6s-1
Linear Programming
Operations Management
William J. Stevenson
8th edition
6s-2
Linear Programming
CHAPTER
6s
Linear Programming
McGraw-Hill/Irwin
Operations Management, Eighth Edition, by William J. Stevenson Copyright 2005 by The McGraw-Hill Companies, Inc. All rights reserved.
6s-3
Linear Programming
Linear Programming
Used to obtain optimal solutions to problems that involve restrictions or limitations, such as:
Materials Budgets Labor Machine time
6s-4
Linear Programming
Linear Programming
Linear programming (LP) techniques consist of a sequence of steps that will lead to an optimal solution to problems, in cases where an optimum exists
6s-5
Linear Programming
Linear Programming Model
Objective: the goal of an LP model is maximization or
minimization
Decision variables: amounts of either inputs or
outputs
Feasible solution space: the set of all feasible
combinations of decision variables as defined by the constraints
Constraints: limitations that restrict the available
alternatives
Parameters: numerical values
6s-6
Linear Programming
Linear Programming Assumptions
Linearity: the impact of decision variables is
linear in constraints and objective function
Divisibility: noninteger values of decision
variables are acceptable
Certainty: values of parameters are known and
constant
Nonnegativity: negative values of decision
variables are unacceptable
6s-7
Linear Programming
Graphical Linear Programming
Set up objective function and constraints in mathematical format Plot the constraints
1.
2.
3.
4. 5.
Identify the feasible solution space
Plot the objective function
Determine the optimum solution
6s-8
Linear Programming
Linear Programming Example
Objective - profit
Maximize Z=60X1 + 50X2
Subject to
Assembly 4X1 + 10X2 <= 100 hours Inspection 2X1 + 1X2 <= 22 hours Storage 3X1 + 3X2 <= 39 cubic feet X1, X2 >= 0
6s-9
Linear Programming
Linear Programming Example
Assembly Constraint 4X1 +10X2 = 100
12 10 8 6 4 2 0
Product X2
10
12
14
16
18
20
22
Product X1
24
6s-10 Linear Programming
Linear Programming Example
Add Inspection Constraint 2X1 + 1X2 = 22
25
Product X2
20 15 10 5 0
10
12
14
16
18
20
22
Product X1
24
6s-11 Linear Programming
Linear Programming Example
Add Storage Constraint 3X1 + 3X2 = 39
25
Inspection Storage Assembly
Product X2
20 15 10 5 0
10
12
14
16
18
20
22
Feasible solution space
Product X1
24
6s-12 Linear Programming
Linear Programming Example
Add Profit Lines
25
Product X2
20 15 10 5 0
Z=900
10
12
14
16
18
20
22
Z=300
Z=600
Product X1
24
6s-13 Linear Programming
Solution
The intersection of inspection and storage Solve two equations in two unknowns
2X1 + 1X2 = 22 3X1 + 3X2 = 39 X1 = 9 X2 = 4 Z = $740
6s-14 Linear Programming
Constraints
Redundant constraint: a constraint that does
not form a unique boundary of the feasible solution space
Binding constraint: a constraint that forms the
optimal corner point of the feasible solution space
6s-15 Linear Programming
Slack and Surplus
Surplus: when the optimal values of decision
variables are substituted into a greater than or equal to constraint and the resulting value exceeds the right side value
Slack: when the optimal values of decision
variables are substituted into a less than or equal to constraint and the resulting value is less than the right side value
6s-16 Linear Programming
Simplex Method
Simplex:
a linear-programming algorithm that can solve problems having more than two decision variables
6s-17 Linear Programming
MS Excel Worksheet for Microcomputer Problem
Figure 6S.15
6s-18 Linear Programming
MS Excel Worksheet Solution
Figure 6S.17
6s-19 Linear Programming
Sensitivity Analysis
Range of optimality: the range of values for
which the solution quantities of the decision variables remains the same
Range of feasibility: the range of values for
the fight-hand side of a constraint over which the shadow price remains the same
Shadow prices: negative values indicating
how much a one-unit decrease in the original amount of a constraint would decrease the final value of the objective function
6s-20 Linear Programming
CHAPTER
Additional PowerPoint slides contributed by Geoff Willis, University of Central Oklahoma.
