Nowadays, one of the most important challenges faced by cold-rolling mills industries is the adjustment of the production mode in order to satisfy market requirements subjected to fluctuations over time, mainly product varieties, demands, etc. Therefore, an order batching model including three objectives is proposed, which are to minimize the difference of delivery, the difference of process route, and adjustment times of mills. Meanwhile, the model meets the requirements of process constraints such as the weight, the width of outlet(inlet), the thickness of outlet(inlet), and the tensile strength. And then a fuzzy clustering algorithm based on an improved particle swarm optimization which is used to solve the model is established. Experiments with real data of a cold-rolling mills production processes indicate that the proposed model and algorithm are superior to the FCM algorithm and can meet the requirement of the batch planning.