Industrial & Systems Engineering

Computer Applications in Industry

LINEAR PROGRAMMING – LP

Assoc. Prof. Ho Thanh Phong

Operations Research 1
INTRODUCTION

• Many decisions in management are related with the best

usage resources of organizations.
• Manager makes Decisions in order to satisfy Objectives,
Goals of organizations.
• Resources: Materials, Machines, Man, Money, Time,
Space.
• Linear Programming (LP) is a mathematical method that
helps managers to make decision related with Resources
Allocation. (references about Nobel laureate: Kantorovich)
• Extensively using computer.

Operations Research 2
 LP PROBLEM

• Problem: Maximize or Minimize some variables, usually Profit/

Cost, called Objective function.
• Constraints: are functions show resources limitation of
companies/ organizations. The problem is to find s solution that
maximize profits (or minimize lost/cost) in given constraints.
Form of constraint funcstionc could be:
• Inequality (form  or )
• Equality

• All Objective function and Constraint functions are linear

functions.

Operations Research 3
Formulating LP Problems - the most important task

Example 1. ABC Furniture Company

• Products: tables and chairs.
• The production process for each is similar (carpentry works,
painting and varnishing)
• ABC Furniture's problem: to determine the best possible
combination of tables chairs to manufacture in order to reach the
maximum profit.
• The data provided in the following table
 ABC Furniture Company Data

Hours required to produce 1 Available

Department unit Hours
(X1) Tables (X2) Chairs

Carpentry 4 3 240
Painting and varnishing 2 1 100
Profit per unit $7 $5
Formulation
• X1 = number of tables to be produced
• X2 = number of chairs to be produced

The problem is formulating:

Maximize profit Z = 7X1 + 5X2
Subject to:
4X1 + 3X2  240
2X1 + 1X2  100
X1 0
X2 0
Graphical Solution
• The graphical method works only when there are two decision
variables, but it provides valuable insight into how larger problems
are structured
• Graphical Representation of Constraints

• Isoprofit-line method
• Corner points method
Graphical Representation of Constraints

100

80 A (X1 = 0, X 2 = 80)
Number of Chairs - X 2

60

4X1+3X2 = 240
40
4X1+3X2 <= 240

20

B (X1 = 60, X 2 = 0)

0 20 40 60 80 100 120
Number of Tables - X 1
 100 C (X1 = 0, X 2 = 100)

80 A (X1 = 0, X 2 = 80)
Number of Chairs - X 2

2X1+1X2 = 100
60

4X1+3X2 = 240
40

D (X1 = 50, X 2 = 0)
20

Feasible B (X1 = 60, X 2 = 0)

Region

0 20 40 60 80 100 120
Number of Tables - X 1
Solving LP graphically by Isoprofit- line method

100

80 A (X 1 = 0, X 2 = 80)
Number of Chairs - X 2

60 I (X 1 = 30, X 2 = 40)

40 Max Profit = 7(30)+5(40) = 410

Feasible
Region 20
D (X1 = 50, X 2 = 0)

0 20 40 60 80 100 120
Number of Tables - X 1
Solving LP problem by Corner-point method
• Objective: Maximize Profit Z = 7X1 + 5X2
• The mathematical theory in LP shows that the optimal
solution must lie at one corner point, or extreme point,
of the feasible region

• At the point (0,0): Profit = 0

• At the point D(50,0): Profit = 7(50) + 5(0) = 350
• At the point A(0,80): Profit = 7(0) + 5(80) = 400
• At the point I(30,40):Profit = 7(30) + 5(40) = 410 *
• The optimal solution is I(30,40) which obtains profit of
410
04 SPECIAL CASES

There are 04 special cases in LP:

• Infeasible solutions
• Unbounded solutions
• Redundant constraints
• Multiple optimal solutions

12
04 SPECIAL CASES

• Case1: Infeasible
• Infeasible solutions occurred when we have conflicting constraints ;
or
• No solution satisfy all constraints; or
• Can not build the feasible solutions region.

Operations Research 13
04 SPECIAL CASES

• Exampleï:
Consider three constraints:
X2 Regions which
Mieàn thoaû maõn raøng
100 satisfy constraints 3
buoäc thöù ba
4X1 + 3X2  240
2X1 + 1X2  100 80 A (X1 = 0, X 2 = 80)

X1  80
60

40
Regions which
satisfy constraints 1 20

and 2
0 20 40 60 80 100 120 X

Operations Research 14
04 SPECIAL CASES

Case 2: Unbounded solutions

• When value of Objective function approached to infinity we said that
the problem is unbounded or missing one or more constraints; or
• The LP did not provide a finite solutions, this implies the objective
function approaches to infinity without violating any constraint. û.
• → Open ended problem

Operations Research 15
04 SPECIAL CASES

• Example:
Maximize Z = 3X1 + 5X2
St: X2
X1  5 X1 = 5

X2  10
X1 + 2 X2  10 15
X 2 = 10
X1, X2  0
10

Mieàn lôøi giaûi

5
X1 + 2X2 = 10

0 5 10 15 X1

Operations Research 16
04 SPECIAL CASES

Case 3: Redundancy of Constraints

• A redundant constraint is a constraint that will not affect to the
solution space
• In reality, this will usually happens when number of constraints and
umber of variables are very large.

Operations Research 17
04 SPECIAL CASES

Example:
X2
Maximize Z = 7X1 + 5X2
100
St:
4X1 + 3X2  240 80 A (X1 = 0, X2 = 80)

2X1 + 1X2  100

Raøng buoäc dö: X1<= 80
X1  80 60

40

20

0 20 40 60 80 100 X1

Operations Research 18
04 SPECIAL CASES

Case : Multiple solutions

• When objective function and one constraint have the same slope we
will faced with the case multiple optimal solutions

Operations Research 19
04 SPECIAL CASES X2

100

Maximize Z = 8X1 + 6X2 80 A (X1 = 0, X 2 = 80)

St:
4X1 + 3X2  240 60
2X1 + 1X2  100
I (X 1 = 30, X 2 = 40)
40

20

0 20 40 60 80

Operations Research 20
COMPUTER APPLICATION

• Excel
• Lingo

Operations Research 21