Farmers typically select crops based on either agronomic suitability or market trends, rarely both together, which can lead to growing a well-suited but comparatively low-profit crop. This paper presents a four-stage system that first predicts crop suitability from soil nutrients (N, P, K), pH, soil moisture, temperature, humidity and rainfall using a Random Forest classifier trained on a public 22-crop dataset (99.32% held-out accuracy), then forecasts market price for every agronomically suitable candidate crop using separate per-crop regression models trained on historical Government of India Agmarknet mandi price records, then computes expected monthly profit (yield × price − cultivation cost, normalized by crop growth duration) rather than ranking candidates by price alone as in prior work, and finally surfaces a forecastreliability risk label and an interactive what-if scenario tool. A case study demonstrates the system recommending cotton over the more agronomically suitable maize (72.7% vs. 27.0% suitability) due to a substantially higher expected monthly profit, illustrating the central contribution: that suitability-only and price-only recommendation strategies can diverge from the profit-maximizing choice, and that explicitly modeling cultivation cost and crop duration closes a gap left open in recent hybrid suitability-price literature.

Khan Ayesha Muhammed Irfan, Dr. Manoj Singh, "An Integrated Decision-Support Framework for Crop Selection Using Suitability Prediction, Market Price Forecasting for Profit Optimization" International Journal of Scientific Research and Engineering Development, V9(5): Page(1292-1297) September - October 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.