A Bayesian-evolutionary game framework for robust multi-agent workforce scheduling in resilient supply chains
This study develops a Bayesian–Evolutionary Hybrid Game with Robust Stochastic Optimisation (BEHG-RSO) framework for coordinated workforce allocation under supply-chain disruptions. Bayesian partner-reliability estimates and evolving cooperation shares enter a two-stage mean–worst-case mixed-integer scheduling model through effective productivity and reserve-sharing capacity. The evaluation combines public benchmark data, multi-agent simulation and a paired 30-seed comparative analysis. The experiments report 93.6% workforce utilisation, a 9.9% reduction in scheduling cost relative to the multi-agent reinforcement learning comparator and a 25.5% reduction in recovery time. Component-ablation, paired uncertainty and computational-scaling analyses further characterise performance under the evaluated disruption scenarios. The results demonstrate the framework’s operational trade-offs and computational requirements.
Authors
- 于诗矛
- Fengbo Cui
Institutions
- Anhui Xinhua University (CN)
Publication Details
- Journal
- Scientific Reports
- Published
- 2026-11-21
- DOI
- https://doi.org/10.1038/s41598-026-80395-3
- Primary Topic
- Supply Chain Resilience and Risk Management
- Type
- article
- Field-Weighted Citation Impact
- 0.00