Multi-product Multi-stage Production Planning with Lead Time on a Rolling

Abstract Responding to a scalable production system in a make-to-order environment requires increased effective decisions. This paper considers challenges brought by lead time existed in a multi-product multi-stage manufacturing system when making a short-term production planning. The order-based production routes enhance the production flexibility, and in the meanwhile complicate the decision-making for each production stage. We define a series of sets to aggregate the production route and then model the problem as a mixed integer linear program (MILP) problem. The model seeks to find optimal production lots with several practical extensions. To efficiently capture the production dynamics induced by the lead times and setup, the production planning is implemented on a rolling horizon basis. The model is divided into several sub-problems as the horizon is rolled forward, of which a fix-and-relax strategy is applied. A study is conducted using data from a real case to evaluate the performance.