A Two-Stage Decision-Support Framework for Crop Recommendation and Harvest-Price Forecasting: A Case Study of Sehore District, Madhya Pradesh, India

This study presented a two-stage framework that links unsupervised crop-suitability assessment with harvest-price forecasting for Sehore district. The clustering stage produced moderately separated groups (silhouette 0.4885 for the K-means and GMM ensemble), and the stacked ensemble of Random Forest, Extra Trees, XGBoost, and Ridge regressors improved on a Random Forest baseline (RMSE 307.99 versus 360.91; R2 0.967 versus 0.947) on a small test set. Both stages are available through a web prototype. The work is a proof of concept. Future work will (i) enlarge the price and soil datasets and evaluate with time-ordered cross-validation against persistence and sequence-model baselines; (ii) validate the suitability stage against yield records and expert assessment; (iii) add prediction intervals and evaluate the profit and risk layer with cost-of-cultivation data; and (iv) test the interface with farmers and extension officers, including local-language support.

        1.         Sehore District Administration. Agriculture. https://sehore.nic.in/?p=24754         2.         ICAR–Central Research Institute for Dryland Agriculture. Agriculture Contingency Plan for District: Sehore, Madhya Pradesh. 2012. https://icar-crida.res.in/CP/madhya%20pradesh/MP16-Sehore-26.6.2012.pdf         3.         Pande SM, Ramesh PK, Anmol A, Aishwarya BR, Rohilla K, Shaurya K. Crop recommender system using machine learning approach. In: 2021 5th International Conference on Computing Methodologies and Communication (ICCMC). Erode, India: IEEE; 2021. https://doi.org/10.1109/ICCMC51019.2021.9418351         4.         Gosai D, Raval C, Nayak R, Jayswal H, Patel A. Crop recommendation system using machine learning. Int J Sci Res Comput Sci Eng Inf Technol. 2021;7(3):554–557. https://doi.org/10.32628/CSEIT2173129         5.         Shastri S, Kumar S, Mansotra V, Salgotra R. Advancing crop recommendation system with supervised machine learning and explainable artificial intelligence. Sci Rep. 2025;15:25498. https://doi.org/10.1038/s41598-025-07003-8         6.         Manogna RL, Dharmaji V, Sarang S. Enhancing agricultural commodity price forecasting with deep learning. Sci Rep. 2025. https://doi.org/10.1038/s41598-025-05103-z         7.         Nayak GHH, Alam MW, Singh KN, Avinash G, Kumar RR, Ray M, Deb CK. Exogenous variable driven deep learning models for improved price forecasting of TOP crops in India. Sci Rep. 2024;14. https://doi.org/10.1038/s41598-024-68040-3         8.         Özden C, Bulut M. Spectral temporal graph neural network for multivariate agricultural price forecasting. Cienc Rural. 2024;54(1). https://doi.org/10.1590/0103-8478cr20220677         9.         Sindhur NM, Pavithra C, Muchikel N. A hybrid machine learning framework for optimizing crop selection via agronomic and economic forecasting. arXiv:2507.08832. 2025. https://arxiv.org/abs/2507.08832        10.       Agricultural Marketing Information Network (Agmarknet), Government of India. https://agmarknet.gov.in