Storage Allocation And AGV Path Planning In Parts-to-Picker Mode

With the rapid development of e-commerce and the continuous change of customer order demands,the rapid response of the order picking system is becoming more and more important.Therefore,the Robotic Mobile Fulfillment System(RMFS)in the parts-to-picker mode has been developed,and the core factors that determine the efficiency of warehousing operations in the parts-to-picker mode are storage allocation and automated guided vehicle(AGV)path planning.In order to improve the efficiency of warehouse picking and reduce costs,this thesis studies the two major problems of storage allocation and AGV path planning on the background of RMFS.The main research contents and results are as follows:Firstly,a two-stage model of storage allocation is established by comprehensively considering the correlation of goods and frequency of goods of out of the warehouse.In the first stage,the aggregation degree of correlation of goods is used as an index,and the goods are clustered based on the index.In the second stage,an optimization model of storage location assignment is established based on the clustering of goods and the frequency of out of warehouse.In order to calculate the result of the storage allocation,the algorithm based on Hierarchical Clustering-Genetic Algorithm(HC-GA)is designed.And the simulation results show that the location assignment scheme is feasible and effective.Secondly,the intelligent warehouse environment model is built based on the grid method,and the single AGV path planning model with the shortest total handling time as the index and the multi-AGV path planning model with the shortest time to complete the task as the index are established.Considering the influence of frequent turns on driving time,the improved A~* algorithm is proposed to plan the global path of single AGV,on this basis,the minimum completion time for each task can be determined,and the idea of greedy is introduced to assign the tasks performed by the multi-AGV and determine order in which shelves are handled.In order to realize frictionless path planning,the global optimal path of AGV is preliminarily planned in the static environment,and the forms of multiple AGV path conflicts and corresponding resolution strategies are analyzed and designed,and a timetable is introduced for collision detection in path planning,and local conflict paths are replanned according to the resolution strategies.The simulation results show that the improved A~*algorithm can effectively reduce the number of turns and reduce the time consumption of handling.Finally,according to the actual demands of J Company and the research results of this thesis,the overall design scheme of location assignment and multi-AGV path planning is proposed,and the effect of solving the demands of J Company is analyzed through simulation analysis.The simulation results show that planning the AGV path based on the storage allocation results in this thesis can shorten the time to complete the task by reducing the number of shelves accessed to the AGV and the handling distance,and the optimal number of AGVs to complete a batch of orders is determined by pre-allocation and then precise allocation in the simulation experiment,and the applicability and effectiveness of the improved A~* algorithm in ordinary order scenarios and special order scenarios are verified by setting up the different comparative algorithms.In order to improve the picking efficiency,the storage allocation and AGV path planning in the parts-to-picker mode are studied,and the related research is theoretically feasible.The research results show that the storage allocation in this thesis has advantages in improving the efficiency of warehousing operations,and the AGV path planning can realize the coordinated operation of multiple AGVs.And the research results in this thesis is verified that can effectively solve the actual needs of J Company and similar warehouses,meanwhile,which can provide certain reference value for related fields.