A two-level stochastic-adaptive model for managing the operational risks of retail network facilities

The article presents an innovative two-level stochastic-adaptive operational risk management model designed for large distributed infrastructure networks. The study solves the problem of the inability of traditional deterministic models to adequately assess the "tail" risks in conditions of high uncertainty of energy consumption, equipment failures and logistical failures. The proposed methodology combines strategic planning and tactical online adaptation. At the top level, two-stage stochastic programming is used to generate robust maintenance and capacity redundancy plans that take into account the probabilistic nature of threats. Intelligent clustering of objects using self-organizing Kohonen maps allows you to divide the network into four categories: critical, high-risk, logistically vulnerable and stable. At the lower level, reinforcement learning agents (PPO and DQN algorithms) adapt operational solutions in real time using customized reward functions for each cluster. Experimental results confirm the effectiveness of the approach: for critical facilities, the share of downtime has been reduced to 2 %, and for stable facilities, maximum resource savings have been achieved. The implementation of the model makes it possible to reduce operating costs by 10–15% and increase the reliability of critical infrastructure by 20–30%. The model ensures transparency of management through clear KPIs and contributes to the implementation of a sustainable development strategy.

1. Miller C., Fu C., Roth J., et al. The ASHRAE Great Energy Predictor III competition: Overview and results. Science and Technology for the Built Environment. 2020;26(10):1427–1447. https://doi.org/10.1080/23744731.2020.1795514

2. Yu L., Qin Sh., Zhang M., et al. A Review of Deep Reinforcement Learning for Smart Building Energy Management. IEEE Internet of Things Journal. 2021;8(15):12046–12063.​ https://doi.org/10.1109/JIOT.2021.3078462

3. Kohonen T. Self-Organizing Maps. Moscow: Laboratoriya znanii; 2017. 660 p. (In Russ.).

4. Pigott A., Crozier C., Baker K., Nagy Z. GridLearn: Multiagent Reinforcement Learning for Grid-Aware Building Energy Management. arXiv.​ URL: https://arxiv.org/abs/2110.06396 [Accessed 28th December 2025].

5. Dong X., Liu Y., Xu Zh., et al. Optimal scheduling of distributed hydrogen-based multi-energy systems for building energy cost and carbon emission reduction. In: 2020 IEEE 16th International Conference on Automation Science and Engineering (CASE), 20–21 August 2020, Hong Kong, China. IEEE; 2020. P. 1526–1531. https://doi.org/10.1109/CASE48305.2020.9216885

6. Sutton R.S., Barto A.G. Reinforcement Learning. Moscow: DMK-Press; 2020. 552 p. (In Russ.).

7. Butusov I.Yu., Tatokhin E.A., Sychev I.V. Consideration of uncertainty in the diagnosis of complex systems. Vestnik of Voronezh Institute of the Ministry of Interior of Russia. 2025;(3):64–71. (In Russ.).

8. Bondarenko A.S., Zaytsev K.S. Using container management systems to build distributed cloud information systems with microservice architecture. International Journal of Open Information Technologies. 2023;11(8):17–23. (In Russ.).

9. Dzheiranian A.D., Lyadova L.N. Integrating an ontology-driven approach to data visualization and AI-based visualization with Plotly. Proceedings of the Institute for System Programming of the RAS. 2025;37(4-2):191–206.

10. Ustimov M.G., Prokhorova O.K., Zalozhnykh D.O. Adaptive risk-based management of retail network facilities based on clusterization and training with reinforcements. Modeling, Optimization and Information Technology. 2025;13(4). (In Russ.). https://doi.org/10.26102/2310-6018/2025.51.4.066