<![CDATA[Objective: The inefficiency of the manual medical check-up process at the Rumah Sakit Mata Padang Eye Center, characterized by paper-based recording, long waiting times, inconsistent symptom documentation, and heavy reliance on limited ophthalmologists, motivated this study. The objective was to develop a web-based expert system that enables patients to perform preliminary self-diagnosis of common eye diseases through interactive yes/no symptom consultation while integrating the forward chaining method with patient data management to support early detection and reduce specialist workload. Method: This study adopted a Research and Development (R&D) approach using the sequential linear (waterfall) model. Data were collected through field observations, interviews with ophthalmologists and staff, questionnaires, and a literature review at Padang Eye Center. The system was implemented using PHP and MySQL, incorporating decision trees, Data Flow Diagrams, and production rules with a Certainty Factor. Results: Functional testing on 50 test cases revealed that the prototype achieved 92% diagnostic accuracy, 94% precision, and an average processing time of 0.82 seconds, representing an 85% reduction in consultation time compared to the manual process. The system successfully generated consistent diagnoses with treatment and prevention recommendations for conditions such as eyelid edema, blepharitis, and trichiasis. Novelty: The novelty of this study lies in the seamless integration of forward chaining inference, symptom weighting, and modular web interfaces specifically designed for real clinical workflows in an Indonesian eye hospital, features rarely combined with previous standalone prototypes. This study provides both practical improvements in service efficiency and theoretical contributions to accessible, explainable rule-based expert systems in ophthalmology.]]>
