x, y ≥ 0

1. Identify the input parameters: Identify the input parameters, such as the coefficients of
the objective function and the constraints.
2. Analyze the effect of changes: Analyze how changes in the input parameters affect the
optimal solution.
3. Interpret the results: Interpret the results of the Sensitivity Analysis to understand the
robustness of the optimal solution.
4. Sensitivity Analysis:

Steps involved in Sensitivity Analysis

The steps involved in Sensitivity Analysis are:

1. Identify the input parameters: Identify the input parameters that are subject to change
or uncertainty.
2. Determine the range of values: Determine the range of values for each input parameter.
3. Analyze the effect of changes: Analyze how changes in each input parameter affect the
optimal solution.
4. Interpret the results: Interpret the results of the Sensitivity Analysis to understand the
robustness of the optimal solution.

Methods used in Sensitivity Analysis

Several methods are used in Sensitivity Analysis:

 1. Graphical Method: This method involves graphing the optimal solution and analyzing
how changes in the input parameters affect the graph.
2. Tabular Method: This method involves creating a table that shows how changes in the
input parameters affect the optimal solution.
3. Computational Method: This method involves using computational software to analyze
how changes in the input parameters affect the optimal solution.

Importance of Sensitivity Analysis

Sensitivity Analysis is important because:

1. Uncertainty: Input parameters are often uncertain or subject to change.

2. Robustness: Sensitivity Analysis helps to understand the robustness of the optimal
solution.
3. Decision-making: Sensitivity Analysis provides valuable insights for decision-makers.

Limitations of Sensitivity Analysis

The limitations of Sensitivity Analysis are:

1. Assumes linearity: Sensitivity Analysis assumes that the relationship between the input
parameters and the optimal solution is linear.
2. Ignores interactions: Sensitivity Analysis ignores interactions between input parameters.
3. Requires accurate data: Sensitivity Analysis requires accurate data on the input
parameters.

1.9 Algebraic Sensitivity Analysis

Algebraic Sensitivity Analysis is a mathematical approach to sensitivity analysis that uses
algebraic techniques to analyze the sensitivity of the optimal solution.
Techniques Used

1. Partial Derivatives: Analyzes how changes in input parameters affect the optimal
solution using partial derivatives.
2. Taylor Series Expansion: Approximates the optimal solution using a Taylor series
expansion.
3. Matrix Algebra: Uses matrix algebra to analyze the sensitivity of the optimal solution.

Advantages of Algebraic Sensitivity Analysis

1. Mathematical Rigor: Provides a mathematically rigorous approach to sensitivity

analysis.
2. Flexibility: Can be applied to a wide range of LP problems.
3. Insight: Provides valuable insights into the sensitivity of the optimal solution.

Applications of Algebraic Sensitivity Analysis

 1. Resource Allocation: Helps to analyze the sensitivity of resource allocation decisions.
2. Supply Chain Optimization: Analyzes the sensitivity of supply chain optimization
decisions.
3. Financial Planning: Helps to analyze the sensitivity of financial planning decisions.

Unit II
DUALITY AND NETWORKS

2.1 Definition of Dual Problem

2.2 Primal
2.3 Dual Relation Ships
2.4 Dual Simplex Methods
2.5 Post Optimality Analysis
2.6 Transportation Model
2.7 Assignment Model
2.8 The Hungarian Model
2.9 Shortest Route Problem
2.10 LP Formulation of Shortest Route Problem

2.1 Definition of Dual Problem

The Dual Problem is a mathematical concept in Linear Programming (LP) that involves creating
a second LP problem, known as the dual, from the original problem, known as the primal.
Key Characteristics

1. Variables: The dual problem has a variable for each constraint in the primal problem.
These variables are known as dual variables.
2. Constraints: The dual problem has a constraint for each variable in the primal problem.
These constraints are known as dual constraints.
3. Objective Function: The dual problem's objective function is derived from the primal
problem's objective function. The dual objective function is to maximize or minimize the
dual objective function.

Relationship Between Primal and Dual Problems

The primal and dual problems are related in the following ways:
 1. Weak Duality: The primal problem's objective function value is greater than or equal to
the dual problem's objective function value.
2. Strong Duality: The primal problem's objective function value is equal to the dual
problem's objective function value.
3. Complementary Slackness: The primal and dual variables are related through
complementary slackness conditions.

