Vehicle dispatching system based on Taguchi-tuned fuzzy rules

Abstract

This paper presents the development of a dispatching system for a fleet of automated guided vehicles in a flexible manufacturing environment which is based on a hybrid Fuzzy-Taguchi approach. A fuzzy decision-making system emulates the human behavior necessary for multi-objective directed decision making in a dynamically evolving environment. A statistical approach based on the Taguchi method tunes the fuzzy rules to achieve near optimal performance. Simulation results demonstrate the effectiveness of this marriage of computational tools in dealing with the well-known NP-complete scheduling problem.