LPP
LPP
To accompany Quantitative Analysis for Management, Tenth Edition, by Render, Stair, and Hanna Power Point slides created by Jeff Heyl
Introduction
Many management decisions involve trying to
materials
mathematical modeling technique designed to help managers in planning and decision making relative to resource allocation Belongs to the broader field of mathematical programming In this sense, programming refers to modeling and solving a problem mathematically
2009 Prentice-Hall, Inc. 72
minimize total shipping cost from several warehouses to various market locations
2009 Prentice-Hall, Inc. 74
Basic Assumptions of LP
We assume conditions of certainty exist and numbers
in the objective and constraints are known with certainty and do not change during the period being studied We assume proportionality exists in the objective and constraints
constancy between production increases and resource utilization if 1 unit needs 3 hours then 10 require 30 hours
equals the sum of the individual activities We assume divisibility in that solutions need not be whole numbers All answers or variables are nonnegative as we are dealing with real physical quantities
2009 Prentice-Hall, Inc. 76
Formulating LP Problems
Formulating a linear program involves
developing a mathematical model to represent the managerial problem The steps in formulating a linear program are 1. Completely understand the managerial problem being faced 2. Identify the objective and constraints 3. Define the decision variables 4. Use the decision variables to write mathematical expressions for the objective function and the constraints
2009 Prentice-Hall, Inc. 77
Formulating LP Problems
One of the most common LP applications is the
product mix problem Two or more products are produced using limited resources such as personnel, machines, and raw materials The profit that the firm seeks to maximize is based on the profit contribution per unit of each product The company would like to determine how many units of each product it should produce so as to maximize overall profit given its limited resources
2009 Prentice-Hall, Inc. 78
inexpensive tables and chairs Processes are similar in that both require a certain amount of hours of carpentry work and in the painting and varnishing department Each table takes 4 hours of carpentry and 2 hours of painting and varnishing Each chair requires 3 of carpentry and 1 hour of painting and varnishing There are 240 hours of carpentry time available and 100 hours of painting and varnishing Each table yields a profit of $70 and each chair a profit of $50
2009 Prentice-Hall, Inc. 79
DEPARTMENT
Carpentry Painting and varnishing Profit per unit Table 7.2
7 10
Maximize profit The constraints are 1. The hours of carpentry time used cannot exceed 240 hours per week 2. The hours of painting and varnishing time used cannot exceed 100 hours per week The decision variables representing the actual decisions we will make are T = number of tables to be produced per week C = number of chairs to be produced per week
7 11
and C Maximize profit = $70T + $50C Develop mathematical relationships for the two constraints For carpentry, total time used is
(4 hours per table)(Number of tables produced) + (3 hours per chair)(Number of chairs produced)
We know that
Carpentry time used Carpentry time available 4T + 3C 240 (hours of carpentry time)
7 12
Painting and varnishing time used Painting and varnishing time available 2 T + 1C 100 (hours of painting and varnishing time) This means that each table produced requires two hours of painting and varnishing time
Both of these constraints restrict production
7 13
subject to 4T + 3C 240 (carpentry constraint) 2T + 1C 100 (painting and varnishing constraint) T, C 0 (nonnegativity constraint)
2009 Prentice-Hall, Inc. 7 14
problems is with the graphical solution approach The graphical method only works when there are just two decision variables When there are more than two variables, a more complex approach is needed as it is not possible to plot the solution on a twodimensional graph The graphical method provides valuable insight into how other approaches work
2009 Prentice-Hall, Inc. 7 15
40
20 0
| | | | | | |
20
40
60
80
100
Figure 7.1
Number of Tables
2009 Prentice-Hall, Inc. 7 16
identify a set or region of feasible solutions To do this we plot each constraint equation on a graph We start by graphing the equality portion of the constraint equations 4T + 3C = 240 We solve for the axis intercepts and draw the line
