Research On Dynamic VRP Based On Real-Time Information

With the rapid development and going mature of information and communication technology, enterprises pay more attention to optimize distribution using these technologies, in order to improve economic efficiency and achieve scientific logistics. We introduce the knowledge of intelligent transport systems on the basis of former studies, so we can arrange vehicle routing based on real-time information so that the results can reflect reality better. In the last part, the usefulness of the model is proved by an example.The model mainly deals with two issues. The first is the acquirement of real-time information, and the second is dynamic vehicle routing, the result of which has the shortest time rather than the shortest distance. Real-time information includes road information and order information. The observation system which links to traffic information center sees to the requirement of real-time information. Order information is communicated directly from outside to the order and fleet management system. The observation system also takes in charge of calculation of the shortest route between two random nodes to provide data support for the plan system. The plan system is responsible for the route on which the vehicle traverses all the client points on the shortest path, and the updating based on the change of real-time information.The main works of research are as follows:(1) The dynamic vehicle routing model is built on the basis of the description and analysis of dynamic vehicle routing problem and the function of all parts of the model are illustrated;(2) After roads' pretreatment, automatic parting and node matching, the road layer in Dalian electric map is topologized, and the redundant data is eliminated, so that the platform for the model is built;(3) The theory of transportation information center's acquirement of road information and how the observation system links to it is analyzed. The demand function of the shortest route between two random nodes is achieved by using Dijsktra algorithm;(4) The primary and dynamic routing procedure is designed based on the idea of Dijsktra algorithm. The optimizing strategy in the dynamic routing is incomplete re-optimizing. By this way the routing time can be saved, and the complexity of the problem is decreased;(5) Found on the above theories, the application of the model is studied by the example of snack flow distribution. The dynamic routing in the given area is achieved and we get satisfactory result. The research is significant on theory and utility. Applying real-time information in logistic distribution can enable the routing process more dynamic, and the result can be closer to reality; The topology of the digital map can provide data support for homologous researches; The plan aims at the rapidest services for clients, which makes distribution service quicker and more flexible.