The Scheduling Method Research Of MTO Supply Chain On Single Machine

Compared with the MTS production mode, MTO requires that customer demand drives production, so it can fully improve customer satisfaction. The relationship between manufacturers and customers becomes closer. In the supply chain, the inventory and the corresponding inventory cost are greatly reduced. Then the production and distribution cost is a main part of the supply chain cost, so the coordinated optimization of both production and distribution has become an important problem to optimize supply chain. Besides, MTO also requires that manufacturers and suppliers can form a strategic alliance relationship and coordinate together to better meet customer needs. Based on the above, we know that the manufacturers operating need to consider the upstream and downstream factors in MTO mode. So we firstly study the coordinated scheduling of production and distribution, and then study the multi-objective scheduling problem considering cost of raw material release.For the first problem, we establish a collaborative optimization model of production and distribution, which is based on manufacturer with a single machine. It assumes that orders have different release times. The scheduling goal is to minimize the sum of delivery cost and penalty cost caused by departure time. For this kind of strong NP-hard problem, this paper analyzes the properties of the optimal solution, and then comes up with a new method of solution presentation. We propose two methods of neighborhood generation, swap neighborhood and insertion neighborhood, and then construct a simulated annealing algorithm. Finally, the experimental results show that the proposed SA algorithm can obtain the satisfied solution with high quality in a reasonable amount of time.For the second problem, it assumes that the release time is a linear decreasing continuous function of extra cost. We consider a multi-objective optimization model of production and supplying. By analyzing the solution to the problem of limited release time, the paper comes up with an algorithm, denoted as JS algorithm. It can obtain the best allocation for a fixed production order. Combined with the characteristics of multi-objective model, the paper constructs an non-domination genetic algorithm with the elite strategy, denoted as NSGA-Ⅱ algorithm. The experimental results show that the proposed NSGA-Ⅱ algorithm can provide multiple sets of Pareto solution set for deciders.