A Knowledge-Based Model Representation and On-Line Solution Method for Dynamic Vehicle Routing Problem

摘要

We propose a knowledge-based model representation and on-line solution method for dynamic vehicle routing problem (DVRP) in the paper in order to realize on-line modeling and solution process of the problem. This knowledge-based model representation is composed of six components-B (Basic Data collector), R (Restrictions), I (Initial state generator), S (State operator), G (Goal state), and C (Controller). We term the representation BRISGC six-component model representation. Based on the representation, an on-line solution approach to DVRP is presented. And a real-world DVRP in e-Commerce is solved by the representation and solution method in case study. The result proves that the proposed approach is effective for online and real-time vehicle routing.