Sistem Pakar Deteksi Penyakit Tanaman Cabai Menggunakan Metode Forward Chaining

Abstract

Chili plants are a vital horticultural commodity in Indonesia, yet their productivity is frequently hindered by disease outbreaks that are difficult for novice farmers to identify. Delays in diagnosing symptoms often lead to improper treatment and significant crop losses. This study aims to develop an expert system for detecting chili plant diseases that provides pesticide recommendations based on observed symptoms. The system is built using the Forward Chaining method for reasoning from symptomatic facts to disease conclusions. The software development follows the Rapid Application Development (RAD) methodology, implemented with the PHP programming language and Laravel framework. Research data were gathered through literature reviews and interviews with plant experts. Black box testing results indicate that all system functions—including symptom data management, disease records, and rule sets—operate as designed. System validation, conducted by comparing results with diagnoses from three experts using the Fleiss' Kappa test, yielded a value of 0.464, placing the agreement level in the "Moderate" category. Consequently, this expert system is a viable tool for farmers to perform early identification of chili plant diseases and accurately determine the appropriate pesticides independently.