A vehicle routing problem with distribution uncertainty in deadlines

This article considers a stochastic vehicle routing problem with probability constraints. The that customers are served before their (uncertain) deadlines must be higher than pre-specified target. It is unrealistic to expect the perfect knowledge on distributions of always available. To this end, we propose distributionally robust optimisation framework study worst bounds problem, which exploits moment information historical observations. includes two steps. We first use Conditional Value-at-Risk (CVaR) as risk approximation missing customer deadlines. resulting nonlinear model then transformed into semi-infinite mixed integer program, using dual form CVaR approximation. A sample approach used address computational challenges. As standard constraints rather conservative, suggest relaxation and develop an iterative algorithm find right value parameter introduced relaxed extensive numerical experiments show policies developed by proposed solution able achieve required target, regardless deadline distributions.