Infeasible Model unbound Excel OpenSolver

ユーザーの入力を待っています。

Adrian Duran

2022年04月18日 10:39 編集済み

• 
• 

Hello to all,

I have a problem since I am using OpenSolver to which I have added Gurobi.
The peculiarity of my problem lies in the fact that my variable has a conditional constraint
 Xi < Upperbound.

This Upperbound must be the minimum between a constant and a function that depends on the sum of Xi-n (n=0 to i-1):

Vi-1 = Sum(Xi-1) -> (Filling of the day before).

This is a filling of a minimum cost store.

This nonlinear term can be easily be linearized:

Xi≤H, where H=min(Constant ,f(Vi-1))

Becomes two inequalities (assuming you have a maximization problem)

Xi≤Constant
Xi≤f(Vi-1)

When I solve separately there are no problems but when I add both constraints I get the error: Unfeasible or Unbounded model.

How can I solve it? When I put in the MIN(Hconstant, Hf(V-1) sentence, it doesn't let me solve either.

I would be very grateful if you could solve my problem, thank you very much in advance.

Thanks in advance,

ADM

 

0

• 正式なコメント

  

  Simranjit Kaur

  • Gurobi Staff

  2025年11月16日 23:39

  This post is more than three years old. Some information may not be up to date. For current information, please check the Gurobi Documentation or Knowledge Base. If you need more help, please create a new post in the community forum, or try Gurobot, our chatbot interface offering instant, expert-level support.
• 

  Maliheh Aramon

  • Gurobi Staff

  2022年04月20日 00:08

  Hi Adrian, 

  What are the coefficient values of variables \(x_i\) in your maximization model? Are they negative?

  The best way to figure out why your model is infeasible is to compute one of its IIS (Irreducible Inconsistent Subsystems). You can do this using the model.computeIIS() method of the Gurobi Python API. I looked at the OpenSolver list of APIs and it seems to me that there is no API to find IIS using OpenSolver directly.

  You can either use one of the Gurobi Python APIs directly to figure out why your model is infeasible. Alternatively, you can save your model as an MPS file following our instructions in How do I write an MPS model file from a third-party API? and share it with us using OneDrive, Google Drive, or Dropbox.

  Best regards,

  Maliheh

  0

投稿コメントは受け付けていません。