LP objective function with unknown parameters

Click For Summary

Discussion Overview

The discussion revolves around linear programming (LP) optimization problems that involve an objective function with an unknown parameter, specifically focusing on graphical methods for minimization and maximization. Participants explore how to approach such problems and the implications of the unknown parameter on the solutions.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested

Main Points Raised

  • One participant presents a specific LP problem with an objective function f = 2x_1 + λx_2 and constraints, seeking guidance on how to approach similar problems.
  • Another participant suggests a method for solving the problem graphically by plotting the constraints and the objective function, indicating that the optimal solution will lie on the boundary defined by the constraints.
  • A different participant argues that it is not possible to determine the maximum and minimum values of the objective function without knowing the value of λ, as the evaluation at the vertices of the feasible region will depend on λ.
  • One participant questions how to ascertain whether the problem has an optimal solution, or if it could be infinite or have no solution, depending on the value of λ.

Areas of Agreement / Disagreement

Participants express differing views on the role of the unknown parameter λ in determining the optimal solution, with some asserting that it is essential for evaluation, while others explore graphical methods without resolving the implications of λ.

Contextual Notes

The discussion highlights the dependence of the solution on the unknown parameter λ, which introduces uncertainty regarding the existence and nature of optimal solutions. There are also unresolved aspects related to the graphical interpretation of the constraints and the objective function.

Who May Find This Useful

This discussion may be useful for individuals interested in linear programming, optimization techniques, and graphical methods for solving mathematical problems involving parameters.

Mark J.
Messages
81
Reaction score
0
Hi All,

I am struggling with minimization(maximization) problems which needs to be solved graphically but they have unknown parameter in objective function:

For example:

f = 2x_1 + \lambda x_2(min)

for conditions:

-x_1 + x_2 \leq 3
x_1 + 2x_2 \leq 12
3x_1 -x_2 \leq 15
x_i \geq 0

More than solution I need to understand the way so I can proceed for similar examples

Regards
 
Physics news on Phys.org
Mark J. said:
Hi All,

I am struggling with minimization(maximization) problems which needs to be solved graphically but they have unknown parameter in objective function:

For example:

f = 2x_1 + \lambda x_2(min)

for conditions:

-x_1 + x_2 \leq 3
x_1 + 2x_2 \leq 12
3x_1 -x_2 \leq 15
x_i \geq 0

More than solution I need to understand the way so I can proceed for similar examples

Regards
I am having trouble reading this. can you redo it?
 
There are 4 straight lines limiting the search area and one as function f to minimize. Draw all 4 lines in cartesian coordinates, i.e a piece of paper with x_1 and x_2 axes. Then determine the areas defined by them (hatch them). Finally draw f and look whether you have to push it up or down to minimalize it's value. The searched point will be on the boundary you drew.
 
No it is not possible to determine max and min without knowing \lambda. The basic "rule" of linear programming is that max and min of a linear function on a convex polygon occurs at a vertex. It is fairly easy to determine the vertices of the given convex polygon but when you evaluate f at the vertices, the value will depend upon\lambda so that knowing which is largest and which is smallest will depend upon \lambda.
 
So how to determine when problem has optimal solution. infinite or no solution depending on lambda??
 

Similar threads

  • · Replies 17 ·
Replies
17
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 0 ·
Replies
0
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 20 ·
Replies
20
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 61 ·
3
Replies
61
Views
13K
  • · Replies 100 ·
4
Replies
100
Views
13K
  • · Replies 61 ·
3
Replies
61
Views
12K