Optimization Methods in Engineering

Unit-01
Introduction

Dr. Deepak Sharma

Assistant Professor
Department of Mechanical Engineering
National Institute of Technology Hamirpur, H.P., India
Topics to be Covered
Introduction: Introduction, Terminologies, Design Variables and Constraints, Objective Function, Variable Bounds, and Problem Formulation.
Gradient Based Methods: Newton-Raphson Method, Bisection Method, Secant Method. Multi-variable Optimization, Algorithms: Optimality
Criteria, Unidirectional Search, Direct Search Methods: Box Method, Hooke-Jeeves Pattern Search Method, Powell's Conjugate Direction
Method, Cauchy's Steepest Descent Method, Newton's Method. Marquan Method, Conjugate Gradient Method, Variable-Metric (DFP)
Method.
Constrained Optimization Methods: Kuhn Tucker Conditions, Transformation Methods: Penn Function Method, Method of Multipliers
(MOM), and Sensitivity Analysis.
Specialized Optimization Methods: Integer Programming: Penalty Function Method, Branch and Bo‘ Method, Geometric Programming.
Non-Traditional Optimization Methods: Genetic Algorithms, Simulated Annealing, Tabu Search and Ant Colony Optimization, Particle Swarm
Optimization; Applications to Engineering Optimization Problems.

References:
1. Optimization for Engineering Design: Algorithms and Examples by Kalyanmoy Deb, PHI Publication.
2. Engineering Optimization: Theory and Practice by S.S Rao, New International (P) Publication.
3. Engineering Optimization - Methods and Applications by Ravindran, Ragsdell and Rekla, John Wiley & Sons Publication.
4. Multi-Objective Optimization using Evolutionary Algorithms by Kalyanmoy Deb, Wiley Publication
 TABLE OF CONTENTS

01 Introduction 04 Objective Function

02 Terminologies 05 Variable Bounds

03 Design Variables and Constraints 06 Problem Formulation

 INTRODUCTION
 Optimization is the act of obtaining the best result under given circumstances.

 In design, construction, and maintenance of any engineering system, engineers have to take many technological and

managerial decisions at several stages.

 The ultimate goal of all such decisions is either to minimize the effort required or to maximize the desired benefit.

 Since the effort required or the benefit desired in any practical situation can be expressed as a function of certain decision

variables.

 So optimization can be defined as the process of finding the conditions that give the maximum or minimum value of a

function.

 There is no single method available for solving all optimization problems efficiently.
 ENGINEERING APPLICATIONS OF OPTIMIZATION
Some typical applications from different engineering disciplines indicate the wide scope of the subject:

1. Design of aircraft and aerospace structures for minimum weight

2. Finding the optimal trajectories of space vehicles

3. Design of civil engineering structures such as frames, foundations, bridges, towers, chimneys, and dams for minimum cost

4. Minimum-weight design of structures for earthquake, wind, and other types of random loading.

5. Design of water resources systems for maximum benefit

6. Optimal plastic design of structures

7. Optimum design of linkages, cams, gears, machine tools, and other mechanical components
 INTRODUCTION

So… what is mathematical optimization, anyway?

“Optimization” comes from the same root as “optimal”, which means
best. When you optimize something, you are “making it best”
 INTRODUCTION
Mathematical Optimization in the “Real World”
Mathematical Optimization is a branch of applied mathematics which is useful in many different fields. Here are a few
examples:
• Manufacturing
• Production
• Inventory control
• Transportation
• Scheduling
• Networks
• Finance
• Engineering
• Mechanics
 OPTIMIZATION PROBLEM

1. Objective Function: 𝒇

2. Design Variables:(𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 )

3. Constraints: 𝒈𝟏 , 𝒈𝟐 … … … 𝒈𝒎 , (𝒉𝟏 , 𝒉𝟐 … … … 𝒉𝒑 )
 OPTIMIZATION PROBLEM
Your basic optimization problem consists of…
• The objective function, f(x), which is the output you’re trying to maximize or minimize.

