Unidirectional AGV guidepath network design: A heuristic algorithm

Abstract

In an AGV system, the design of the guidepath network is one of the most important factors that determine the system effectiveness. Many alternative guidepath design schema have been introduced in the literature. However, the methods generally address the static production environment where the product mix or machine routings are assumed to be stable over time. In today's dynamic production systems with small lot sizes and short product cycles, the assumption of an unchanging product mix over an extended period is unrealistic in some situations. Considerable system inefficiency can be introduced when a network designed under the assumption of a stable product mix is used when indeed the product mix has changed. To avoid this kind of hidden source of inefficiency, what is needed is a tool that can recognize when a previously designed network is inappropriate for a new product mix and then uses the new product routing information to reconfigure a new network appropriate for the new production condition. A heuristic algorithm for the design of AGV guidepath in an environment with changing product mix is proposed here. The algorithm not only makes it possible to adapt the system network as the product mix changes, but also produces new designs at reasonably short computational time.