A simulated annealing method directed by a shifting-bottleneck for solving serial multi-factory scheduling with batch transportations

This paper investigates the production and delivery scheduling coordination where a set of jobs should be scheduled for delivery to other factories for further processing and finally to the customer. The high delivery cost in manufacturing systems is the main motivation of integration of production and delivery scheduling which contains a more holistic view of the supply chain problem. So constituting batches of jobs may reduce the transportation cost but on the other hand, it may cause extreme increase in job’s tardiness. The optimization criterion is the minimization of the sum of total tardiness and delivery cost. A lower bound is proposed for this problem. Based on the theory of constraints, in such tandem production lines the performance of the system is significantly related to the performance of the weakest factory in the system. So, the main focus of this heuristic is on finding the bottleneck and also a good sequence for jobs on the bottleneck factory. Jobs should also be scheduled on downstream factories through forward scheduling and on upstream factories through backward scheduling. We proposed a simulated annealing approach directed with this heuristic for our problem. We conduct computational experiments to test the effectiveness of the method. The comparison is done on the solution obtained from presented method and lower bound. The computational results show that the good performance of the heuristic for most of the problem instances.