Analysing Rice Productivity in India using a Fuzzy Logic Model based on Soil Quality Parameters

This study integrates fuzzy logic modelling to evaluate the nonlinear interactions of soil quality parameters (i.e., soil pH, organic matter, moisture, and nutrient levels) on rice productivity across diverse agro-climatic regions of India, offering improved accuracy over conventional yield prediction models. Membership functions are developed using empirical field data and expert knowledge. The fuzzy inference system, including fuzzification and rule bases, is implemented using MATLAB. Fuzzy surface analyses identify optimal productivity ranges, particularly at pH levels of 5.0–7.5 and organic matter between 0.5% and 2.5%, supporting the application of sustainable soil management practices. Additional surface analyses involving interactions (such as soil pH–moisture and nutrient–moisture) confirm the model's validity, with predicted yields aligning closely with actual agricultural data. Sensitivity analysis indicates that soil pH and nutrient levels have the greatest impact on rice yields, while organic matter has more influence when considered in combination with other factors. The model identifies critical soil factors and provides practical guidance for applying sustainable interventions (such as organic fertilization, efficient irrigation, and crop rotation). It serves as a decision-support tool for policymakers, agricultural planners, and farmers, enabling data-driven, ecologically sound strategies to improve rice productivity and support long-term food security.