Research On Optimal Scheduling Of Automated Guided Vehicles In Flexible Manufacturing Systems Based On Petri Nets

A flexible manufacturing system(FMS)is a complicated and technology-intensive automation system composed by logistics flow and information flow which has drawn much attention of the academic researchers.Its schedule and control system,which is the focus of FMS,is ensure for the coordination and efficient operation of the entire system.As an important tool for logistics information and automation,automated guided vehicles(AGV)are widely applied in FMS.However,there are so many resources in a FMS that how to schedule AGVs to maximize its profit is a challenging problem.Companies are increasingly concerned about their overall operating efficiency,and require advanced logistics scheduling algorithms.Petri net(PN)is a modeling and analysis tool for distributed systems that can be easily used to simulate manufacturing processes in the industry.This work intends to present an approach to the optimal schedule problem of an FMS with AGVs based on PN.Its main research contents are summarized as follows:1.The method is presented to modeling an FMS with AGVs as a timed PN.Starting from the problem,the work analyzes each functional subject in the system,divides the production chain into two parts.The first subnets is to model the routs for AGVs,the second subnets is constructed to describe the production tasks,and the other ones are for the transformation tasks,where we can gradually refine every detail in the subnets.In turn,these subnets are composed into a whole one,which is called a plant net.2.For the scheduling problem of AGVs in FMS,this work uses A* search algorithm to search the Petri net model.Based on the plant net,a heuristic search algorithm is proposed to search the evolvement path by which all tasks can be taken in the shortest time.It should be noted that the heuristic function is obtained by the reachable-tree algorithm and the subnet modeling of the production tasks.In the DEV C++ development environment,a C language program is programmed for the algorithm,and it is used to compute the optimal schedule strategies for an FMS with different numbers of AGVs.3.The simulation software eM-Plant is used to establish the simulation system of an FMS,and the optimal schedule strategies obtained by the proposed algorithm are verified by this simulation system.