Artificial Intelligence Supported Agricultural Decision Support Systems: The Farming of the Future

Abstract

This study proposes an artificial intelligence (AI) based Agricultural Decision Support System (ADSS) that addresses the lack of data, uncertainties caused by climate change, and the inadequate transfer of scientific knowledge to the field in agricultural production in Turkiye. While digital agriculture technologies generate large volumes of data, most of these data are used solely for monitoring and visualization and fail to deliver actionable, personalized recommendations to farmers. The proposed system compares sensor data and farmer declarations with optimum agronomic parameters defined in the literature to generate meaningful and applicable recommendations on irrigation scheduling, fertilizer management, disease risk, and planting time. Machine learning algorithms support crop recommendation, yield prediction, and risk forecasting, enabling more accurate strategic, tactical, and operational decisions. Reference data are managed exclusively by experts, safeguarding scientific accuracy; regional climatic averages, kc coefficients (crop coefficient), and phenological stage information are consolidated in a centralized database. The system also supports sensorless usage scenarios, making it accessible to small-scale enterprises with limited technology. Image processing and deep learning achieve 97–99% accuracy in disease diagnosis, while IoT sensors detect water stress to enable more efficient autonomous irrigation. The applicability of the system depends on regular and accurate data entry by farmers and on the transparency and explainability of the AI-based decision engine. Data privacy and ethical use are critical for the system's sustainability. Overall, the proposed AI supported agricultural decision support system represents an innovative digital transformation model that aggregates Turkiye's agricultural data on an integrated platform and provides farmers with personalized, applicable guidance shifting agriculture from experience-driven to data-driven, predictable, and sustainable management.

Keywords: Artificial Intelligence, Agricultural Decision Support System, Digital Agriculture