A multi-stage stochastic programming approach for overhaul and supply chain planning of modular physical assets: case study

Effective overhaul planning is vital for optimizing performance and cost-efficiency in industrial organizations reliant on physical assets. Focusing on modular subsystems within a closed-loop supply chain, this paper addresses the challenge of determining the optimal timing for overhauling repairable modules to minimize costs, meet deadlines, and ensure asset availability. The closed-loop structure encompasses inventory holding, transportation, and overhaul operations, capturing the interdependencies between logistics, maintenance, and spare parts management. To manage uncertainties from major asset failures and variable demand, a multi-stage stochastic programming model is proposed. A sample average approximation method is employed to address the computational complexity arising from the large scenario space. Real data from a port operator’s asset management case is used to evaluate the model. Results show a potential 17% reduction in total ownership costs. Additionally, the model provides managerial insights, enabling decision-makers to determine the required number of spare units and the level of investment needed to reduce future uncertainty.