Heavy Equipment Industry have various business counterparts, including mining industries, infrastructure contractors, and as well as any kind of manufactures. Currently companies in the similar business are working hard on how to optimize the maintenance activities on their heavy equipment. Maintenance of those equipment could be very crucial to the business continuity. This paper provides an alternative to optimize such an activity through an approach called condition-based maintenance. We conducted our research in one international heavy equipment rental company based in Singapore and has a branch in Indonesia. The company's core business is on heavy equipment rental including Excavator. The research focused on utilizing data generated by sensors attached to the Excavator with the main aim is to predict the Remaining Useful Life (RUL) of Oil Grease Pump which is a crucial component of the Excavator. We used some machine learning techniques such as Linear Regression, Decision Tree Regression, and Random Forest methodology to build models to predict the RUL. The results from each models were compared each other to gain a deeper insight on the predictive ability of each model using the data provided. It turns out that the linear regression model gives the highest predictive accuracy with 61% of RMSE.
展开▼
