<p>Machine breakdowns in the production line mostly finish in more than 18<br />minutes, since the machine that needs repair more time is done on the<br />production line, not in the machine warehouse. Historical machine<br />breakdown data is digitally recorded through the Andon system, but it is still<br />not being used adequately to aid decision-making. This research introduces<br />an analysis of historical machine breakdown data to provide predictions of<br />repair time intervals with a focus on finding the best algorithm accuracy.<br />The research method uses machine learning techniques with a classification<br />model. There are five algorithms used: logistic regression (LR), naive bayes<br />(NB), k-nearest neighbor (KNN), support vector machine (SVM), and<br />random forest (RF). The results of this study prove that historical machine<br />breakdown data can be optimized to predict machine repair time intervals in<br />the production line. The accuracy of LR algorithm is slightly better than the<br />other algorithms. Based on the receiver operating characteristic–area under<br />curve (ROC-AUC) performance evaluation metric, the quality value of the<br />accuracy of LR model is satisfied with a percentage of 69% with a<br />difference of 0.5% between the train and test data.</p>