2009 Prentice-Hall, Inc. 7 17
carpentry constraint is 4(0) + 3C = 240 3C = 240 C = 80 Similarly for no chairs 4T + 3(0) = 240 4T = 240 T = 60 This line is shown on the following graph
2009 Prentice-Hall, Inc. 7 18
(T = 0, C = 80)
40
20 0
| | | | | | | | | | | |
(T = 60, C = 0)
20
40
60
80
100
Figure 7.2
Number of Tables
2009 Prentice-Hall, Inc. 7 19
plot will not violate the restriction Any point above the plot will violate the restriction
40
20 0
| |
(30, 40)
(70, 40)
(30, 20)
| | | | | | | | | |
20
40
60
80
100
Figure 7.3
Number of Tables
2009 Prentice-Hall, Inc. 7 20
40
20 0
| | | | | | | | | | | |
(T = 50, C = 0)
20
40
60
80
100
Figure 7.4
Number of Tables
2009 Prentice-Hall, Inc. 7 22
departments must be used We need to find a solution that satisfies both constraints simultaneously A new graph shows both constraint plots The feasible region (or area of feasible solutions) is where all constraints are satisfied Any point inside this region is a feasible solution Any point outside the region is an infeasible solution
7 23
Painting/Varnishing Constraint
40
20 Feasible Region 0
| | | | | | | | | | | |
Carpentry Constraint
20
40
60
80
100
Figure 7.5
Number of Tables
2009 Prentice-Hall, Inc. 7 24
2009 Prentice-Hall, Inc. 7 25
Painting constraint
7 26
to find the optimal solution from the many possible solutions The speediest way to do this is to use the isoprofit line method Starting with a small but possible profit value, we graph the objective function We move the objective function line in the direction of increasing profit while maintaining the slope The last point it touches in the feasible region is the optimal solution
7 27
$2,100 = 70T + 50C Solving for the axis intercepts, we can draw the graph This is obviously not the best possible solution Further graphs can be created using larger profits The further we move from the origin while maintaining the slope and staying within the boundaries of the feasible region, the larger the profit will be The highest profit ($4,100) will be generated when the isoprofit line passes through the point (30, 40)
2009 Prentice-Hall, Inc. 7 28
40
(0, 42)
(30, 0)
20
40
60
80
100
Number of Tables
2009 Prentice-Hall, Inc. 7 29
40
20 0
| | | | | | | | | | | |
20
40
60
80
100
Figure 7.7
Number of Tables
2009 Prentice-Hall, Inc. 7 30
40
20 0
| | | | | | | | | | | |
20
40
60
80
100
Figure 7.8
Number of Tables
2009 Prentice-Hall, Inc. 7 31
employs the corner point method It involves looking at the profit at every corner point of the feasible region The mathematical theory behind LP is that the optimal solution must lie at one of the corner points, or extreme point, in the feasible region For Flair Furniture, the feasible region is a four-sided polygon with four corner points labeled 1, 2, 3, and 4 on the graph
2009 Prentice-Hall, Inc. 7 32
2
Number of Chairs 80 60
40
20
1 |
0 Figure 7.9
20
40
60
80
100
Number of Tables
2009 Prentice-Hall, Inc. 7 33
Point 4 : (T = 50, C = 0)
Point 3 : (T = 30, C = 40)
is the optimal solution To find the coordinates for Point 3 accurately we have to solve for the intersection of the two constraint lines The details of this are on the following slide
7 34
7 35
3. Compute the profit (or cost) at each of the feasible corner points. 4. select the corner point with the best value of the objective function found in Step 3. This is the optimal solution.
Table 7.3
2009 Prentice-Hall, Inc. 7 36
software to solve big LP problems While there are differences between software implementations, the approach each takes towards handling LP is basically the same Once you are experienced in dealing with computerized LP algorithms, you can easily adjust to minor changes
7 37
First select the Linear Programming module Specify the number of constraints (non-negativity is assumed) Specify the number of decision variables Specify whether the objective is to be maximized or minimized For the Flair Furniture problem there are two constraints, two decision variables, and the objective is to maximize profit
7 38
Program 7.1A
7 39
Program 7.1B
7 40
Program 7.1C
7 41
Program 7.1D
2009 Prentice-Hall, Inc. 7 42
objective such as cost instead of maximizing a profit function Minimization problems can be solved graphically by first setting up the feasible solution region and then using either the corner point method or an isocost line approach (which is analogous to the isoprofit approach in maximization problems) to find the values of the decision variables (e.g., X1 and X2) that yield the minimum cost