Advantages of Dual Problem

The dual problem has several advantages:

1. Simpler Problem: The dual problem can be simpler to solve than the primal problem.
2. Insight into Primal Problem: The dual problem provides insight into the primal
problem.
3. Sensitivity Analysis: The dual problem can be used to perform sensitivity analysis on the
primal problem.

Example of Dual Problem

Suppose we have the following primal problem:

Maximize: 2x + 3y
Subject to:
x + 2y ≤ 4
x+y≥2
x, y ≥ 0
The dual problem would be:
Minimize: 4w + 2z
Subject to:
w+z≥2
2w + z ≥ 3
w, z ≥ 0
Note that the dual problem has variables w and z, which correspond to the constraints in the
primal problem. The dual problem also has constraints that correspond to the variables in the
primal problem
2.2 Definition of Primal Problem
The Primal Problem is the original Linear Programming (LP) problem that is being solved. It
represents the real-world problem that we are trying to optimize.
Key Characteristics of Primal Problem
 1. Variables: The primal problem has variables that represent the decisions being made.
These variables are known as primal variables or decision variables.
2. Constraints: The primal problem has constraints that represent the limitations on the
decisions. These constraints can be equalities or inequalities.
3. Objective Function: The primal problem's objective function represents the goal being
optimized. The objective function can be either maximization or minimization.

Types of Primal Problems

There are several types of primal problems:

1. Maximization Problem: The goal is to maximize the objective function.

2. Minimization Problem: The goal is to minimize the objective function.
3. Mixed-Integer Programming (MIP) Problem: The problem involves both continuous
and integer variables.
4. Binary Integer Programming (BIP) Problem: The problem involves only binary
variables.

Example of Primal Problem

Suppose we have the following primal problem:

Maximize: 2x + 3y
Subject to:
x + 2y ≤ 4
x+y≥2
x, y ≥ 0
In this example, x and y are the primal variables, and the constraints represent the limitations on
the decisions. The objective function represents the goal being optimized, which is to maximize
the value of 2x + 3y.
Solving the Primal Problem

The primal problem can be solved using various methods, including:

1. Graphical Method: A graphical method can be used to solve small-scale primal

problems.
2. Simplex Method: The simplex method is a popular algorithm for solving primal
problems.
3. Interior Point Method: The interior point method is another algorithm for solving
primal problems.
2.3 Dual Relationships:
Definition of Dual Relationships

The Dual Relationships refer to the connections between the primal and dual problems. These
relationships provide a deeper understanding of the primal and dual problems and are essential
for solving Linear Programming (LP) problems.
Key Relationships

There are three key relationships between the primal and dual problems:

1. Weak Duality: The primal problem's objective function value is greater than or equal to
the dual problem's objective function value. This relationship is also known as the weak
duality theorem.
2. Strong Duality: The primal problem's objective function value is equal to the dual
problem's objective function value. This relationship is also known as the strong duality
theorem.
3. Complementary Slackness: The primal and dual variables are related through
complementary slackness conditions. This relationship provides a connection between the
primal and dual variables.

Weak Duality

Weak duality states that the primal problem's objective function value is greater than or equal to
the dual problem's objective function value. This relationship can be expressed mathematically
as:
Primal Objective Function Value ≥ Dual Objective Function Value
Strong Duality

Strong duality states that the primal problem's objective function value is equal to the dual
problem's objective function value. This relationship can be expressed mathematically as:
Primal Objective Function Value = Dual Objective Function Value
Complementary Slackness

Complementary slackness provides a connection between the primal and dual variables. It states
that the primal and dual variables are related through the following conditions:

1. Primal Slackness: If a primal constraint is not binding, then the corresponding dual
variable is zero.
2. Dual Slackness: If a dual constraint is not binding, then the corresponding primal
variable is zero.
Example of Dual Relationships

Suppose we have the following primal problem:

Maximize: 2x + 3y
Subject to:
x + 2y ≤ 4
x+y≥2
x, y ≥ 0
The dual problem would be:
Minimize: 4w + 2z
Subject to:
w+z≥2
2w + z ≥ 3
w, z ≥ 0
In this example, the primal and dual problems are related through the weak duality, strong
duality, and complementary slackness conditions.
Importance of Dual Relationships

The dual relationships are essential for solving LP problems because they:

1. Provide a connection between the primal and dual problems: The dual relationships
provide a connection between the primal and dual problems, which is essential for
solving LP problems.
2. Help to identify the optimal solution: The dual relationships help to identify the optimal
solution to the LP problem.
3. Provide insight into the primal and dual problems: The dual relationships provide
insight into the primal and dual problems, which is essential for understanding the LP
problem.

