中文摘要：

针对冷轧薄板生产线流程长、产品种类多的特点，为提高合同的交货准确度，将一段时期内的生产合同进行组批处理，并对合同各批量的机组作业时间进行模糊预测，建立了具有模糊处理时间的冷轧生产线合同组批调度模型。模型将批量合同作为基本调度单元，将其归结为一类模糊Job Shop问题，并采用改进的单亲遗传算法与特殊的邻域搜索策略相结合的方法进行求解。通过上海宝钢冷轧薄板厂大量的生产数据进行仿真实验，批量平均交货期满意度达到90％以上，表明该调度模型与算法能够达到较好的应用效果。

英文摘要：

Aiming at the characteristics of long process and lots of product types in cold rolling sheet production line and in order to increase the delivery on-time ratio of production order, the orders within a certain period were pro- cessed in groups and the batches＇ processing time on each machine was predicted by a fuzzy method. A fuzzy batch production scheduling model based on cold roiling line was established, in which each batch was regarded as the basic scheduling unit. This model was summarized as a kind of fuzzy Job Shop scheduling problem, which was solved by the Modified Partheno-Genetic Algorithm （MPGA） and a special neighborhood search strategy. A large number of production data of a cold rolling plant in Shanghai Baosteel Co. Ltd. were simulated to verify the effectiveness of the proposed model and algorithm. Experimental results showed that the average satisfaction ratio of due-date could reach to more than 90%, which revealed that this presented method had satisfactory application effects.