2014
EMSS 2014 Proceeding
Modeling the multi-compartment vehicle routing problem with stochastic demands
Authors: Jan Melechovsky
Abstract
Vehicle routing problems represent a class of combinatorial optimization problems widely studied in the literature. Practical applications often involve a level of uncertainty, which complicates the decision making process. This paper presents two models of the multi- compartment vehicle routing problem with stochastic demands (MCVRPSD). The problem consists in finding a minimum cost set of vehicle routes serving a set of customers. Each customer can require the delivery of m products. The products must be transported in different compartments due to their physical incompatibility. The problem is modeled as a stochastic program with recourse. The recourse action consists in a return to the depot vertex whenever the servicing vehicle cannot satisfy the customer?s demand for a particular product. In an alternative model the failure can also occur due to time constraints. A hybridized evolutionary algorithm is presented to address the problem.
I3M Scientific Sponsors
I3M Industrial Sponsors
I3M Media Sponsors
