Linear Programming

© 2021 McGraw-Hill Education Limited.

 Learning Objectives
 Formulate a linear programming model
from the description of a problem.
 Solve two-variable linear programming
problems using the graphical solution
method.
 Solve linear programming problems
using Excel’s Solver.

© 2021 McGraw-Hill Education Limited.

Examples of Linear Programming
Applications
Establishing
locations for
emergency
equipment and
personnel Optimal
Optimal mix of schedules for
products in a airlines, pilots,
factory ground
personnel

Shipping plans
that will
Financial plans
minimize
shipping costs

Optimal blends of
Optimal diet
animal feed
plans
mixes
© 2021 McGraw-Hill Education Limited.
Linear Programming Model
A mathematical representation of a constrained
optimization problem
Objective
Function: Constraints:
Mathematical Limitations
Expression

Decision Parameters:
Variables: Choices Coefficients
Components

© 2021 McGraw-Hill Education Limited.

 Linear Programming Assumptions

Linearity Divisibility

Non-
Certainty
negativity

© 2021 McGraw-Hill Education Limited.

 Linear Programming Assumptions
 Linearity:
• The impact of decision variables is linear in constraints
and objective function
 Divisibility:
• Non integer values of decision variables are acceptable
 Certainty:
• Values of parameters are known and constant
 Non negativity:
• Negative values of decision variables are unacceptable

© 2021 McGraw-Hill Education Limited.

 Graphical Solution Method
Involves drawing the feasible solution space and contours
of the objective function and finding the optimal solution to
two variable problems by observation
1. Set up objective function and constraints in mathematical
format
2. Plot the constraints
3. Identify the feasible solution space
4. Plot the objective function, determine the optimal point
5. Redundant constraints
6. Minimization
7. Slack and Surplus

© 2021 McGraw-Hill Education Limited.

 1. Set up the Model
x1  quantity of type 1 computer to produce per day
x2  quantity of type 2 computer to produce per day
Maximize Z 60 x1  50 x2
subject to
Assembly 4 x1  10 x2  100 hours
Inspection 2 x1  1x2  22 hours
Storage 3 x1  3 x2  39 cubic feet
x1 , x2  0 (Non negativity constraints)

© 2021 McGraw-Hill Education Limited.

 2. Plotting Constraints

2. Determine 3. Mark axes

1. Replace where line intersections,
inequality sign intersects each connect with a
with = axis straight line

4. Indicate by shading whether inequality

Repeat steps is > or <
1-4 for each (Rule: pick point not on the line, such as
constraint 0,0 and see if it will satisfy the inequality)

© 2021 McGraw-Hill Education Limited.

 2. Plotting Constraints

Figure 6S-1
© 2021 McGraw-Hill Education Limited.
 2. Plotting Constraints

Figure 6S-2
© 2021 McGraw-Hill Education Limited.
3. Feasible Solution Space

Figure 6S-4

Figure 6S-3

© 2021 McGraw-Hill Education Limited.

 3. Feasible Solution Space

Figure 6S-6

Figure 6S-5
© 2021 McGraw-Hill Education Limited.
 4. Plot an Objective Function Line

Figure 6S-7

© 2021 McGraw-Hill Education Limited.

 4. Plot an Objective Function Line

Figure 6S-9

Figure 6S-8

© 2021 McGraw-Hill Education Limited.

4. Optimal Solution at Corner Point

Figure 6S-10
© 2021 McGraw-Hill Education Limited.
4. Optimal Solution at Corner Point

Figure 6S-10
© 2021 McGraw-Hill Education Limited.
 5. Redundant Constraints
 Redundant constraint: does not form a unique boundary
of feasible solution space
 Binding constraint: forms optimal corner point of
feasible solution space

Figure 6S-11
© 2021 McGraw-Hill Education Limited.
 Minimization

Minimize Z 8 x1  12 x 2
subject to
5 x1  2 x 2  20
4 x1  3 x 2  24
x2 2
x1 , x 2  0

Figure 6S-13
© 2021 McGraw-Hill Education Limited.
 Slack and Surplus
 Surplus: when optimal values of decision
variables are substituted into a  constraint
and resulting value > right side value.
x1  10, x 2  20  Slack: when optimal
3 x1  2 x 2  100 values of decision
310   220   70  SLACK  30 variables are
substituted into a 
x1  10, x 2  15 constraint and resulting
4 x1  x 2  50 value is
410   15  55  SURPLUS  5
< right side value.

© 2021 McGraw-Hill Education Limited.

 Simplex Method
A linear-programming algorithm that can solve
problems having more than two decision
variables.
• Determine the coordinates of each corner
point of the feasible solution space
• Compute the value of the objective function
at those corner points
• Identify the best one

© 2021 McGraw-Hill Education Limited.

Linear Programming (LP) Software

 For problems with > 2 decision variables.

 Solver is an Add-on to Excel.

© 2021 McGraw-Hill Education Limited.

 Computer Solutions
computer problem, using Excel’s Solver.

x1  quantity of type 1 computer t o produce per day

x 2  quantity of type 2 computer t o produce per day
Maximize Z 60 x1  50 x 2
subject to
Assembly 4 x1  10 x 2  100 hours
Inspection 2 x1  1x 2  22 hours
Storage 3 x1  3 x 2  39 cubic feet
x1 , x 2  0 (Non negativity constraint s)
© 2021 McGraw-Hill Education Limited.
 Excel Worksheet Example

Figure 6S-15
© 2021 McGraw-Hill Education Limited.
 Excel Worksheet Example
Excel worksheet for the computer
problem with formula showing.

Figure 6S-16
© 2021 McGraw-Hill Education Limited.
 Solver Parameters

Figure 6S-17
© 2021 McGraw-Hill Education Limited.
 Excel Worksheet Solution

Figure 6S-18
© 2021 McGraw-Hill Education Limited.
 Solver Sensitivity Report

Figure 6S-20

© 2021 McGraw-Hill Education Limited.