<![CDATA[Children with Autism Spectrum Disorder (ASD) require early intervention during their developmental stages. Currently, the availability of experts capable of accurately classifying the independence levels of children with ASD remains limited. Determining these independence levels is crucial, as it serves as the basis for establishing appropriate early interventions. The system aims to assist specialists in conducting more consistent and efficient assessments. This study contributes a novel application of a forward chaining–based expert system for classifying ASD children’s independence levels, integrating rule-based reasoning with user-centered evaluation, which distinguishes it from previous studies that primarily focus on diagnosis rather than functional independence assessment. Data were collected from three institutions: two Public Special Need Schools and a Regional Technical Implementation Unit of Children with Special Needs in East Java. The dataset consists of 400 records encompassing five daily activities: eating, drinking, brushing teeth, dressing, and taking off clothes. The independence levels are classified into three categories: independent, partially independent, and dependent. This research consists of seven stages, namely data collection, rule based system using forward chaining, database design using CDM and PDM, user interface development, implementation of the Next.js framework system and PostgreSQL database, system testing, and system evaluation. The results of the study showed that the accuracy was 98.5% and the user satisfaction score was 80.85%. These results indicate that the proposed method is effective in supporting therapists in determining the level of independence of children with ASD based on rules established by experts.]]>