Minimize objective function Maximize objective function

𝑨𝒓𝒆𝒂 = 𝒂 × 𝒃 𝑨𝒓𝒆𝒂 = 𝒂 × 𝒃
 OPTIMIZATION PROBLEM

OBJECTIVE FUNCTION
 OPTIMIZATION PROBLEM

DECISION VARIABLES
 OPTIMIZATION PROBLEM

DECISION VARIABLES
 OPTIMIZATION PROBLEM

Minimize 𝑓(𝑥1 , 𝑥2 … … … 𝑥𝑛 )
Subject to
𝒈𝟏 𝒇(𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 ) ≥ 𝒃𝟏
𝒈𝟐 𝒇(𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 ) ≥ 𝒃𝟐
Inequality Constraint
……………
𝒈𝒎 𝒇(𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 ) ≥ 𝒃𝒎

𝒉𝟏 𝒇 𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 = 𝒅𝟏
𝒉𝟐 𝒇 𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 = 𝒅𝟐
Equality Constraint
……………
𝒉𝒑 𝒇 𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 = 𝒅𝒑
 OPTIMIZATION PROBLEM

Maximize 𝑓(𝑥1 , 𝑥2 … … … 𝑥𝑛 )
Subject to
𝒈𝟏 𝒇(𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 ) ≤ 𝒃𝟏
𝒈𝟐 𝒇(𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 ) ≤ 𝒃𝟐
Inequality Constraint
……………
𝒈𝒎 𝒇(𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 ) ≤ 𝒃𝒎

𝒉𝟏 𝒇 𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 = 𝒅𝟏
𝒉𝟐 𝒇 𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 = 𝒅𝟐
Equality Constraint
……………
𝒉𝒑 𝒇 𝒙𝟏 , 𝒙𝟐 … … … 𝒙𝒏 = 𝒅𝒑
OPTIMIZATION PROBLEM
Need for Optimization

Choose Design Variables

Formulate Constraints

Formulate Objective
Function

Set up variable bounds

Choose an optimization
algorithm

Obtain solution (s)

 CLASSIFICATION

a) No. of Objectives
i) Single Objective Optimization
ii) Multi Objective Optimization
b) No. of Variables
i) Single Variable Optimization
ii) Multi Variable Optimization
c) Constraints
i) Unconstrained Optimization
ii) Constrained Optimization
 CLASSIFICATION

d) Type of Objective
i) Minimization
ii)Maximization
e) Type of Objective Function
i) Linear Problem
ii) Non-linear Problem
f) Methodology
i) Traditional Methods: Analytical or Numerical
ii) Non-Traditional Methods: Nature inspired numerical methods
 CLASSIFICATION
Traditional Optimization Methods
A. Single Variable Optimization Methods
1. Direct Search Methods
i) Exhaustive search method
ii) Bounding phase method
iii) Internal halving method
iv) Fibonacci search method
v) Golden section method

2. Gradient Based Methods

i) Newton Raphson method
ii) Bisection method
iii) Secant method
 CLASSIFICATION
B. Multivariable Optimization Methods
1. Direct Search Methods
i) Simplex search method
ii) Hooke-Jeeves pattern search method
iii) Powell’s conjugate direction method

2. Gradient Based Methods

i) Cauchy’s steepest descent method
ii) Newton’s method
iii) Marquardt method
 CLASSIFICATION

Traditional Optimization Methods

A. Unconstrained Optimization Methods
B. Constrained Optimization
i) Penalty function method
ii) Method of multipliers
iii) Variable elimination method
iv) Complex search method
v) Random search method
vi) Cutting plane method
 OPTIMIZATION PROBLEM
Types of Optimization Problems
• Some problems have constraints and some do not.
 OPTIMIZATION PROBLEM
Types of Optimization Problems
• Some problems have constraints and some do not.
• There can be one variable or many.
 OPTIMIZATION PROBLEM

Types of Optimization Problems

• Some problems have constraints and some do not.
• There can be one variable or many.
• Variables can be discrete (for example, only have integer values) or continuous.
 OPTIMIZATION PROBLEM

Types of Optimization Problems

• Some problems have constraints and some do not.
• There can be one variable or many.
• Variables can be discrete (for example, only have integer values) or continuous.
• Some problems are static (do not change over time) while some are dynamic (continual adjustments must
be made as changes occur).
 OPTIMIZATION TERMINOLOGIES

