IMPLEMENTASI SISTEM CERDAS BERBASIS RULE-BASED UNTUK ESTIMASI RENCANA ANGGARAN BIAYA (RAB) INSTALASI LISTRIK BERBASIS WEB

Abstract

Across building construction projects, estimating the cost of electrical installations has traditionally leaned on manual, spreadsheet-driven calculation—an approach prone to human error, slow to finish, and awkward to update whenever project specifications shift. This work reports the design and development of a web-based intelligent system that applies a rule-based inference mechanism to automate Rencana Anggaran Biaya (RAB) estimation for electrical installations in small- to medium-scale buildings. Development followed the Waterfall model, pairing PHP on the server side with MySQL as the relational database and HTML, CSS, and JavaScript on the client side. From the parameters a user enters—the number of light points, socket outlets, and switches, the total cable length, and the applicable service rate—the rule-based engine automatically derives both the required material volumes and the overall cost. Functional verification through black-box testing over ten scenarios showed every core feature operating as intended. A side-by-side comparison further indicates that estimation time drops from roughly 45-60 minutes by hand to under 2 minutes, while the risk of arithmetic mistakes is effectively removed.