2009 Prentice-Hall, Inc. 7 43
buying two different brands of turkey feed and blending them to provide a good, low-cost diet for its turkeys Let
X1 = number of pounds of brand 1 feed purchased X2 = number of pounds of brand 2 feed purchased
Minimize cost (in cents) = 2X1 + 3X2 subject to: 5X1 + 10X2 90 ounces (ingredient constraint A) 4X1 + 3X2 48 ounces (ingredient constraint B) 0.5X1 1.5 ounces (ingredient constraint C) X1 0 (nonnegativity constraint) X2 0 (nonnegativity constraint)
2009 Prentice-Hall, Inc. 7 44
A B C
Cost per pound Table 7.4
5 4 0.5 2 cents
10 3 0 3 cents
90 48 1.5
7 45
point method First we construct the feasible solution region The optimal solution will lie at on of the corners as it would in a maximization problem
20 Pounds of Brand 2
Ingredient C Constraint
15
a
Feasible Region
10
Ingredient B Constraint 5
b 0 |
| |
Ingredient A Constraint
|
Figure 7.10
10 15 20 Pounds of Brand 1
25 X1
7 46
C and B 4X1 + 3X2 = 48 X1 = 3 Substituting 3 in the first equation, we find X2 = 12 Solving for point b with basic algebra we find X1 = 8.4 and X2 = 4.8 Solving for point c we find X1 = 18 and X2 = 0
7 47
function we find
Cost = 2X1 + 3X2 Cost at point a = 2(3) + 3(12) = 42 Cost at point b = 2(8.4) + 3(4.8) = 31.2 Cost at point c = 2(18) + 3(0) = 36
The lowest cost solution is to purchase 8.4
pounds of brand 1 feed and 4.8 pounds of brand 2 feed for a total cost of 31.2 cents per turkey
7 48
Pounds of Brand 2
Feasible Region
20
15
10
5
(X1 = 8.4, X2 = 4.8)
| |
0 | Figure 7.11
10 15 20 Pounds of Brand 1
25 X1
7 49
Program 7.3
7 50
7 51
young attorneys Ivan wants to make lawyer-to-client assignments in the most effective manner He identifies four lawyers who could possibly be assigned new cases Each lawyer can handle one new client The lawyers have different skills and special interests The following table summarizes the lawyers estimated effectiveness on new cases
7 52
Let where
1 if attorney i is assigned to case j Xij = 0 otherwise i = 1, 2, 3, 4 stands for Adams, Brooks, Carter, and Darwin respectively j = 1, 2, 3, 4 stands for divorce, merger, embezzlement, and exhibitionism 2009 Prentice-Hall, Inc.
7 53
6X11 Maximize effectiveness = + 2X12 + 8X13 + 5X14 + 9X21 + 3X22 + 5X23 + 8X24 + 4X31 + 8X32 + 3X33 + 4X + 6X41 + 7X42 + 6X43 + 4X44
subject to X11 + X21 + X31 + X41 = 1 X12 + X22 + X32 + X42 = 1 X13 + X23 + X33 + X43 = 1 X14 + X24 + X34 + X44 = 1 X11 + X12 + X13 + X14 = 1 X21 + X22 + X23 + X24 = 1 X31 + X32 + X33 + X34 = 1 X41 + X42 + X43 + X44 = 1
(divorce case) (merger) (embezzlement) (exhibitionism) (Adams) (Brook) (Carter) (Darwin) 2009 Prentice-Hall, Inc. 7 54
Program 8.4
2009 Prentice-Hall, Inc. 7 55
7 56
7 57
X1+2X2 <=6
2X1+X2 <=8 X1 >=7
solution
X2
8 6 4 2 0 Figure 7.12
X1
7 59
15
X1 5
10
X2 10
0|
5
Figure 7.13
10
15
X1
7 60
7 61
a redundant constraint
20
15 10 5 0 Figure 7.14 Feasible Region
| | | | | |
Redundant Constraint
X1 25 X1 + X2 20
10
15
20
25
30
X1
7 62
7 63
Maximize 3X1 + 2X2 Subj. To: 6X1 + 4X2 < 24 X1 <3 X1 , X2 > 0
5 4 3 2
B 1 Feasible Region | | 0 1 2
| |
Figure 7.15
8 X1
7 64
Sensitivity Analysis
Optimal solutions to LP problems thus far have
been found under what are called deterministic assumptions This means that we assume complete certainty in the data and relationships of a problem But in the real world, conditions are dynamic and changing We can analyze how sensitive a deterministic solution is to changes in the assumptions of the model This is called sensitivity analysis, postoptimality analysis, parametric programming, or optimality analysis
2009 Prentice-Hall, Inc. 7 65
Sensitivity Analysis
Sensitivity analysis often involves a series of
what-if? questions concerning constraints, variable coefficients, and the objective function
What if the profit for product 1 increases by 10%? What if less advertising money is available?
where values are changed and the entire model is resolved The preferred way is to use an analytic postoptimality analysis
After a problem has been solved, we determine a range of changes in problem parameters that will not affect the optimal solution or change the variables in the solution without re-solving the entire problem
2009 Prentice-Hall, Inc. 7 66
Sensitivity Analysis
Sensitivity analysis can be used to deal not
only with errors in estimating input parameters to the LP model but also with managements experiments with possible future changes in the firm that may affect profits.