1. Constraint: Any restriction, requirement or interaction that limits the values of the decision variables. Also called a general
constraint. Some special constraints include:
• (Closed) linear constraint: A constraint that can be written as a linear function on the decision variables that is set equal to or
less than or equal to a constant. We typically omit stating “closed” which is the condition that only weak inequalities and
equalities are allowed, e.g., x < 5 is not a (closed) linear constraint.
• Variable bounds: An upper or lower bound on a single decision variable, e.g., x ≥ 0 or y ≤ 3 for decision variables x and y.
Note that a variable bound is also a linear constraint.
2. Decision variable: An unknown quantity for an optimization model that represents a decision to be made. In this class we
work with two kinds of decision variables (note these are constraints!):
• Continuous variable: Decision variables constrained to be real numbers.
• Discrete variable: Decision variables constrained to have values from a discrete set. E.g., x ∈ Z restricts x to the set of
integers.
 OPTIMIZATION TERMINOLOGIES
3. Feasible region: The set of allowable values for the decision variables.
4. Feasible problem: An optimization model is feasible if and only if a feasible solution exists for the problem.
5. Feasible solution: A solution to an optimization problem that satisfies all constraints.
6. Integer program: An optimization model where one or more of the decision variables must be integer.
7. Linear function: A function f : R n → R is linear if, and only if, there are constants c1, . . . , cn ∈ R such that for all x ∈ R n ,
f(x) = c1x1 + . . . + cnxn.
8. Linear program: An optimization model where
(a) the decision variables are continuous;
(b) the objective function is linear; and
(c) there are a finite number of (closed) linear constraints, i.e., only equality and weak inequality linear constraints are
permitted.
 LINEAR PROGRAMMING

Linear Programing
 Linear programming is a method of optimizing operations with some constraints.
 The main objective of linear programming is to maximize or minimize the numerical value.
 It consists of linear functions which are subjected to the constraints in the form of linear equations or in the
form of inequalities.

Different types of linear programming

 Solving linear programming by Simplex method.
 Solving linear programming using Big M Method.
 Solving linear programming by Graphical method.
LINEAR PROGRAMMING

Multiple
Solution

Unbounded Feasible
Solution Region

Linear
Programming
Model

Infeasible
No Solution
Solution

Optimal
Solution
 LINEAR PROGRAMMING

Transportation
Problem

Manpower Assignment
Management Problem

Application
Marketing
Management of LPP Marketing
Management

Problems
Production Military
Management Applications

Agriculture
Applications
 LINEAR PROGRAMMING

Simplex Method

When decision variables are more than two, we always use Simplex method.

Slack Variable: Variable added to a (≤) constraint to convert it to an equation (=).

 A slack variable represents unused resources.

 A slack variable contributes nothing to the objective function value.

Surplus Variable: Variable subtracted from a (≥) constraint to convert it to an equation (=).

 A surplus variable represents an excess above a constraint requirement level.

 Surplus variable contributes nothing to the calculated value of the objective function value.
 LINEAR PROGRAMMING

Basic Feasible Solution

It is of two types:

 Degenerate Basic Feasible Solution: If one or more basic variables are zero.

 Non- Degenerate Basic Feasible Solution: All basic variables are non-zero.

Optimal Basic Feasible Solution: A solution that optimizes the objective function.
 LINEAR PROGRAMMING

Steps for Solving the Simplex Problem

i) Convert the given general LPP into standard LPP. If it is minimized then convert it into a problem of

maximization by 𝑀𝑎𝑥𝑍 ′ = −𝑀𝑖𝑛 𝑍.

a) Check all the decision variables are greater than zero.

b) Express the problem in standard form by introducing the slack as well as surplus variable to convert the

inequality constraint into equality equation.

c) All the values of right hand side must be positive.

ii) Write the values of initial basic feasible solution.

 LINEAR PROGRAMMING
iii) Write the standard form of LPP into matrix form.

iv) Construct the initial simplex table.

v) Calculate the values of 𝑍𝑗 − 𝐶𝑗 and check the basic feasible solution for optimality.

