A path-dependent multi-period rebalancing model for stochastic U-shaped disassembly lines

With the increasing environmental concerns and stricter regulations, manufacturers face serious pressure to design sustainable and efficient end-of-life (EoL) product recovery systems. Disassembly plays a central role in these systems, yet keeping disassembly lines balanced and efficient under the changing conditions of real-world remains a major challenge. In practice, these conditions often require rebalancing the line. Unlike existing studies that consider rebalancing as a one-time independent adjustment, this study proposes a Mixed-Integer Linear Programming (MILP) model for multi-period, path-dependent rebalancing of U-shaped disassembly lines reflecting the sequential nature of real operations where each decision shapes the next. To address uncertainty, a chance constraint is incorporated for stochastic task times, while the objective minimizes both task relocation and workstation operating costs as demand vary. Tested and benchmarked on different instances, the findings showed that the proposed model achieved cost savings of approximately 20% under high-demand conditions, 17% under cyclical demand, and 5% under low-demand scenarios outperforming the independent rebalancing strategy commonly adopted in the literature. To the best of our knowledge, this is the first study integrating U-shaped layouts, stochastic task times, and path-dependent rebalancing, providing a flexible decision-support tool for managers in dynamic environments.