<![CDATA[The development of transportation, especially two-wheeled motorized vehicles, drives the increasing demand for maintenance services and the availability of spare parts. However, many workshops still face challenges in managing spare parts stock, which is handled manually. This study aims to design and develop a spare parts recommendation system at Robbi Motor Workshop using the Apriori Algorithm, as well as to test the performance of the developed system. The method used is data mining with association techniques, where the Apriori Algorithm is applied to discover spare parts purchasing patterns from transaction data. The system enables users to analyze transactions based on a selected time range without the need to manually input minimum support and confidence values. The results show that the system is capable of generating relevant association rules, such as: “If consumers buy Engine Oil, then consumers will also buy Axle Oil”, with a support value of 67% and a confidence value of 86%. In addition, the system’s accuracy was tested using the lift value against two recommendation rules: (1) Engine Oil → Axle Oil with a lift value of 0.9949, and (2) Inner Tire → Axle Oil with a lift value of 1.0714. A lift value > 1 indicates that the combination of items has a stronger association than random occurrence. The system is implemented as a web-based application using the Laravel framework, equipped with features for transaction data management, Apriori analysis, analysis history, and exporting analysis results to PDF format. Testing using the blackbox method shows that the system operates according to specifications and produces accurate outputs. With this recommendation system, it is expected that the workshop can improve the efficiency of spare parts stock management.]]>
Copyrights © 2025
