100% found this document useful (1 vote)
797 views13 pages

Parametric Linear Programming-1

Parametric linear programming involves solving a linear program where parameters are systematically varied. The objective function or constraint coefficients are modified by adding terms that are multiplied by a parameter θ. The optimal solution is tracked as θ changes. An example problem tracks the optimal solution of a glass production problem as coefficients in the objective function and right hand side change with θ. The procedure involves solving the initial problem at θ=0 and then incrementing θ, resolving the problem when θ causes a coefficient to change sign using sensitivity analysis.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
100% found this document useful (1 vote)
797 views13 pages

Parametric Linear Programming-1

Parametric linear programming involves solving a linear program where parameters are systematically varied. The objective function or constraint coefficients are modified by adding terms that are multiplied by a parameter θ. The optimal solution is tracked as θ changes. An example problem tracks the optimal solution of a glass production problem as coefficients in the objective function and right hand side change with θ. The procedure involves solving the initial problem at θ=0 and then incrementing θ, resolving the problem when θ causes a coefficient to change sign using sensitivity analysis.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 13

Parametric Linear Programming

Parametric Linear Programming-1


Systematic Changes in cj
n
• Objective function Z   c j x j is replaced by
j 1

Z      c j   j x j
n

j 1

• Find the optimal solution as a function of θ


Z*(θ)

0 θ1 θ2 θ
Parametric Linear Programming-2
Example: Wyndor Glass Problem

• Z(θ) = (3 + 2θ) x1+(5 - θ) x2

Parametric Linear Programming-3


Example: Wyndor Glass Problem

Range Basic Z x1 x2 x3 x4 x5 RHS


of θ Var.

Z(θ) 1 0 0 0 (9-7θ)/6 (3+2θ)/3 36-2θ

x3 0 0 0 1 1/3 -1/3 2
0 ≤ θ ≤ 9/7
x2 0 0 1 0 1/2 0 6

x1 0 1 0 0 -1/3 1/3 2

Parametric Linear Programming-4


Example: Wyndor Glass Problem

Range Basic Z x1 x2 x3 x4 x5 RHS


of θ Var.

Z(θ) 1 0 0 (-9+7θ)/2 0 (5-θ)/2 27+5θ

x4 0 0 0 3 1 -1 6
9/7 ≤ θ ≤ 5
x2 0 0 1 -3/2 0 1/2 3

x1 0 1 0 1 0 0 4

Parametric Linear Programming-5


Example: Wyndor Glass Problem

Range Basic Z x1 x2 x3 x4 x5 RHS


of θ Var.

Z(θ) 1 0 -5+θ 3+2θ 0 0 12+8θ

x4 0 0 2 0 1 0 12
θ≥5
x5 0 0 2 -3 0 1 6

x1 0 1 0 1 0 0 4

Parametric Linear Programming-6


Procedure Summary for Systematic
Changes in cj
1. Solve the problem with θ = 0 by the simplex method.
2. Use the sensitivity analysis procedure to introduce the
Δcj = αjθ changes into Eq.(0).
3. Increase θ until one of the nonbasic variables has its
coefficient in Eq.(0) go negative (or until θ has been
increased as far as desired).
4. Use this variable as the entering basic variable for an
iteration of the simplex method to find the new optimal
solution. Return to Step 3.

Parametric Linear Programming-7


Systematic Changes in bi
n
• Constraints a x
j 1
ij j  bi for i  1,2, ,m are replaced by

a x
j 1
ij j  bi  i for i  1,2, ,m
• Find the optimal solution as a function of θ
Z*(θ)

0 θ1 θ2 θ
Parametric Linear Programming-8
Example: Wyndor Glass Problem

• y1 + 3y3 ≥ 3 + 2θ
2y2 + 2y3 ≥ 5 - θ

Parametric Linear Programming-9


Example: Wyndor Glass Problem

Range Basic Z y1 y2 y3 y4 y5 RHS


of θ Var.

Z(θ) 1 2 0 0 2 6 -36+2θ

0 ≤ θ ≤ 9/7 y3 0 1/3 0 1 -1/3 0 (3+2θ)/3

y2 0 -1/3 1 0 1/3 -1/2 (9-7θ)/6

Parametric Linear Programming-10


Example: Wyndor Glass Problem

Range Basic Z y1 y2 y3 y4 y5 RHS


of θ Var.

Z(θ) 1 0 6 0 4 3 -27-5θ

9/7 ≤ θ ≤ 5 y3 0 0 1 1 0 -1/2 (5-θ)/2

y1 0 1 -3 0 -1 3/2 (-9+7θ)/2

Parametric Linear Programming-11


Example: Wyndor Glass Problem

Range Basic Z y1 y2 y3 y4 y5 RHS


of θ Var.

Z(θ) 1 0 12 6 4 0 -12-8θ

θ≥5 y5 0 0 -2 -2 0 1 -5+θ

y1 0 1 0 3 -1 0 3+2θ

Parametric Linear Programming-12


Procedure Summary for Systematic
Changes in bi
1. Solve the problem with θ = 0 by the simplex method.
2. Use the sensitivity analysis procedure to introduce the
Δbi = αiθ changes to the right side column.
3. Increase θ until one of the basic variables has its value
in the right side column go negative (or until θ has been
increased as far as desired).
4. Use this variable as the leaving basic variable for an
iteration of the dual simplex method to find the new
optimal solution. Return to Step 3.

Parametric Linear Programming-13

You might also like