Robust remanufacturing job shop scheduling under carbon regulatory policies using modified discrete particle swarm optimization algorithm

Remanufacturing job shop scheduling (RJSS) faces the dual challenges of the inherent stochasticity of End-of-Life (EOL) products and the compliance pressure from stringent carbon regulatory policies. However, existing sustainable scheduling research predominantly relies on deterministic assumptions, overlooking interactions between these operational uncertainties and dynamic carbon costs. To bridge this gap, this study proposes a new robust RJSS model under two carbon regulatory policies (RRJS-CRP). The proposed model employs a discrete scenario set to characterise uncertain arrival and processing times of EOL products. It aims to achieve robust solutions that can be effectively implemented in various scenarios by minimising expected total cost and cost discrepancies. To solve the proposed model, a modified discrete particle swarm optimisation algorithm (MDPSO) is presented. The algorithm integrates a particle update mechanism, a path relinking procedure, a local search strategy, an elite replacement strategy, and a stagnation reduction strategy to enhance the algorithm’s performance. Finally, comparative experiments, statistical analysis, and a real-world case study demonstrate the algorithm’s competitive performance against the Gurobi solver and five baseline algorithms. Moreover, sensitivity analysis reveals that emission reduction decisions are inelastic to carbon tax variations, whereas a moderate emission cap optimally balances economic efficiency and environmental performance.