Dynamic conflict-free routing of automated guided vehicles

This paper presents work performed to address design and operational control issues for an Automated Guided Vehicle (AGV) based material handling system. Various design issues connected with flow path design, such as traffic flow along the guide paths, location and representation of intersections, buffers, pickup and dropoff points, are outlined. A network representation of the AVG system (AGVS) is presented. Operational control factors, such as demand selection and assignment, route planning and traffic management, idle AGV positioning, and AGV characteristics, are addressed. In particular, the performance of six demand selection policies, four demand assignment policies, and two idle AGV positioning policies are outlined. A very effective route-generation technique that provides conflict-free routes for multiple AGVs with varying speeds is presented. A large-scale simulation of a dynamic batch type manufacturing system to test the feasibility of the proposed scheme for real time control of AGVs and to determine the effect of several factors on several performance measures is presented. Our findings indicate that performance is significantly affected by vehicle speed, number of vehicles, demand arrival interval, idle AGV positioning, and the demand selection and assignment policy in use.