2.4 Dual Simplex Methods:

Definition of Dual Simplex Methods

The Dual Simplex Methods are algorithms used to solve Linear Programming (LP) problems
using the dual problem. These algorithms are based on the simplex method, but they start with
the dual problem instead of the primal problem.
Key Characteristics of Dual Simplex Methods
 1. Starting with the Dual: The algorithm starts with the dual problem and iteratively
improves the dual solution.
2. Pivot Operations: The algorithm uses pivot operations to move from one dual solution
to another. Pivot operations involve exchanging one basic variable with another.
3. Optimality Test: The algorithm checks for optimality using the complementary
slackness conditions. The complementary slackness conditions provide a connection
between the primal and dual variables.

How Dual Simplex Methods Work

Here's a step-by-step overview of how Dual Simplex Methods work:

1. Initialize the Dual Solution: Initialize the dual solution by setting the dual variables to
zero.
2. Check for Optimality: Check if the current dual solution is optimal using the
complementary slackness conditions.
3. Perform Pivot Operation: If the current dual solution is not optimal, perform a pivot
operation to move to a new dual solution.
4. Update the Dual Solution: Update the dual solution by exchanging one basic variable
with another.
5. Repeat Steps 2-4: Repeat steps 2-4 until the optimal dual solution is found.

Advantages of Dual Simplex Methods

Dual Simplex Methods have several advantages:

1. Efficient: Dual Simplex Methods are efficient and can solve large-scale LP problems
quickly.
2. Robust: Dual Simplex Methods are robust and can handle degenerate and infeasible
problems.
3. Flexible: Dual Simplex Methods can be used to solve a wide range of LP problems,
including those with multiple objectives and constraints.

Disadvantages of Dual Simplex Methods

Dual Simplex Methods also have some disadvantages:

1. Complexity: Dual Simplex Methods can be complex and difficult to implement.

2. Computational Requirements: Dual Simplex Methods require significant
computational resources, especially for large-scale problems.
3. Limited Applicability: Dual Simplex Methods are limited to solving LP problems and
cannot be used to solve other types of optimization problems.
2.5 Post-Optimality Analysis:
What is Post-Optimality Analysis?

Post-optimality analysis, also known as sensitivity analysis or post-optimal analysis, is a

technique used to examine how changes in the input parameters of a linear programming
problem affect the optimal solution.
Why is Post-Optimality Analysis Important?

Post-optimality analysis is crucial because it helps decision-makers understand:

1. How sensitive the optimal solution is to changes in the input parameters.

2. Which input parameters have the most significant impact on the optimal solution.
3. What changes can be made to the input parameters to improve the optimal solution.

Types of Post-Optimality Analysis

There are several types of post-optimality analysis:

1. Sensitivity Analysis: Analyzing how changes in the objective function coefficients affect
the optimal solution.
2. Parametric Analysis: Analyzing how changes in the right-hand side values (resource
availability) affect the optimal solution.
3. Range Analysis: Analyzing the range of values for which the optimal solution remains
unchanged.
4. Shadow Price Analysis: Analyzing the shadow prices (opportunity costs) of the
resources.

Steps Involved in Post-Optimality Analysis

The steps involved in post-optimality analysis are:

1. Solve the original linear programming problem.

2. Identify the input parameters to be analyzed.
3. Perform sensitivity analysis, parametric analysis, range analysis, or shadow price
analysis.
4. Analyze the results and draw conclusions.

Example

Suppose we have a linear programming problem:

Maximize: 2x + 3y
Subject to:
x + 2y ≤ 4
 3x + 2y ≤ 5
x, y ≥ 0
After solving the problem, we get the optimal solution: x = 1, y = 1.
Now, let's perform sensitivity analysis on the objective function coefficient of x. We want to
know how much the coefficient can change before the optimal solution changes.
After performing the analysis, we find that the coefficient can change by ±10% before the
optimal solution changes.
This information can be useful for decision-makers to understand the sensitivity of the optimal
solution to changes in the input parameters.

