Zone design and control for vehicle collision prevention and load balancing in a zone control AGV system

Abstract

Because of their routing flexibility, automated guided vehicles (AGVs) have been used in many manufacturing systems, especially those in which parts with diverse and complex processing routes are made. In recent years, there have been many studies on AGV-related problems. One of them is preventing the collision of vehicles. Some traditional vehicle collision prevention strategies are zone strategies that divide guide paths into several non-overlapping zones and restrict the presence of only one vehicle in any zone at any time. Traditional zone strategies are fixed zone strategies in which the area of a zone is fixed and vehicles are not allowed to help each other. Because of these restrictions, fixed zone strategies cannot always satisfy the system’s transportation demand if there is a load imbalance between vehicles of different zones. In this paper, a new zone strategy is proposed for a zone control AGV system with a network guide path. The proposed strategy is a dynamic zone strategy, which is different from a fixed zone strategy. It relies on two methods – zone partition design and dynamic zone control – to prevent the collision of vehicles and to maintain the load balance between the vehicles of different zones. The zone partition design defines a relationship coefficient, which measures both the distance relationship and the flow relationship between workstations, and uses it to find an initial zone partition design. We then improve this initial design by an SA (Simulated Annealing) based improvement procedure to achieve a better load balance result. The dynamic zone control uses two methods – zone repartition and load sharing – to ensure that vehicle collision can be prevented and the system’s load balance can be maintained when the system is in operation. Simulation experiments were conducted to understand the performance of the proposed strategy. The simulation results show that the proposed strategy outperform the fixed zone strategy in throughput, WIP inventory, and flow time. The results also show that the proposed strategy is able to adapt to any changes (in the system) that cause the load imbalance problem between zones.