In: Math
A primal maximization problem is given.
Maximize
f = 60x1 + 30x2
subject to
3x1 | + | 2x2 | ≤ | 150 | |
x1 | + | x2 | ≤ | 70 | . |
(a) Form the dual minimization problem. (Use
y1 and y2 as the variables
and g as the function.)
Minimize g =
subject to
≥ | 60 | ||
≥ | 30 | ||
y1, y2 | ≥ | 0 | . |
(b) Solve both the primal and dual problems with the simplex
method.
primal | x1 | = | |
primal | x2 | = | |
primal | f | = | |
dual | y1 | = | |
dual | y2 | = | |
dual | g | = |
a) the dual minimization problem is given by
subject to
b) simplex method for primal problem
primal problem is
maximize f = 60x1+30x2
subject to
3x1+2x2150
x1+x270
x1,x2
introducing slack variables x3,x4 to the system we get
subject to
table 1
60 | 30 | 0 | 0 | ||
basic variable | b | ||||
3 | 2 | 1 | 0 | 150 | |
1 | 1 | 0 | 1 | 70 | |
f | -60 (enters) | -30 | 0 | 0 | 0 |
table 2
basic variable | b | ||||
1 | 0 | 50 | |||
0 | 1 | 20 | |||
f | 0 | 10 | 20 | 0 | 3000 |
simplex table stops here as the last row values are all non negative
the solution is
primal x1=50
primal x2=0
primal f = 3000
solution to the dual problem
introducing surplus variables y3,y4 into the problem we get
minimize g = 150 y1+70y2+0y3+0y4
subject to
3y1+2y2+y3=60
y1+y2+y4=30
table 1
basic variable | RHS | ||||
3 | 2 | 1 | 0 | 60 | |
1 | 1 | 0 | 1 | 30 | |
g | -150(enters) | -70 | 0 | 0 | 0 |
table 2
basic variable | RHS | ||||
1 | 0 | 20 | |||
0 | 1 | 10 | |||
g | 0 | 30 | 50 | 0 |
3000 |
simplex stops here as the values of last row are non - negative
solution is
dual y1=20
dual y2=0
dual g = 3000