2.6 Transportation Model:

What is the Transportation Model?

The transportation model is a type of linear programming problem that deals with finding the
most efficient way to transport goods from sources to destinations. It's a classic problem in
Operations Research and Management Science.
Characteristics of the Transportation Model

1. Sources and Destinations: The model involves multiple sources (e.g., factories,
warehouses) and multiple destinations (e.g., customers, retail stores).
2. Transportation Costs: Each source-destination pair has a associated transportation cost.
3. Supply and Demand: Each source has a limited supply of goods, and each destination
has a specific demand for goods.
4. Objective: The objective is to minimize the total transportation cost while satisfying the
supply and demand constraints.

Formulation of the Transportation Model

The transportation model can be formulated as a linear programming problem:

Minimize: ∑∑ c_ij x_ij (total transportation cost)
Subject to:
∑ x_ij ≤ S_i (supply constraint at source i)
∑ x_ij ≥ D_j (demand constraint at destination j)
x_ij ≥ 0 (non-negativity constraint)
where:

 c_ij is the transportation cost from source i to destination j

 x_ij is the amount of goods transported from source i to destination j
 S_i is the supply at source i
 D_j is the demand at destination j
Solution Methods

The transportation model can be solved using various methods, including:

1. Northwest Corner Method: A simple and intuitive method for finding an initial feasible
solution.
2. Least Cost Method: A method that selects the source-destination pair with the lowest
transportation cost.
3. Vogel's Approximation Method (VAM): A method that uses a penalty function to guide
the search for an optimal solution.
4. Linear Programming: The transportation model can be solved using linear
programming techniques, such as the simplex method.

Applications

The transportation model has numerous applications in:

1. Logistics and Supply Chain Management: Optimizing the transportation of goods from
suppliers to customers.
2. Manufacturing: Optimizing the transportation of raw materials to production facilities.
3. Retailing: Optimizing the transportation of goods from warehouses to retail stores.

2.7 Assignment Model:

What is the Assignment Model?

The assignment model is a special type of transportation model that involves assigning tasks or
jobs to resources. It's a classic problem in Operations Research and Management Science.
Characteristics of the Assignment Model

1. Tasks or Jobs: The model involves a set of tasks or jobs that need to be assigned to
resources.
2. Resources: The model involves a set of resources (e.g., machines, personnel) that can
perform the tasks or jobs.
3. Cost or Profit: Each task-resource pair has an associated cost or profit.
4. Objective: The objective is to minimize the total cost or maximize the total profit while
ensuring that each task is assigned to exactly one resource and each resource is assigned
to at most one task.

Formulation of the Assignment Model

The assignment model can be formulated as a linear programming problem:

Minimize: ∑∑ c_ij x_ij (total cost)
Subject to:
∑ x_ij = 1 (each task is assigned to exactly one resource)
 x_ij ∈ {0, 1} (assignment variable)
∑ x_ij ≤ 1 (each resource is assigned to at most one task)

where:

 c_ij is the cost of assigning task i to resource j

 x_ij is the assignment variable (1 if task i is assigned to resource j, 0 otherwise)

Solution Methods

The assignment model can be solved using various methods, including:

1. Hungarian Algorithm: A efficient algorithm for solving assignment problems.

2. Linear Programming: The assignment model can be solved using linear programming
techniques, such as the simplex method.

Applications

The assignment model has numerous applications in:

1. Workforce Scheduling: Assigning tasks to workers based on their skills and availability.
2. Machine Scheduling: Assigning jobs to machines based on their processing times and
capacities.
3. Resource Allocation: Assigning resources to tasks based on their availability and
requirements.

2.8 Hungarian Model:

What is the Hungarian Model?

The Hungarian model, also known as the Hungarian algorithm, is a combinatorial optimization
algorithm used to solve assignment problems efficiently. It was developed by Harold Kuhn in
1955 and is considered one of the most efficient algorithms for solving assignment problems.
How Does the Hungarian Model Work?

