<![CDATA[The diagnostic process for laptop damage in computer service centers often requires considerable time due to the similarity of symptoms across different types of damage and the reliance on technicians’ experiential knowledge, while studies that specifically compare the performance of Forward Chaining and Backward Chaining in web-based expert systems within local service contexts remain limited. This study aims to evaluate the effectiveness of these two inference methods in an expert system for detecting laptop damage at Warrior Computer Mendalo Jambi. A quantitative approach with a case study and system testing design was employed, involving 30 real laptop damage cases as the test sample. Data were collected through observation of the service process, technician interviews for rule-base construction, and case documentation, then analyzed by comparing the diagnostic accuracy of the system with the technicians’ diagnoses. The results indicate that the Forward Chaining method achieved an accuracy of 86.70% (26 out of 30 cases matched), whereas Backward Chaining achieved an accuracy of 80.00% (24 out of 30 cases matched). These findings extend the understanding of rule-based reasoning applications in laptop damage diagnosis and demonstrate that Forward Chaining is more suitable for symptom-based input, while Backward Chaining is more effective for verifying damage hypotheses. The study concludes that integrating both methods in an expert system is crucial to enhance the flexibility of the diagnostic process and support more efficient technician services in computer repair centers.]]>
Copyrights © 2026
