We study a collaborative scheduling problem where the dominant manufacturer (DM) outsources several processes to collaborative manufacturers (CMs). We consider multiple types of products defined by different collaborative production networks which may share some common CMs. The DM assigns the orders to the CMs and coordinates the production scheduling among the CMs. We analyse a practical order-merging strategy which combines a subset of same-type orders throughout the production aming at the cost benefit by continuous processing. Such a strategy is easy to implement but may increase the scheduling cost because of increased deviation between the completion time and the required delivery window of some orders. To this end, a heuristic algorithm based on a learning mechanism and ant colony optimisation is proposed for solving the collaborative scheduling problem under the order-merging plans. Simulation case study was carried out to compare the costs under different merging options. We conclude that merging more orders could lead to a better cost in a few cases, and the overall system performance is more sensitive to the merging schemes when the network capacity is tighter due to disruption. Our study provides an effective solution approach and managerial insights for collaborative scheduling problems with practical order-merging strategies.