Abstract: The supply chain characteristic of manufacturers which produce non-standard parts is presented. To collaborate optimize order assignment, job shop schedule and batch delivery schedule, a supply chain scheduling problem with multiple plants and customers is applied. A mathematical model for the problem is first built to minimize a weighted sum of the maximum lead time and total cost. Moreover, based on the optimal condition analysis, its hybrid taboo search algorithm with vector array representation is designed. The hybrid algorithm systematically partitions the feasible region, applies taboo search algorithm with insert and swap neighborhood operations to select sub-region, and uses taboo search algorithm with block structure neighborhood operations to search the optimal solution of the selected sub-region. The hybrid algorithm is applied to optimize the example and the results by different solution structure and heuristic algorithms are compared. Results and analysis demonstrate the validity of the proposed model and its algorithm.

Key words: Batch delivery, Multiple plants, Supply chain scheduling, Taboo search