Linear Regression Modeling to Predict the Lifespan of Final Drive Components in Heavy Equipment

This research is motivated by the importance of optimal performance of heavy equipment in various industrial sectors that are highly dependent on the reliability of each component, especially in the context of maintenance planning and parts replacement. Problems arise when there is a mismatch between the expected life of the component and the actual life, which has the potential to cause a decrease in operational efficiency and an increase in unexpected costs. This research identifies a gap between the expectation and reality of component life that has not been widely studied empirically. The research methodology involved linear regression analysis to evaluate the relationship between expected component life and actual life. The results show that there is a strong relationship between the two variables with a coefficient of determination of 0.8605, however, there is a mean difference of 7892.79 which indicates that actual life tends to be shorter than expected. In addition, the significant variation in the difference, with a standard deviation of 4043.82, indicates a discrepancy that requires further analysis. The implication of this research is the need to validate component life predictions to improve accuracy and efficiency in machine management, and to encourage the development of more comprehensive prediction models.