A hybrid differential evolution algorithm to vehicle routing problem with fuzzy demands (Q2389564)

The authors consider a non-deterministic version of the vehicle routing problem. The deterministic version of the problem consists in finding routes of minimal total cost, beginning and ending in a central depot for a fleet of vehicles to serve a number of customers. Each customer must be visited exactly once by one vehicle possibly under some additional constraints. The authors describe a fuzzy version of credibility measure theory. Using this theory, a vehicle routing problem with fuzzy demand is introduced. A chance constrained program model is presented. After that, stochastic simulation and differential evolution algorithms are integrated to design a hybrid algorithm, which focuses on minimizing the total traveled distance. A dispatcher preference index expressing dispatcher's attitude to risk is introduced and its in a certain sense best value is obtained by the hybrid algorithm. The effectiveness of the hybrid algorithm is illustrated by numerical examples in the concluding part of the paper.