𝑍𝑗 − 𝐶𝑗 =𝐶𝐵 𝑋𝑗 − 𝐶𝑗

a) If all 𝑍𝑗 − 𝐶𝑗 ≥ 0 the optimal solution will obtained.

b) If at least one (𝑍𝑗 − 𝐶𝑗 ) is –ve then indicate that column by arrow and that column is called key column.

c) If more than one (𝑍𝑗 − 𝐶𝑗 ) is –ve then choose the most negative of them & then that column is called key

column.
 LINEAR PROGRAMMING

vi) Calculate minimum ratio = 𝑋𝐵ൗ𝐶𝐾 , 𝐶𝑘 = 𝑘𝑒𝑦 𝑐𝑜𝑙𝑢𝑚𝑛 > 0

vii) Construct the new simplex table by entering incoming vector.

viii) Repeat step (v) and (vi)

Q-1. Objective Function Max Z = 3𝑥1 + 2𝑥2 + 5𝑥3

Subject to: 𝑥1 + 2𝑥2 + 𝑥3 ≤ 430
3𝑥1 + 𝑥2 ≤ 460
𝑥1 + 4𝑥2 ≤ 420
𝑥1 , 𝑥2 , 𝑥3 ≥ 0
 LINEAR PROGRAMMING
Sol: By introducing slack variables 𝑠1 , 𝑠2 , 𝑠3 and then convert the problem in standard form.
Max Z = 3𝑥1 + 2𝑥2 + 5𝑥3 + 0. 𝑠1 + 0. 𝑠2 + 0. 𝑠3
𝑥1 + 2𝑥2 + 𝑥3 + 𝑠1 = 430
3𝑥1 + 𝑥2 + 𝑠2 = 460
𝑥1 + 4𝑥2 + 𝑠3 = 420
𝑥1 , 𝑥2 , 𝑥3 , 𝑠1 , 𝑠2 , 𝑠3 ≥ 0
An initial basic feasible solution is given by
𝑥1 = 𝑥2 = 𝑥3 = 0, 𝑠1 = 430, 𝑠2 = 460, 𝑠3 = 420
Writing in Matrix Form AX =B
𝑥1
𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 𝑠3 𝑥2 430
1 2 1 1 0 0 𝑥3
= 460
3 0 2 0 1 0 𝑠1
1 4 0 0 0 1 𝑠2 420
𝑠3
 LINEAR PROGRAMMING
Cj 3 2 5 0 0 0 Min Ratio

CB B XB x1 x2 x3 S1 S2 S3 XB/CK

0 S1 430 1 2 1 1 0 0 430/1 = 430

0 S2 460 3 0 2 0 1 0 460/2 = 230

0 S3 420 1 4 0 0 0 1 420/0 = ∞

Zj-Cj -3 -2 -5 0 0 0

R2 = R2/2, R1= R1-R2

 LINEAR PROGRAMMING
𝑅1 430 1 2 1 1 0 0
𝑅2 230 3/2 0 1 0 1/2 0
𝑅1 -𝑅2 200 -1/2 2 0 1 -1/2 0

Cj 3 2 5 0 0 0 Min Ratio

CB B XB x1 x2 x3 S1 S2 S3 XB/CK

0 S1 200 -1/2 2 0 1 -1/2 0 200/2 = 100

5 x3 230 3/2 0 1 0 1/2 0 230/0 = ∞

0 S3 420 1 4 0 0 0 1 420/4 = 105

Zj-Cj 9/2 -2 0 0 5/2 0

R1 = R1/2, R3= R3-4R1

 LINEAR PROGRAMMING
𝑅3 420 1 4 0 0 0 1
4𝑅1 400 -1 4 0 2 -1 0
𝑅3-4𝑅1 20 2 0 0 -2 1 1

Cj 3 2 5 0 0 0 Min Ratio
CB B XB x1 x2 x3 S1 S2 S3 XB/CK
2 x2 100 -1/4 1 0 1 -1/4 0
5 x3 230 3/2 0 1 0 1/2 0
0 S3 20 2 0 0 0 1 1
Zj-Cj 4 0 0 1 2 0

