Modern automotive technology in premium vehicles, particularly the BMW 3-Series produced between 1990 and 2006, contains interdependent mechanical components, sensors, actuators, and electronic control units. These relationships make fault diagnosis challenging because different failures can produce similar symptoms. At CV. Ryuga Spareparts, initial fault identification depends substantially on the experience and memory of technicians. This condition can create variation in diagnostic decisions, extend troubleshooting time, and make junior technicians dependent on senior personnel. This study develops a web-based expert system to support the initial diagnosis of BMW 3-Series faults using Forward Chaining and Certainty Factor. Research activities include knowledge acquisition, symptom and fault modeling, production-rule formulation, web application development, and functional testing. The knowledge base contains ten symptom indicators, ten fault classifications, and ten IF–THEN rules. The inference engine treats selected symptoms as initial facts and evaluates rule premises to infer a possible fault. Certainty Factor weighting represents confidence associated with user-observed symptoms and expert knowledge. The application is implemented using Laravel and PHP with a MySQL relational database following an MVC architecture. Black Box Testing was conducted on six core modules: authentication, symptom management, fault management, rule configuration, diagnosis, and report export. The source study reports that all tested functions passed, giving 100% functional validity for the tested scenarios. This result indicates functional conformity, not a measured 100% mechanical diagnostic accuracy. The application can support a more systematic preliminary diagnosis and preserve workshop knowledge in a structured form.
Copyrights © 2026