Author:
Hari Nugraha Nurjaman, Dwi Dinariana, Fitry Suryani, Danil Rahman, Prijasambada, Suwito, Siti Sujatini, Dimas Putra Anugerah, Ana Maria Tey Seran

Doi: 10.26480/jtin.02.2025.39.47

This is an open access article distributed under the Creative Commons Attribution License CC BY 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited

Precast and prestressed technology have been the main solution for Indonesia’s infrastructure since 1962, with many innovations coming from the Indonesian experts. However, traditional learning methods are often ineffective due to limited access. The Expert System introduced in the same year is a systemic method for compiling and transmitting technical knowledge. Meanwhile, Artificial Intelligence (AI), which has developed since the 1950s, presents interactive learning tools. OpenAI applications like ChatGPT are too generic and not suitable for specific contexts such as Indonesian construction. The Expert Intelligence System is here as an effort to combine Expert Systems and AI, starting from Indonesian precast and prestressed technology. The goal is to develop an Expert Intelligent System based on Indonesian Development Standards and national innovations, with more accurate results in a local context than OpenAI applications. The research began with the preparation of an Expert System database that was validated through scientific publications. This data is integrated into the AI Retrieval-Augmented Generation (RAG) model to generate contextual answers. The model is tested through the learning process and compared to open AI in terms of precision and clarity. The result is an Expert Intelligence System web application for precast and prestressed technology in Indonesia. Evaluations demonstrate excellence in understanding the local context and can be continuously improved through the latest scientific publications. The system is focused on Indonesian data and Standard Nasional Indonesia (SNI) standards, and is prioritized for academics, stakeholders, and policymakers. This limitation actually strengthens the relevance and validity of the answers given.