Predictive Modeling of Rice Production in South America Using Machine Learning
Resumo
Research Context: Rice is a staple food for nearly half of the global population and is highly vulnerable to climate variability. In South America, accurate forecasting models are critical to ensure food security and guide agricultural decision-making. Scientific and/or Practical Problem: Traditional methods often fail to capture the complexity of agricultural systems shaped by environmental and socioeconomic factors. The lack of predictive systems adapted to South America restricts the ability of governments and producers to anticipate risks. This research addresses both a scientific challenge, by advancing predictive modeling, and a practical challenge, by supporting sustainable agricultural planning. Proposed Solution and/or Analysis: The study integrates agricultural production data from FAOSTAT (2004–2023) with climate variables from NASA POWER (2011–2023). It compares Linear Regression, Support Vector Regression (SVR), and XGBoost Regressor to identify the most reliable configuration for rice production forecasting. Related IS Theory: The work aligns with decision support systems (DSS) and information systems for sustainability, contributing to the Brazilian Grand Challenges in Information Systems (GranDSI-BR 2016–2026). Research Method: The approach combined exploratory analysis, training with multiple time windows (5, 10, 15 years), and cross-validation, using Mean Absolute Percentage Error (MAPE) as the main evaluation metric. Summary of Results: Longer windows (10 years) improved stability. XGBoost achieved the lowest MAPE and greater robustness, while climate data integration reduced errors in atypical years. Contributions and Impact to IS area: The study demonstrates the potential of machine learning–based predictive systems as components of agricultural information systems, reinforcing their role in enhancing food security, supporting public policies, and promoting sustainable resource management in South America and beyond.
Referências
Bostrom, R. P. and Heinen, J. S. (1977). Mis problems and failures: A socio-technical perspective. part i: The causes. MIS quarterly, pages 17–32.
Chen, T. and Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 785–794.
Daft, R. L. and Lengel, R. H. (1986). Organizational information requirements, media richness and structural design. Management science, 32(5):554–571.
Dias, L. G., Hacke, A., dos Santos Souza, E., Nath, S., Canesin, M. R., Vilella, O. V., Geloneze, B., Pallone, J. A. L., Cazarin, C. B. B., Blakeslee, J. J., Mariutti, L. R. B., and Bragagnolo, N. (2022). Comparison of chemical and nutritional compositions between aromatic and non-aromatic rice from brazil and effect of planting time on bioactive compounds. Journal of Food Composition and Analysis, 111:104608.
FAO (2025). Yield forecasting with catboost: Fao pilot project on machine learning for agriculture. Technical report, Food and Agriculture Organization of the United Nations, Rome.
FAOSTAT (2025). Crops and livestock products. [link]. Accessed: 2025-09-25.
GranDSI-BR (2016). Grandes desafios de pesquisa em sistemas de informação no brasil (2016–2026). Technical report, SBC – Sociedade Brasileira de Computação.
Hamim, A. A. I., Mohaimin, A., Ishmum, A. F., Ifty, R. A., and Patwary, M. J. A. (2025). Advances in machine learning for crop yield prediction: A comprehensive review of techniques, trends, and challenges. pages 1–6.
Hyndman, R. J. and Koehler, A. B. (2006). Another look at measures of forecast accuracy. International Journal of Forecasting, 22(4):679–688.
Khan, I. and Khare, B. K. (2025). Integrating ai in plant science: A review of applications and future prospects. Plant Gene, page 100542.
Kumar, A., Kumar, A., Singh, R. P., Kumar, P., and Singh, P. V. (2024). A comparative study of detrending methods on crop yield time series for drought studies. Asian Research Journal of Agriculture, 17(3):191–204.
Küçüktopcu, E., Cemek, E., Cemek, B., and Simsek, H. (2023). Hybrid statistical and machine learning methods for daily evapotranspiration modeling. Sustainability, 15(7).
Li, S., Wu, F., Zhou, Q., and Zhang, Y. (2024). Adopting agronomic strategies to enhance the adaptation of global rice production to future climate change: a meta-analysis. Agronomy for Sustainable Development, 44(3).
Mishra, S., Mishra, D., and Santra, G. H. (2016). Applications of machine learning techniques in agricultural crop production: A review paper. Indian Journal of Science and Technology, 9(38).
Montgomery, D. C., Peck, E. A., and Vining, G. G. (2012). Introduction to Linear Regression Analysis. Wiley, New Jersey, 5th edition.
Neog, B. J. (2022). Temperature shocks and rural labour markets: evidence from india. Climatic Change, 171(16).
Noorunnahar, M., Chowdhury, A. H., and Mila, F. A. (2023). A tree based extreme gradient boosting (xgboost) machine learning model to forecast the annual rice production in bangladesh. PLOS ONE, 18(3):e0283452.
Parry, M. A. J. and Hawkesford, M. J. (2010). Food security: increasing yield and improving resource use efficiency. Proceedings of The Nutrition Society, 69(4):592–600.
Ray, D. K., Mueller, N. D., West, P. C., and Foley, J. A. (2013). Yield trends are insufficient to double global crop production by 2050. PLOS ONE, 8(6):1–8.
Rodrigues, G. C. and Braga, R. P. (2021). Evaluation of nasa power reanalysis products to estimate daily weather variables in a hot summer mediterranean climate. Agronomy, 11(6):1207.
Samal, P., Babu, S. C., Mondal, B., and Mishra, S. N. (2022). The global rice agriculture towards 2050: An inter-continental perspective. Outlook on Agriculture, 51(2):164–172.
Satpathi, A., Setiya, P., Das, B., Nain, A. S., Jha, P. K., Singh, S., and Singh, S. (2023). Comparative analysis of statistical and machine learning techniques for rice yield forecasting for chhattisgarh, india. Sustainability, 15(3):2786.
Smola, A. J. and Schölkopf, B. (2004). A tutorial on support vector regression. Statistics and Computing, 14(3):199–222.
Stackhouse, P. W. J., Whitlock, C. H., Chandler, W. S., Hoell, J. M., Zhang, T., Westberg, D., Hoell, A., Macpherson, B., and Mikovitz, J. C. (2018). Power release: Nasa prediction of worldwide energy resources. Technical report, NASA.
Turban, E., Sharda, R., and Delen, D. (2010). Decision Support and Business Intelligence Systems. Prentice Hall Press, USA, 9th edition.
Ujjwal, T. R., Murty, B. N. S., Lenka, N. K., and Lal, S. B. (2023). Comparative analysis of statistical and machine learning techniques for rice yield forecasting for chhattisgarh, india. Sustainability, 15(3):2786.
USDA (2025). Production, supply and distribution (psd) online. [link]. Accessed: 2025-09-25.
Watson, R. T., Boudreau, M.-C., and Chen, A. J. (2010). Information systems and environmentally sustainable development: energy informatics and new directions for the is community. MIS quarterly, pages 23–38.
Zhao, T., Chen, G., Suraphee, S., Phoophiwfa, T., and Busababodhin, P. (2025). A hybrid tcn-xgboost model for agricultural product market price forecasting. PLOS ONE, 20.
