Machines. Technologies. Materials.
Vol. 20 (2026), Issue 1, pg(s) 14-17

TECHNOLOGIES

Forecast-oriented optimization of production plans: closed-loop decision-making and model risk management

  • 1 Department of Information and Communication Technologies – The Federal State Autonomous Educational Institution of Higher Education "National Research Technological University "MISIS", Moscow, Russia

Abstract

: This paper presents a forecast-oriented optimization framework for production and technological planning in modern metallurgical manufacturing systems. Forecasts of demand, raw material availability, and external constraints are widely used in the development of production plans; however, in traditional optimization models, forecast information is typically treated as a fixed input parameter, which leads to model risk and reduced robustness of decision-making. Within the proposed framework, forecasts are interpreted as dynamic and potentially inaccurate elements of the control system, and planning is performed in a closed-loop feedback structure integrating forecasting, model risk management, and optimization. The paper describes the architecture of an integrated forecast-oriented planning system and reports the results of computational experiments on the problem of forming a technological production plan for secondary aluminum alloys. The results demonstrate the impact of forecast accuracy on product cost and the computational complexity of the optimization process.

Keywords

References

  1. Ben-Tal A., Nemirovski A. Robust optimization – methodology and applications // Mathematical Programming. 2002. Vol. 92, no. 3. P. 453–480. DOI:10.1007/s101070100286.
  2. Bertsimas D., Sim M. The price of robustness // Operations Research. 2004. Vol. 52, no. 1. P. 35–53. DOI:10.1287/opre.1030.0065.
  3. Bertsimas D., Sim M. Robust discrete optimization and network flows // Mathematical Programming. 2003. Vol. 98, no. 1–3. P. 49–71. DOI:10.1007/s10107-003-0396-4.
  4. Bertsimas D., Thiele A. A robust optimization approach to inventory theory // Operations Research. 2006. Vol. 54, no. 1. P. 150–168. DOI:10.1287/opre.1050.0238.
  5. Bertsimas D., Gupta V., Kallus N. Data-driven robust optimization // Mathematical Programming. 2018. Vol. 167, no. 2. P. 235–292. DOI:10.1007/s10107-017-1125-8.
  6. Sadana U., Bampis C., Gupta V. A survey of contextual optimization methods for decision-making under uncertainty // European Journal of Operational Research. 2024. Article 110369. DOI:10.1016/j.ejor.2024.03.020.
  7. Elmachtoub A. N., Grigas P. Smart “Predict, then Optimize” // Management Science. 2021. Vol. 67, no. 2. P. 1029–1052. DOI:10.1287/mnsc.2020.3922.
  8. Zhang W., Huang X., Xu C. Closed-loop process optimization in smart manufacturing: A systematic review and future trends // Computers in Industry. 2024. Vol. 150. Article 104131. DOI:10.1016/j.compind.2024.104131.
  9. Schwenzer M., Kirsch A., Hölzle W. Review on model predictive control: an engineering perspective // The International Journal of Advanced Manufacturing Technology. 2021. Vol. 114, no. 5–6. P. 1701–1724. DOI:10.1007/s00170-021-07682-3.
  10. Wu J., Tang Z., Liao T. Approximated dynamic programming for production and inventory planning problem in cold rolling process of steel production // Mathematics. 2024. Vol. 12, no. 24. Article 3922. DOI:10.3390/math12243922.

Article full text

Download PDF