Development of an Expert System for Diagnosis of Pests and Diseases in Soybean Plants Using the Forward Chaining Method

This research focuses on developing an expert system to detect pests and diseases affecting soybean plants (Glycine max), which often reduce yield. The system employs forward chaining with the best-first search decision-making algorithm, which was developed using the waterfall methodology. Data utilized includes comprehensive information on symptoms, types of pests, diseases, and their respective management solutions gathered through case studies and expert interviews. Users of the system can input observed symptoms in soybean plants, and the system provides diagnoses and treatment recommendations based on established knowledge rules. Feasibility testing of the system was conducted using the TAM approach to assess technology acceptance among users and BlackBox Testing to ensure system reliability from a technical perspective. Test results indicate that the expert system is viable, achieving a feasibility rate of 83.7% based on TAM criteria and 100% across eight modules using BlackBox Testing, demonstrating significant potential in effectively supporting the diagnosis and management of pests and diseases in soybean plants.