Web-Based Expert System for Computer Troubleshooting Diagnosis Using the Forward Chaining Method

Abstract

Computer fault diagnosis at SMKN 2 Padang Panjang is currently performed manually and relies on a single technician, resulting in lengthy identification and repair processes. Furthermore, the lack of organized records regarding computer fault history hinders the tracking of failure patterns in specific units. This study aims to develop a web-based expert system utilizing the Forward Chaining method to assist in computer troubleshooting and diagnosis based on user-reported symptoms. The research employs the Research and Development (R&D) approach with the Rapid Application Development (RAD) model, comprising the stages of Requirements Planning, User Design, Construction, and Cutover. A knowledge base (rule base) is automatically generated from training data using the C4.5 Decision Tree algorithm and subsequently processed via a Forward Chaining inference engine. The system incorporates QR Code features for unique computer identification and a diagnosis history log. System validation involved Black Box Testing, White Box Testing, expert evaluation, and user testing. Black box and white box testing results confirmed that all system functions operated according to the design. Expert evaluation yielded an average score of 95.8% (categorized as highly valid), while testing with 25 users demonstrated a 96% alignment between the system's diagnoses and those of experts. These results demonstrate that the developed web-based expert system is effective, valid, and suitable for use as an initial diagnostic tool for computer troubleshooting in a school laboratory environment.