Troubleshooting Generator Sets using Expert System

Abstract

PT. Zaman Teknindo (PT. ZT) is a company engaged in Mechanical and Engineering field which is registered as one PT. Telkomsel vendors. The problems that occur at PT. ZT, if the power outage and generator set (generator) does not start automatically. The corrective team on duty at that time will go to the field and find a solution to the problem. With a lack of knowledge from the corrective team, they need help from the mechanical team. The mechanical team is an external team of PT. ZT. To bring a mechanical team requires an enormous cost and a relatively long time needed to get to the location. Based on the problem above, this study proposes a forward chaining expert system that is by depth-first search using the certainty factor method. To prove whether a fact is certain or not, it must be in the metric form in generator troubleshooting. The research methodology used the Software Development Life Cycle (SDLC) starting from problem identification, analysis, design, coding, testing and maintenance. This system is web-based, so users can easily access and choose symptoms of the damage. With this system makes it easy for PT. ZT especially the corrective team in the field can easily find out the damage symptoms without having to meet with experts directly.