6s-21 Linear Programming
Linear Programming in Excel - Part 1
One of Excels useful features is the ability to solve linear programming problems (especially those beyond our graphical abilities). The feature is invoked by creating a spreadsheet containing the objective function and constraints, selecting Tools from the main menu, and Solver from the submenu
6s-22 Linear Programming
Linear Programming in Excel - Part 2
Corn (bushels) Sugar (pounds) Jugs Hours White Lightning 2 3 1 3 12 10 Rotgut 1 2 1 2 8 20 =C60*$C$52+D60*$D$52 =C61*$C$52+D61*$D$52 =C62*$C$52+D62*$D$52 =C63*$C$52+D63*$D$52 =C65*C64+D65*D64 <= <= <= <= Profit 40 70 50 72
Here is an Excel version of the moonshine problem
Corn (bushels) Sugar (pounds) Jugs Hours White Lightning Rotgut 2 1 3 2 1 1 3 2 $12.00 $8.00 10 20 40 70 30 70 <= <= <= <= 40 70 50 72
$280.00 Profit
The top view shows formulas and the bottom view shows initial calculations.
6s-23 Linear Programming
Linear Programming in Excel - Part 3
Once the basic set of equations has been entered, launch Solver and fill in the dialog boxes with references to your sheet. Target cell - The objective function value (E52) Equal to - Choose max or min based on the problem By changing cells - The decision variables (C52:D52) Subject to the constraints - Add all constraints one at a time by referencing their function values (e.g., the amount of corn used, E47 must be less than the amount of corn on hand, G47) Once all constraints have been entered, choose Options and check the boxes for Assume Linear Model and Assume Non-Negative. Finally, choose Solve and wait for Excel to work its magic
6s-24 Linear Programming
Solver Output Reports Part 1
Answer Report - contains the basic answer to the problem and reveals which constraints had an impact on your situation. Sensitivity Report - tells you reduced costs and shadow prices Limits Report - dont bother asking for this one. We wont use its information.
6s-25 Linear Programming
Solver Output Reports Part 2
Microsoft Excel 8.0e Answer Report The value of the Worksheet: [P/OM.xls]Sheet1 objective function at the Report Created: 7/28/2000 11:03:04 AM optimal solution Target Cell (Max) Cell Name Original Value Final Value The optimal values $E$52 Profit $20.00 $280.00 of the decision Adjustable Cells variables Cell Name Original Value Final Value $C$52 White Lightning 1 10 $D$52 Rotgut 1 20 Constraints Cell Name Cell Value Formula Status Slack $E$47 Corn 40 $E$47<=$F$47 Binding 0 $E$48 Sugar 70 $E$48<=$F$48 Binding 0 $E$49 Jugs 30 $E$49<=$F$49 Not Binding 20 $E$50 Hours 70 $E$50<=$F$50 Not Binding 2 A binding constraint is one that limits the value our objective function can assume. We use up all of our corn and sugar (we have no slack).
If a constraint is not binding, then we have some left over (slack) when we implement the optimal solution. We have 20 extra jugs and 2 hours to spare.
6s-26 Linear Programming
Solver Output Reports Part 3
Microsoft Excel 8.0e Sensitivity Report Worksheet: [P/OM.xls]Sheet1 Increases or decreases within these ranges will result in the same Report Created: 7/28/2000 11:03:05 AM product mix (but a different objective function value). Adjustable Cells Final Reduced Objective Allowable Allowable Cell Name Value Cost Coefficient Increase Decrease $C$52 White 10 0 12 4 0 $D$52 Rotgut 20 0 8 0 2 Constraints
Cell $E$47 $E$48 $E$49 $E$50
Name Corn Sugar Jugs Hours
Final Value 40 70 30 70
Shadow Price 4 -
Constraint R.H. Side 40 70 50 72
Allowable Increase 6.67 2 1E+30 1E+30
Allowable Decrease 5 10 20 2
A one unit decrease in the original amount of sugar available will decrease our profit by this amount Extra amount of resource needed for a binding constraint to become non-binding. Note that this doesnt apply to nonbinding constraints, hence the huge amounts indicated.
Amount of resource to be taken away for a non-binding constraint to become binding, or a binding constraint to become more so.
