Multi-objective Vehicle Scheduling Problem For JIT Purchasing

With the increasingly fierce market competition,JIT(Just In Time)production mode with multi-variety and small batch characteristics is more and more widely used in enterprise manufacturing.Therefore,JIT purchasing and delivery in small batch,multi-cycle and on time has become an inevitable trend.Due to the diversity of raw materials demand in continuous production process,delivery of raw materials to production in an accurate time is the key to ensure the uninterrupted production.In order to ensure the uninterrupted production,the traditional mode of procurement and transportation is mostly driven production.Large quantities of raw materials are purchased before production,and the response speed to sudden orders is slow.Moreover,due to the large storage of raw materials,the quality of products is reduced.Production costs have been raised.In this paper,the production of SEDL company is taken as the research object,and combined with the existing knowledge of vehicle transportation,JIT(Just In Time)procurement and lean production,a multi-cycle,small batch and split procurement mode is proposed to replace the previous inventory-oriented procurement and distribution mode,so as to realize the timely delivery of raw materials between production lines and suppliers without passing through it.He stores it in the middle.According to the actual situation of production demand,aiming at minimizing the completion time of procurement and transportation and the number of vehicles used,the impact of procurement route scheduling on transportation cycle and procurement volume is studied.A JIT procurement-continuous production scheduling model is established,and the characteristics of the problem are analyzed.A grid-based adaptive multi-objective artificial bee colony algorithm(Grid-based adaptive multi-objective artificial bee colony algorithm)is designed.-Objective artificial bee colony algorithm,GAMOABC.In the algorithm,we first use grid to save the optimal solution,and update and maintain the optimal solution in the adaptive grid,guarantee the diversity of the Pareto solution set,and design the location shared artificial bee colony algorithm in the grid to improve the accuracy of the solution set.Local search and dynamic random search are performed in the grid to optimize the solution or generate a new Pareto solution in other grids.Secondly,two dimensional matrix coding is used to represent priority values of vehicles and production materials.In the decoding process,in order to ensure the continuity of production,the scheduling set is determined by using the current consumed completion time of raw materials,and the heuristic information of selection is designed for the number of vehicles used and the transportation time.Finally,the test results and comparative experiments show that compared with NSGA-II and MOEAS algorithms,the Pareto disassembly set obtained by GAMOABC algorithm is more diverse and accurate.Compared with other JIT procurement and transportation,the effectiveness of multi-cycle,small batch,separable transportation strategy and heuristic information design is verified.By optimizing the scheduling of procurement and transportation routes in the production line,SEDL company has greatly improved the efficiency of procurement and transportation and reduced production costs.