Since all 𝑍𝑗 − 𝐶𝑗 ≥ 0 the solution is optimum and given by

X2 = 100, X3 = 230, X1 = 0 and value of
Max Z = 𝟐 × 𝟏𝟎𝟎 + 𝟓 × 𝟐𝟑𝟎 + 𝟎 × 𝟐𝟎 = 𝟏𝟑𝟓𝟎
 LINEAR PROGRAMMING
Q-2. Objective Function Max Z = 13𝑥1 + 11𝑥2
Subject to: 4𝑥1 + 5𝑥2 ≤ 1500
5𝑥1 + 3𝑥2 ≤ 1575
𝑥1 + 2𝑥2 ≤ 420
𝑥1 , 𝑥2 , 𝑥3 ≥ 0
Sol: By introducing slack variables 𝑠1 , 𝑠2 , 𝑠3 and then convert the problem in standard form.
Max Z = 13𝑥1 + 11𝑥2 + 0. 𝑠1 + 0. 𝑠2 + 0. 𝑠3
4𝑥1 + 5𝑥2 + 𝑠1 = 1500
5𝑥1 + 3𝑥2 + 𝑠2 = 1575
𝑥1 + 2𝑥2 + 𝑠3 = 420
𝑥1 , 𝑥2 , 𝑥3 , 𝑠1 , 𝑠2 , 𝑠3 ≥ 0
An initial basic feasible solution is given by
𝑥1 = 𝑥2 = 𝑥3 = 0, 𝑠1 = 1500, 𝑠2 = 1575, 𝑠3 = 420
 LINEAR PROGRAMMING
Writing in Matrix Form AX =B
𝑥1
𝑥1 𝑥2 𝑠1 𝑠2 𝑠3 1500
𝑥2
4 5 1 0 0
𝑠1 = 1575
5 3 0 1 0
𝑠2
1 2 0 0 1 420
𝑠3

Cj 13 11 0 0 0 Min Ratio
CB B XB x1 x2 S1 S2 S3 XB/CK
0 S1 1500 4 5 1 0 0 1500/4 = 375
0 S2 1575 5 3 0 1 0 1575/5 = 315
0 S3 420 1 2 0 0 1 420/1 = 420
Zj-Cj -13 -11 0 0 0

R2 = R2/5, R1= R1-4R2 , R3= R3-R2

 LINEAR PROGRAMMING

𝑅1 1500 4 5 1 0 0 𝑅3 420 1 2 0 0 1
4𝑅2 1260 1 12/5 0 4/5 0 𝑅2 315 1 3/5 0 1/5 0
𝑅1 -4𝑅2 240 0 13/5 1 -4/5 0 𝑅3 -𝑅2 105 0 7/5 0 -1/5 1

Cj 13 11 0 0 0 Min Ratio
CB B XB x1 x2 S1 S2 S3 XB/CK
0 S1 240 0 13/5 1 -4/5 0 240*5/13 =92.3
13 x1 315 1 3/5 0 1/5 0 315*3/5 = 525
0 S3 105 0 7/5 0 -1/5 1 105*5/7 = 75
Zj-Cj 0 -16/5 0 13/5 0

R3 = 5*R3/7, R1= R1-13/5R3 , R2= R2-3/5R3

 LINEAR PROGRAMMING

𝑅1 240 0 13/5 1 -4/5 0 𝑅2 315 1 3/5 0 1/5 0

13/5𝑅3 195 0 13/5 0 -13/35 65/35 3/5𝑅3 45 0 3/5 0 -3/35 15/35
𝑅1 -13/5𝑅3 45 0 0 1 -3/7 -13/7
𝑅2 -3/5𝑅3 270 1 0 0 2/7 -3/7

Cj 13 11 0 0 0 Min Ratio
CB B XB x1 x2 S1 S2 S3 XB/CK
0 S1 45 0 0 1 -3/7 -13/7
13 x1 270 1 0 0 2/7 -3/7
11 x2 75 0 1 0 -1/7 5/7
Zj-Cj 0 0 0 15/7 16/7

