Artificial intelligence

Abstract: At the current stage, modern methods of mathematical modeling are used to build forecasting models, with machine learning and artificial neural network technologies playing a leading role. Climate factors play a crucial role in fluctuations in crop yields. Machine learning methods, including linear and nonlinear regression models, were used to analyze their impact on wheat yields. The main goal of the study was to compare linear and nonlinear regression models with different numbers of parameters. Given the complex changes in economic mechanisms that have taken place in Ukrainian agriculture over the past thirty years, we come to the realization that statistical research can only be conducted on data after 2000. The study is based on the data on average monthly temperature and precipitation during the wheat growing season (April-June) for 2000-2021. The data were grouped by agroclimatic zones, which allows for the consideration of natural and climatic features of the regions, as well as for improving the accuracy of modeling. The results of the study confirm that taking into account the nonlinear influence of climate factors, such as the squares of variables and their products, significantly improves the accuracy of yield forecasting. Nonlinear models demonstrated almost twice the efficiency of linear models, as evidenced by the values of the coefficient of determination (R²). The increase in the number of parameters in the models also had a positive impact on their quality, although the main role was played by nonlinear relationships. The obtained models allow forecasting wheat yields with a three-month horizon, which ensures their practical value for the agricultural sector. The proposed approach can be adapted for other crops and used in different agroclimatic zones, contributing to the efficiency of management decisions in the face of climate change.

Keywords: crop yield, climate factors, machine learning.

References:

1. M. Ashfaq, I. Khan, A. Alzahrani, M. U. Tariq, H. Khan and A. Ghani. (2024) Accurate Wheat Yield Prediction Using Machine Learning and Climate-NDVI Data Fusion. IEEE Access, Vol. 12, pp. 40947-40961. [Online]. Available: https://doi.org/10.1109/ACCESS.2024.3376735.
2. D. Muller, A. Jungandreas, F. Koch, F. Shirhorn. (2016) The impact of climate change on wheat production in Ukraine. Report on agricultural policy (APD).
3. Petro Hrytsiuk, Maksym Havryliuk. (2024) Modeling of the nonlinear impact of climatic factors on wheat yield using machine learning techniques. 19th International Conference ICTERI-2024, Lviv, 16-21 September 2024.
4. Ali, M.A.; Hassan, M.; Mehmood, M.; Kazmi, D.H.; Chishtie, F.A.; Shahid, I. (2022) The Potential Impact of Climate Extremes on Cotton and Wheat Crops in Southern Punjab, Pakistan. Sustainability, Vol. 14, p. 1609. https://doi.org/10.3390/su14031609.
5. Agatha Popescu, Toma Adrian Dinu, Elena Stoian and Valentin Serban. (2023) Climate Change and its Impact on Wheat, Maize and Sunflower Yield in Romania in the Period 2017-2021. Scientific Papers Series Management, Economic Engineering in Agriculture and Rural Development, Vol. 23(1).
6. Özden C., Karadoğan N. (2024) Wheat Yield Prediction for Turkey Using Statistical Machine Learning and Deep Learning Methods. PAKISTAN JOURNAL OF AGRICULTURAL SCIENCES, Vol. 61(1), pp.1-7.
7. Fiorentini, M., Schillaci, C., Denora, M., Zenobi, S., Deligios, P., Orsini, R., Santilocchi, R., Perniola, M., Montanarella, L., & Ledda, L. (2024). A machine learning modeling framework for Triticum turgidum subsp. durum Desf. yield forecasting in Italy. Agronomy Journal, Vol. 116, pp. 1050–1070. [Online]. Available: https://doi.org/10.1002/agj2.21279.
8. Navid Mahdizadeh Gharakhanlou, Liliana Perez. (2024) From data to harvest: Leveraging ensemble machine learning for enhanced crop yield predictions across Canada amidst climate change. Science of The Total Environment. Vol. 951, p. 175764. [Online]. Available: https://doi.org/10.1016/j.scitotenv.2024.175764.
9. McKinney, W. (2018) Python for Data Analysis. O’Reilly Media.
10. Schierhorn, F., Hofmann, M., Gagalyuk, T. et al. (2021) Machine learning reveals complex effects of climatic means and weather extremes on wheat yields during different plant developmental stages. Climatic Change, Vol. 169(39). [Online]. Available: https://doi.org/10.1007/s10584-021-03272-0.
11. P. Hrytsiuk, T. Babych, S. Baranovsky, M. Havryliuk. (2023) Assessing of Climate Impact on Wheat Yield using Machine Learning Techniques. CEUR Workshop Proceedings, 3513, pp. 314–329.
12. Iqbal N, Shahzad MU, Sherif E-SM, Tariq MU, Rashid J, Le T-V, Ghani A. (2024) Analysis of Wheat-Yield Prediction Using Machine Learning Models under Climate Change Scenarios. Sustainability. Vol. 16(16), p. 6976. [Online]. Available: https://doi.org/10.3390/su16166976.