Computer-aided distribution network designing system

Abstract

To enhance distribution efficiency, optimizing a large-scale distribution network is necessary. However, in such a network, there are hundreds of distribution I collection locations, and dozens of distribution centers each having tens or hundreds of trucks. Furthermore, the networks optimality should be considered from various aspects such as social, political, and environmental factors, which needs human decision. Thus, the design to optimize such a distribution network is a large-scale combinatorial and man-in-the-loop problem.To solve this problem, a computer-aided distribution network designing system is proposed, which has intelligent Genetic Algorithm (GA) and goal-oriented strategic Knowledge Based (KB) control framework. The former guarantees interactive responsiveness and realizes experts' level accuracy. The latter supports to design optimal networks through almost automatically designing distribution networks that satisfy human input goals. Experimental results proved that the proposed method was useful for supporting the design of large-scale distribution networks totally optimal from various aspects such as social or political factors.