Since all 𝑍𝑗 − 𝐶𝑗 ≥ 0 the solution is optimum and given by

X2 = 75, X1 = 270 and value of
Max Z = 𝟒𝟑𝟑𝟓
 LINEAR PROGRAMMING
Q-3. Objective Function Max Z = 3𝑥1 + 5𝑥2 + 4𝑥3
Subject to: 2𝑥1 + 3𝑥2 ≤ 8
2𝑥1 + 5𝑥3 ≤ 10
3𝑥1 + 2𝑥2 + 4𝑥3 ≤ 15
𝑥1 , 𝑥2 , 𝑥3 ≥ 0

Sol: By introducing slack variables 𝑠1 , 𝑠2 , 𝑠3 and then convert the problem in standard form.
Max Z = 3𝑥1 + 5𝑥2 + 4𝑥3 + 0. 𝑠1 + 0. 𝑠2 + 0. 𝑠3
2𝑥1 + 3𝑥2 + 𝑠1 = 8
2𝑥1 + 5𝑥3 + 𝑠2 = 10
3𝑥1 + 2𝑥2 + 4𝑥3 + 𝑠3 = 15
𝑥1 , 𝑥2 , 𝑥3 , 𝑠1 , 𝑠2 , 𝑠3 ≥ 0
An initial basic feasible solution is given by
𝑥1 = 𝑥2 = 𝑥3 = 0, 𝑠1 = 8, 𝑠2 = 10, 𝑠3 = 15
 LINEAR PROGRAMMING
Writing in Matrix Form AX =B
𝑥1
𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 𝑠3 𝑥2 8
2 3 0 1 0 0 𝑥3
= 10
0 2 5 0 1 0 𝑠1
3 2 4 0 0 1 𝑠2 15
𝑠3

Cj 3 5 4 0 0 0 Min Ratio
CB B XB x1 x2 x3 S1 S2 S3 XB/CK
0 S1 8 2 3 0 1 0 0 8/3 = 2.6
0 S2 10 0 2 5 0 1 0 10/2 = 5
0 S3 15 3 2 4 0 0 1 15/2 = 7.5
Zj-Cj -3 -5 -4 0 0 0

R1 = R1/3, R2= R2-2R1 , R3= R3-2R1

 LINEAR PROGRAMMING

𝑅3 15 3 2 4 0 0 1 𝑅2 10 0 2 5 0 1 0
2𝑅1 16/3 4/3 2 0 2/3 0 0 2𝑅1 16/3 4/3 2 0 2/3 0 0
𝑅3-2𝑅1 29/3 5/3 0 4 -2/3 0 1 𝑅2 -2𝑅1 14/3 -4/3 0 5 -2/3 1 0

Cj 3 5 4 0 0 0 Min Ratio
CB B XB x1 x2 x3 S1 S2 S3 XB/CK
5 x2 8/3 2/3 1 0 1/3 0 0 -
0 S2 14/3 -4/3 0 5 -2/3 1 0 14/3*1/5 = 14/15
0 S3 29/3 5/3 0 4 -2/3 0 1 29/3*1/4 = 29/12
Zj-Cj 1/3 0 -4 5/3 0 0

R2 = R2/5, R3= R3-4R2

 LINEAR PROGRAMMING
𝑅3 29/3 5/3 0 4 -2/3 0 1

4𝑅2 56/15 -16/15 0 4 -8/15 4/5 0

𝑅3-4𝑅2 89/15 41/15 0 0 -2/15 -4/5 1

Cj 3 5 4 0 0 0 Min Ratio
CB B XB x1 x2 x3 S1 S2 S3 XB/CK
5 x2 8/3 2/3 1 0 1/3 0 0 8/3*3/2=4
4 x2 14/15 -4/15 0 1 -2/15 1/5 0 -
0 S3 89/15 41/15 0 0 -2/15 -4/5 1 89/15*41/15=89/41
Zj-Cj -11/15 0 0 17/15 4/5 0

R3 = 15/41R3, R2= R2+4/15R3 , R1= R1-2/3R3

 LINEAR PROGRAMMING

Cj 3 05 4 0 0 0 Min Ratio
CB B XB x1 x2 x3 S1 S2 S3 XB/CK
5 x2 50/41 0 1 0 15/41 8/41 -10/41
4 x2 62/41 0 0 1 -6/41 5/41 4/41
4 x3 89/41 1 0 0 -2/41 -12/41 15/41
Zj-Cj 0 0 0 45/41 24/41 11/41

