A Heuristic Approach for a Multistage Lotsizing Problem with Dynamic Product Structure

We consider a multistage lotsizing problem with capacity-constrained resources, general product structures and nonzero lead times. This problem is motivated by a real-life application from a large manufacturer of electronic devices and deals with the determination of a production plan for the end items and their components to meet external prioritized demand in each period of a given planning horizon. Since resources are scarce, the main objective is to obtain a production plan that minimizes the total number of late orders. The problem is substantially complicated by a number of side-constraints that arise in many industrial settings. In addition, as a result of short product life cycles, the bill-of-materials (BOM) has a dynamic structure meaning that components are replaced by new ones during the planning horizon. Since real-life problems are too hard to solve optimally in reasonable computing time, one is compelled to use heuristic solution methods. We propose a two-step greedy heuristic which first generates a solution to the uncapacitated version of the problem. In the second step, capacity feasibility is attained by shifting production of product sub-structures backward and forward in time.