How to compute the objective and feasibility of a manually given solution?

Answered

liu heng

December 07, 2023 01:35

• 
• 
• 

Suppose "sol" is a list that consists of the value of each variable.(sol is given by user)

How to compute the objective and feasiblity of sol

0

• 

  Jaromił Najman

  • Gurobi Staff

  December 07, 2023 09:26

  One easy way is to fix every variable to its sol value. The following pseudo code should give you an idea

  sol = ... # given solution values
  for var in model.getVars():
    # fix variable to its sol value
    var.lb = sol[var]
    var.ub = sol[var]
  
  # optimize model
  model.optimize()
  
  if model.status == GRB.OPTIMAL:
    print(model.ObjVal) # print objective value
    print(model.MaxVio) # print maximum violation of the solution
  
  if model.status == GRB.INFEASIBLE:
    # if given sol values are infeasible generate an IIS file called iis.ilp
    # to easily check why it is infeasible
    model.computeIIS()
    model.write("iis.ilp")

  For additional information please refer to the Attributes documentation and How does Gurobi compute the IIS for infeasible models?

  Best regards, 
  Jaromił

  0

Please sign in to leave a comment.