borelian / adaptivebenders

Code and computational experiments of the paper "Benders Adaptive-Cuts Method for Two-Stage Stochastic Programs" by Cristian Ramírez-Pico, Ivana Ljubić and Eduardo Moreno. arXiv:2203.00752

Adaptive Benders

Source code and instances for the computational experiments of the paper "Benders Adaptive-Cuts Method for Two-Stage Stochastic Programs" by Cristian Ramírez-Pico, Ivana Ljubić and Eduardo Moreno. arXiv:2203.00752

It applies different Benders methods and other optimization methods to solve three stochastic network flow problems. Each problem as its own class file and a run-file to execute an instance

Problems are:

• Stochastic Capacity Planning Problem
  • Class: cpp.py
  • Run file: runCPP.py
  • Instances: CPP_instances/
• Stochastic Multicommodity Flow Problem
  • Class: smcf.py
  • Run file: runSMCF.py
  • Instances: SMCF_instances/
• Facility Location with CVaR
  • Class: flcvar.py
  • Run file: runFLcvar.py
  • Instances: instancesFlcvar/

It requires NumPy library and Gurobi (https://www.gurobi.com) as optimization solver.