<![CDATA[Machine breakdowns in the production line mostly finish in more than 18minutes, since the machine that needs repair more time is done on theproduction line, not in the machine warehouse. Historical machinebreakdown data is digitally recorded through the Andon system, but it is stillnot being used adequately to aid decision-making. This research introducesan analysis of historical machine breakdown data to provide predictions ofrepair time intervals with a focus on finding the best algorithm accuracy.The research method uses machine learning techniques with a classificationmodel. There are five algorithms used: logistic regression (LR), naive bayes(NB), k-nearest neighbor (KNN), support vector machine (SVM), andrandom forest (RF). The results of this study prove that historical machinebreakdown data can be optimized to predict machine repair time intervals inthe production line. The accuracy of LR algorithm is slightly better than theother algorithms. Based on the receiver operating characteristic–area undercurve (ROC-AUC) performance evaluation metric, the quality value of theaccuracy of LR model is satisfied with a percentage of 69% with adifference of 0.5% between the train and test data.]]>
Copyrights © 2023
