APPLICATION OF MACHINE LEARNING METHODS FOR FORECASTING AND RESOURCE MANAGEMENT BASED ON INTELLIGENT TIME SERIES ANALYSIS

This study focuses on the application of machine learning methods and statistical time series modeling to historical data on natural gas production in Kazakhstan (2000-2024) in order to build a reliable predictive model. The study considers and compares ARIMA, Holt-Winters, linear regression with lag variables, Random Forest, and gradient boosting models. The accuracy of these models was evaluated using standard metrics: MAE, RMSE, and the coefficient of determination R². The comparison results showed that the Holt–Winters exponential smoothing method provides the highest forecast accuracy among all the approaches tested. This model was chosen to generate the forecast of natural gas production volumes for 2025-2027. According to the forecast, a further gradual increase in natural gas production is expected in 2025-2027 while maintaining the identified trend and seasonal patterns. The results obtained demonstrate the effectiveness of integrating modern machine learning algorithms with classical time series analysis methods when working with historical statistical data. The practical significance of this work lies in the fact that the developed forecasting model can contribute to more substantiated strategic planning in the gas industry and improved efficiency of resource management.

1. Time Series Analysis: Forecasting and Control / G.E.P. Box et al. – 5th ed. – Hoboken: John Wiley & Sons, 2015. – 712 p. https://doi.org/10.1111/jtsa.12194/.

2. Hyndman R.J. Forecasting: Principles and Practice / R.J. Hyndman, G. Athanasopoulos. – 2nd ed. – Melbourne: OTexts, 2018. – 384 p.

3. Experimental Study on Damage Fracture Law of Coal from Solid-Propellant Blasting / H. Chu et al // Energies. – 2022. – Vol. 15(21). – Р. 8104. https://doi.org/10.3390/en15218104.

4. Forecasting long-term world annual natural gas production by machine learning / D. Sen et al // Resources Policy. – 2023. – Vol. 80. – Р. 103224. https://doi.org/10.1016/j.resourpol.2022.103224.

5. Integrated Data-Driven Framework for Forecasting Tight Gas Production / F. Yao et al // Processes. – 2025. – Vol. 13(4). – Р. 1162. https://doi.org/10.3390/pr13041162.

6. Synergizing Machine Learning and Physical Models for Enhanced Gas Production Forecasting: A Comparative Study of Short- and Long-Term Feasibility / BK Raoof et al // Energies. – 2025. – № 18(5). – Р. 1187. https://doi.org/10.3390/en18051187.

7. A Study on a Novel Production Forecasting Method of Unconventional Oil and Gas Wells Based on Adaptive Fusion / D. Hou et al // Processes. – 2024. – Vol.12(11). – Р. 2515. https://doi.org/10.3390/pr12112515.

8. A Natural Gas Consumption Forecasting System for Continual Learning Scenarios / S. Radek et al // arXiv preprint. 2023. arXiv:2309.03720. URL: https://arxiv.org/abs/2309.03720.

9. Tussupova K. Optimal Allocation of Resources in an Open Economic System with Cobb–Douglas Production and Trade Balances / K. Tussupova, Z. Murzabekov // Economies. – 2025. – № 13(7). – Р. 184. https://doi.org/10.3390/economies13070184.

10. Murzabekov Z.N. Development of a model of efficient resource allocation in an open threesector economy for balanced growth / Z.N. Murzabekov, K.B. Tussupova // Journal of Mathematics, Mechanics and Computer Science. – 2025. – Vol. 124(4). – Р. 59-70. https://doi.org/10.26577/JMMCS2024-v124-i4-a5.

11. Tusupova K.B. Razrabotka algoritma dlya chislennogo resheniya zadachi optimal'nogo upravleniya resursami v tryohsektornoj ekonomike // VESTNIK AUES. – 2024. – № 67(4). – S. 172-184. https://doi.org/10.51775/2790-0886_2024_67_4_172.