Research On Feeding Problem Of Mixed-flow Assembly Line Based On Just-in-time Cycle Distribution

In the context of the diversification of demand in today's automotive market,JIT has become a new requirement for automotive manufacturing and logistics systems.The use of material supermarkets and tow train for cyclical distribution of feeding stations is widely used by automobile manufacturers to achieve just-in-time logistics.Therefore,in-depth study of this issue has important theoretical and practical significance.In order to propose a more timely production model of mixed-model assembly line based on supermarket production,this paper considers a series of production factors and constraints of the mixed-model assembly line itself,and proposes an optimization model based on this: A series of production factors in the mixed-model assembly line are transformed into methods for establishing the parameters required to optimize the model.Based on this,a distribution cost and inventory cost are optimized,and the loading and unloading time,capacity and safety of the tank are additionally considered.A mixed integer programming model for inventory and improved objective functions.A small-scale example was designed to verify the solution using the Gurobi solver.The results show that the proposed model is realistic and solvable.Since the problem is NP-hard,in order to solve large-scale problems,this paper proposes a two-stage heuristic algorithm to solve the model.The numerical experiments show that the method using a single scheduling rule has theoretical defects.In order to solve this problem,a Memetic algorithm-based heuristic method embedding the Memetic algorithm into the heuristic algorithm is proposed and the heuristic algorithm based on simulated annealing is used.The results show that the heuristic method based on Memetic algorithm can find the optimal solution in the small-scale problem,and the accuracy and stability of the solution are superior to the other two algorithms in large-scale problems.Finally,on the basis of considering the single tow train model,considering the adverse effect of loading and unloading bin time and load constraints on the distribution cost of bicycle model,a multi-trailer model based on segmented distribution is proposed and the algorithm is improved.Experimental results show that the proposed algorithm is feasible and effective.In the same large-scale study,the parameters were analyzed in terms of the number of tow trains and the load.The results show that as the number of tow trains increases,the sum of distribution costs and inventory costs decreases,but the marginal benefits of increasing tow trains are also reduced.At the same time,the experimental results are summarized.The conclusions provide a useful reference for selecting the number,load,price and allocation scheme of the assembly station in practical applications.