VRASP

Vehicle Routing and Appointment Scheduling Problem(VRASP) with stochastic traveling and service time. Use Sample Average Approxiamation(SAA) to transform the stochastic problem into a deterministic one. Contain both CPLEX 12.10 and Variable Neighorhood Search(VNS) impementation.

Dependencies

Anaconda Python 3.7
cplex 12.10
docplex 2.11
numpy 1.21.6
pandas 1.3.5
scipy 1.7.3

Files

• .\experiemnt: data for computational experiments
  • problem size--[5, 10, 20, 30, 40]
  • number of instances--20
  • nunber of samples in SAA implementation--[30, 50, 80, 100]
• .\utils\data_model: Class
• .\utils\gen_data: generate experiment data, save them into .txt file
• .\utils\readwrite: load the experiment data
• .\vns_de: VNS algorithm for deterministic VRASP
• .\vns_saa: VNS algorithm for stochastic VRASP using SAA
• .\cplex_solver: CPLEX implementation for both deterministic and stochastic VRASP
• .\vns_main: VNS runner