The Hungarian model works by:

1. Creating a cost matrix: A matrix is created where the rows represent the tasks or jobs,
and the columns represent the resources or agents. The entries in the matrix represent the
cost of assigning a task to a resource.
2. Finding the maximum number of independent zeros: The algorithm finds the
maximum number of independent zeros in the cost matrix. A zero is considered
independent if it is not in the same row or column as another zero.
3. Covering the zeros: The algorithm covers the independent zeros with lines, where each
line covers a row or column.
 4. Finding the smallest uncovered value: The algorithm finds the smallest uncovered
value in the cost matrix.
5. Augmenting the path: The algorithm augments the path by subtracting the smallest
uncovered value from all uncovered elements and adding it to all elements that are
covered twice.

Steps Involved in the Hungarian Model

The steps involved in the Hungarian model are:

1. Initialize the cost matrix: Create a cost matrix where the rows represent the tasks or
jobs, and the columns represent the resources or agents.
2. Find the maximum number of independent zeros: Find the maximum number of
independent zeros in the cost matrix.
3. Cover the zeros: Cover the independent zeros with lines, where each line covers a row or
column.
4. Find the smallest uncovered value: Find the smallest uncovered value in the cost
matrix.
5. Augment the path: Augment the path by subtracting the smallest uncovered value from
all uncovered elements and adding it to all elements that are covered twice.
6. Repeat steps 3-5: Repeat steps 3-5 until an optimal assignment is found.

Advantages of the Hungarian Model

The Hungarian model has several advantages:

1. Efficient: The Hungarian model is an efficient algorithm for solving assignment

problems.
2. Optimal solution: The Hungarian model guarantees an optimal solution.
3. Simple to implement: The Hungarian model is simple to implement.

Applications of the Hungarian Model

The Hungarian model has numerous applications:

1. Assignment problems: The Hungarian model is used to solve assignment problems,

where tasks or jobs need to be assigned to resources or agents.
2. Scheduling: The Hungarian model is used in scheduling, where tasks or jobs need to be
assigned to time slots.
3. Resource allocation: The Hungarian model is used in resource allocation, where
resources need to be allocated to tasks or jobs.

2.9 Shortest Route Problem and its LP Formulation:

Shortest Route Problem
The shortest route problem involves finding the minimum-distance path between two nodes in a
network. This problem is fundamental to many fields, including:

1. Transportation: Finding the shortest route between two cities.

2. Logistics: Finding the shortest route for delivery trucks.
3. Telecommunications: Finding the shortest path for data transmission.

2.10 LP Formulation of Shortest Route Problem

The shortest route problem can be formulated as a linear programming problem. The goal is to
minimize the total distance traveled while satisfying the constraints:

1. Flow conservation: The flow into each node must equal the flow out of each node.
2. Non-negativity: The flow on each arc must be non-negative.

The LP formulation involves:

1. Decision variables: x_ij, representing the flow on arc (i, j).

2. Objective function: Minimize the total distance traveled.
3. Constraints: Flow conservation, non-negativity, and bounds on the decision variables.

Steps to Formulate the LP

To formulate the LP, follow these steps:

1. Define the decision variables: x_ij, representing the flow on arc (i, j).
2. Define the objective function: Minimize the total distance traveled.
3. Define the constraints: Flow conservation, non-negativity, and bounds on the decision
variables.
4. Specify the bounds: Specify the lower and upper bounds for each decision variable.

Example

Suppose we have a network with 4 nodes and 5 arcs:

Node 1 → Node 2 (distance 3)
Node 1 → Node 3 (distance 2)
Node 2 → Node 4 (distance 4)
Node 3 → Node 4 (distance 1)
Node 3 → Node 2 (distance 5)
The LP formulation would involve:
Decision variables: x_12, x_13, x_24, x_34, x_32
Objective function: Minimize 3x_12 + 2x_13 + 4x_24 + x_34 + 5x_32
Constraints:
Flow conservation: x_12 + x_13 = x_24 + x_34
Non-negativity: x_ij ≥ 0 for all i, j
Solving the LP

The LP can be solved using various methods, including:

1. Simplex method: A popular method for solving LPs.

2. Interior point methods: A class of methods that solve LPs by iterating through the
interior of the feasible region.

Interpretation of Results

Once the LP is solved, the results can be interpreted as follows:

1. Optimal solution: The values of the decision variables that minimize the total distance
traveled.
2. Shadow prices: The shadow prices of the constraints, which represent the change in the
objective function value for a unit change in the constraint.