How to reduce model run time for an MIP problem

Question

Hi Team,I am trying to solve a furnace loading problem. There are 60 items to be loaded over the furnaces. The number of furnaces are 32 where 16 of them are identical in nature and other 16 are identical in nature. I have modeled it as MIP problem with objective of capacity maximization in terms of weights. Model is giving correct answer but I am facing run time issues, due to the exponential complexity model is taking longer time to solve (almost more than 1 hour). Note that I have created virtual location inside the furnaces due to other constraints and I am also using some constraints to break the symmetry part. One way to reduce run time is to run it through loop but I do not want to use the loop. Is there any alternative way I can handle.

Marcel Hunting · Accepted Answer

Hi @RwitamJ. Some tips:I highly recommend reading the paperPractical guidelines for solving difficult mixed integer linear programs byEd Klotz and Alexandra M.Newman (2013).Try different solvers: CPLEX, Gurobi and ODH-CPLEX, using the latest version.Print the solver log and analyze it (and maybe shareit here).In my experience the best way to improve the performance is byimprovingthe model formulation. We can only help you with that if you share your project.

Reply

Sign up

Login to the community

Scanning file for viruses.

This file cannot be downloaded