Originally posted by: labrecheMustapha

Hi,

I implement a VRP with special constraints. the execution works but the model is apparently unfeasible.
Can you help me please.

Cordially.

Your problem looks feasible here. Are you sure you attached the right data? Can you please show the engine log as well? What is the version of CPLEX you are using?

Originally posted by: labrecheMustapha

Hi,

sorry this is the right files.

thank u for reply.

Indeed, this model is infeasible. And the conflict refiner comes back with a very blatant infeasibility:

 c1:   y(2)(1) + y(2)(2) + y(2)(3) + y(2)(4)  = 1
 c114: y(2)(1) + y(2)(2) + y(2)(3) + y(2)(4)  = 0

The sum of y variables in these constraints is required to be 1 and 0 at the same time! This should be easy to track down in your model. You may want to take a look at the respective chapter in the user manual: https://www.ibm.com/support/knowledgecenter/SSSA5P_12.8.0/ilog.odms.ide.help/OPL_Studio/usroplexamples/topics/opl_mp_examples_relaxation.html

Originally posted by: labrecheMustapha

Thanks @DanielJunglas 548c249b-bb6b-42f6-99b8-f0376c80829e for ur help. I cheked my data file and i found some wrong values.

I would like to know if it is not asked too much and if it is not confidential, how did you get this writing of constraints. Because i am writing a model with special constraints (VRP for the inspection of structures) I will often have to introduce difficult restrictions and this representation of the constraints will be of great help

First of all, you should read the chapter in the user manual, I pointed to. This explains how you can handle conflicts within the IDE.

If that does not help, here is how I obtained the list of conflicting constraints for your model:

1. Create a settings (.ops) file

2. In that file, choose SAV as "export format"

3. Solve the model. This will create a file the name of which ends in ".sav"

4. Use the interactive optimizer to find the conflict:

CPLEX> read yourmodel.sav
CPLEX> conf
CPLEX> disp conf all

Originally posted by: labrecheMustapha

Thanks so much for it @DanielJunglas.