Decision Support System for predicting crop yield using ARIMA, SARIMA and fuzzy logic models

This study introduces an innovative approach for accurate crop yield prediction, integrating fuzzy logic and multi-parameter analysis. In contrast to conventional methods relying solely on temperature and rainfall, this research incorporates pH and humidity parameters to capture a holistic view of crop growth influencers. By leveraging fuzzy logic, it effectively handles data uncertainty and intricacies, unveiling intricate parameter-crop relationships. To bolster predictions, the model employs ARIMA and SARIMA models for rainfall and temperature, respectively. These time series forecasting techniques enhance accuracy by considering historical trends. The comprehensive dataset spanning multiple seasons ensures robust training and validation. Results highlight the model’s superiority over traditional methods, showcasing its potential in aiding agricultural decisions. This research pioneers more reliable crop yield prediction by incorporating a wider range of factors and using advanced forecasting, enabling better agricultural planning for farmers and policy makers.