Ch4 Goal Programming
Ch4 Goal Programming
●
Same as linear programming, but we add goal.
●
Can have more than one goal.
●
Goal → not necessarily achieved.
Goal Programming Example
●
Example on problem without goal:
●
An electronics firm manufactures two types of electronic calculators, the
“student” and the “professor.”
If 60 h is available each week for each department, how many calculators should the company
produce to maximize profit?
Goal Programming Example
● Max f = 7.5x1 + 10x2
subject to
x1 + 2x2 <= 60
1.5x1 + 1.5x2 <= 60
x1, x2 >= 0
Goal Programming Example
● Max f = 7.5x1 + 10x2
subject to
x1 + 2x2 <= 60
1.5x1 + 1.5x2 <= 60
x1, x2 >= 0
●
Set goal of achieving 1000$ profit
Goal Programming Example
●
Min f = d- d- → underachievement
●
subject to d+ → overachievement
x1 + 2x2 <= 60
more than or equal to goal →
1.5x1 + 1.5x2 <= 60 minimize d-
7.5x1 + 10x2 + d- − d+ = 1000
less than or equal to goal →
x1, x2,d-,d+ >= 0 minimize d+
Goal Programming Example
●
Min f = d-
●
subject to
x1 + 2x2 + x3 = 60
1.5x1 + 1.5x2 + x4 = 60 Standard Form
7.5x1 + 10x2 + d- − d+ = 1000
x1, x2,d-,d+ >= 0
Goal Programming Example
Out (Pivot
Row)
In (Pivot Column)
Goal Programming: Multiple Goals
●
Min f = d-
●
subject to
x1 + 2x2 <= 60
1.5x1 + 1.5x2 <= 60
7.5x1 + 10x2 + d- − d+ = 1000
x1, x2,d-,d+ >= 0
●
Add two more goals → sell at least 10 of each of the calculators
Goal Programming: Multiple Goals
● Min f = d1- + d2- + d3-
●
subject to
x1 + 2x2 <= 60
1.5x1 + 1.5x2 <= 60
7.5x1 + 10x2 + d1- − d1+ = 1000
x1 + d2- - d2+ = 10
x2 + d3- - d3+ = 10
x1, x2,d1-,d1+,d2-,d2+,d3-,d3+ >= 0
Goal Programming: Priority
●
Consider this problem:
● Minimize f = d1- + d2-
Subject to
x1 – x2 + d1- - d1+ = 50
-2x1 + 3x2 + d2- - d2+ = 0
x1 + x2 <= 1000
x1,x2,d1-,d1+,d2-,d2+ >= 0
●
The two goals have no priority.
Goal Programming: Priority
● Minimize f = d1- + d2-
Subject to
x1 – x2 + d1- - d1+ = 50
-2x1 + 3x2 + d2- - d2+ = 0
x1 + x2 <= 1000
x1,x2,d1-,d1+,d2-,d2+ >= 0
●
The two goals have no priority.
● To assign priority to goals, multiply most important goal by P 1, second most
important goal by P2, and so on.
Goal Programming: Priority
● Minimize f = P1d1- + P2d2-
Subject to
x1 – x2 + d1- - d1+ = 50
-2x1 + 3x2 + d2- - d2+ = 0
x1 + x2 <= 1000
x1,x2,d1-,d1+,d2-,d2+ >= 0
Goal Programming: Priority
● Minimize f = P1d1- + P2d2-
Subject to
x1 – x2 + d1- - d1+ = 50
-2x1 + 3x2 + d2- - d2+ = 0 Standard Form
x1 + x2 + x3 = 1000
x1,x2,d1-,d1+,d2-,d2+ >= 0
Goal Programming: Priority
Out (Pivot
Row)
In (Pivot Column)
Goal Programming: Priority
In (Pivot Column)