7 67
quality CD players and stereo receivers Products require a certain amount of skilled artisanship which is in limited supply The firm has formulated the following product mix LP model
Maximize profit = $50X1 + $120X2 Subject to 2X1 + 4X2 80 3X1 + 1X2 60 X1, X2 0
2009 Prentice-Hall, Inc. 7 68
0
Figure 7.16
10
20
30 40 c = (20, 0)
50
60 X1 (CD players)
2009 Prentice-Hall, Inc. 7 69
objective functions fluctuate periodically Graphically, this means that although the feasible solution region remains exactly the same, the slope of the isoprofit or isocost line will change We can often make modest increases or decreases in the objective function coefficient of any variable without changing the current optimal corner point We need to know how much an objective function coefficient can change before the optimal solution would be at a different corner point
2009 Prentice-Hall, Inc. 7 70
20 a 10
0 Figure 7.17
| c
10
20
30
40
50
60
X1
2009 Prentice-Hall, Inc. 7 71
data
Program 7.5A
Program 7.5B
2009 Prentice-Hall, Inc. 7 72
reflect changes in the state of technology If the amount of resources needed to produce a product changes, coefficients in the constraint equations will change This does not change the objective function, but it can produce a significant change in the shape of the feasible region This may cause a change in the optimal solution
7 73
X2 60
X2 60
40 20 a
40 20 a
40
2X1 + 4X2 80
| |
2X1 + 4X2 80 e | | 20 30 40
|
20 16 f 0
2X1 + 5 X2 80
| |
40
X1
X1
40
X1
CD Players
Figure 7.18
2009 Prentice-Hall, Inc. 7 74
often represent resources available to the firm If additional resources were available, a higher total profit could be realized Sensitivity analysis about resources will help answer questions such as:
How much should the company be willing to pay
for additional hours? Is it profitable to have some electricians work overtime? Should we be willing to pay for more audio technician time?
2009 Prentice-Hall, Inc. 7 75
changed, the feasible region will change (unless the constraint is redundant) and often the optimal solution will change The amount of change (increase or decrease) in the objective function value that results from a unit change in one of the resources available is called the dual price or dual value
2009 Prentice-Hall, Inc. 7 76
right-hand side of a resource is limited If the number of hours increases beyond the upper bound (or decreases below the lower bound), then the objective function would no longer increase (decrease) by the dual price.
or the objective function may change by an amount different from the dual price. Thus, the dual price is relevant only within limits.
If the dual value of a constraint is zero
The slack is positive, indicating unused resource Additional amount of resource will simply increase
If the electricians hours are changed from 80 to 100, the new optimal solution is (0,25) with profit of $3,000. The extra 20 hours resulted in an increase in profit of $600 or $30/hour
Constraint Representing 60 Hours of Audio Technicians Time Resource a b
(a)
40
25 20
20
40
50
60
X1
Figure 7.19
2009 Prentice-Hall, Inc. 7 78
If the hours are decreased to 60, the new solution is (0,15) and the profit is $1,800. Reducing hours by 20 results in a $600 decrease in profit or $30/hour
40
20 15 a b 0 c
| |
20
30
40
60
X1
Figure 7.19
2009 Prentice-Hall, Inc. 7 79
(c)
20
20
40
60
80
100
120
X1 Figure 7.19
2009 Prentice-Hall, Inc. 7 80
the optimal solution would be (0,60) with a profit of $7,200. This is $2,400 (the original solution) + $30 (dual price)*160 hours(240-80)
If the hours increases beyond 240, then the
optimal solution would still be (0,60) and profit would not increase. The extra time is slack during which the electricians are not working
7 81
data
Program 7.5A
Program 7.5B
2009 Prentice-Hall, Inc. 7 82
Dual/Value Carpentry/1.5
Painting/.5
100101
Solution 7 5
7 83
advertising? For the optimal solution, how much square footage will be used? Would the solution change if the advertising budget were only $300 instead of $400? Why? What would the optimal solution be if the profit on the large spaces were reduced from $50 to $45? How much would earnings increase if the square footage requirement were increased from 8,000 to 9,000?
1. 2*60 + 4*40 = 280 (note slack of 120 on advertising constraint) 2. 100*60 + 50*40 = 8,000 (no slack) 3. Advertising has a lower bound of 280 and upper bound of Infinity. 300 is within the
bounds so the solution does not change. 4. The profit on large spaces has a range of 40 to Infinity, 45 is within that range so the solution would not change. However, earnings would be reduced from $3,800 to 45*60 + 20*40 = $3,500. 5. The dual price for square footage is .4 and the upper bound is 9,500. By increasing the square footage by 1,000 we increase the earning by .4*1,000 = $400
7 84