Since all 𝑍𝑗 − 𝐶𝑗 ≥ 0 the solution is optimum and given by

X2 = 62/41, X3 = 50/41, X1 = 89/41 and value of
Max Z = 765/41
 LINEAR PROGRAMMING
Q-3. Objective Function Min Z = 𝑥1 − 3𝑥2 + 2𝑥3
Subject to: 3𝑥1 −𝑥2 +3𝑥3 ≤ 7
−2𝑥1 + 4𝑥2 ≤ 12
−4𝑥1 + 3𝑥2 + 8𝑥3 ≤ 10
𝑥1 , 𝑥2 , 𝑥3 ≥ 0
Solution:
𝑀𝑎𝑥 𝑍 ′ = −𝑥1 + 3𝑥2 − 2𝑥3 𝑤ℎ𝑒𝑟𝑒 𝑍 ′ = 𝑍
Subject to: 3𝑥1 −𝑥2 +3𝑥3 + 𝑠1 = 7
−2𝑥1 + 4𝑥2 + 𝑠2 = 12
−4𝑥1 + 3𝑥2 + 8𝑥3 + 𝑠3 = 10
𝑥1 , 𝑥2 , 𝑥3 , 𝑠1 , 𝑠2 , 𝑠3 ≥ 0
An initial basic feasible solution is given by
𝑥1 = 𝑥2 = 𝑥3 = 0, 𝑠1 = 7, 𝑠2 = 12, 𝑠3 = 10
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING BIG M METHOD
LINEAR PROGRAMMING
LINEAR PROGRAMMING
LINEAR PROGRAMMING
LINEAR PROGRAMMING
LINEAR PROGRAMMING
 LINEAR PROGRAMMING
Max Z = 𝟐𝑿𝟏 + 𝑿𝟐
𝟒𝑿𝟏 + 𝟑𝑿𝟐 ≤ 𝟏𝟐
𝟒𝑿𝟏 + 𝑿𝟐 ≤ 𝟖
𝟒𝑿𝟏 − 𝑿𝟐 ≤ 𝟖
𝑿𝟏 , 𝑿𝟐 ≥ 𝟎

Cj 2 1 0 0 0 Min Ratio
CB B XB X1 X2 S1 S2 S3 XB/CK S1/CK S2/CK
0 S1 12 4 3 1 0 0 12/4=3 ¼=0.25 -
0 S2 8 4 1 0 1 0 8/4=2 0/4=0 ¼=0.25
0 S3 8 4 -1 0 0 1 8/4=2 0/4=0 0/4=0
Zj-Cj -2 -1 0 0 0
R3 = R3/4, R1= R1-4R3, R2= R2-4R3
 LINEAR PROGRAMMING
Cj 2 1 0 0 0 Min Ratio
CB B XB X1 X2 S1 S2 S3 XB/CK
0 S1 4 0 4 1 0 -1 4/4=1
0 S2 0 0 2 0 1 -1 0/2=0
2 X1 2 1 -1/4 0 0 1/4
Zj-Cj 0 -3/2 0 0 1/2

R2 = R2/2, R1= R1-4R2, R3= R3+1/4R2

Cj 2 1 0 0 0 Min Ratio
CB B XB X1 X2 S1 S2 S3 XB/CK
0 S1 4 0 0 1 -2 1 4/1=4
1 X2 0 0 1 0 1/2 -1/2 -
2 X1 2 1 0 0 -1/8 1/8 2/1/8=16
Zj-Cj 0 0 0 3/4 -1/4

R2= R2+1/2R1, R3= R3-1/8R1

 LINEAR PROGRAMMING

Cj 2 1 0 0 0 Min Ratio
CB B XB X1 X2 S1 S2 S3 XB/CK
0 S3 4 0 0 1 -2 1
1 X2 2 0 1 1/2 -1/2 0
2 X1 3/2 1 0 -1/8 3/8 0
Zj-Cj 0 0 1/4 1/4 0
LINEAR PROGRAMMING
LINEAR PROGRAMMING
LINEAR PROGRAMMING
LINEAR PROGRAMMING
LINEAR PROGRAMMING
LINEAR PROGRAMMING