An Android-Based Expert System for Smartphone Damage Diagnosis Using the Forward Chaining Method

Abstract

The increasing use of Android-based smartphones has been accompanied by a growing need for repair services to address various hardware and software problems. At Alfa Cell Randan Batu, the diagnosis process is still conducted manually based on technicians’ experience and information provided by customers. This process requires considerable time, is less efficient, and may result in inconsistent diagnoses. This study aimed to design and develop an expert system to assist in the initial diagnosis of Android smartphone damage in a faster, more efficient, and structured manner, while also supporting the reduction of customer queues. The study employed the Prototype method, consisting of communication, quick planning, quick design modeling, Prototype construction, and deployment and feedback. Data were collected through observation, interviews with Alfa Cell technicians, and literature studies. The system was developed as a web-based application using PHP, MySQL, XAMPP, and Bootstrap, with the Forward chaining method as the inference engine. The knowledge base consisted of 37 symptoms, 13 damage categories, 13 recommended solutions, and 13 diagnostic rules using IF–THEN structures with the OR operator. The evaluation results showed that all 13 diagnostic cases produced results consistent with technician validation. Black-box testing demonstrated that all tested system functions operated successfully. User testing involving 30 respondents obtained an overall mean score of 4.41, equivalent to 88.10%, categorized as strongly agree. In conclusion, the developed Forward chaining-based expert system can assist in the initial diagnosis of Android smartphone damage in a more structured and consistent manner, support faster diagnosis, and provide recommended solutions based on selected symptoms. The results should still be confirmed by a technician through direct device examination.