Enhancing the Prediction of Escalator Failure Using Logistic Regression over Random Forest

In the specific paper's context, the main aim is to use the logistic regression data set to look into the forecast of the failed escalator through the random forest machine learning algorithm. Analyzing the sources of data, the set of data which is under discussion in this article was derived from data of 6 escalators. Flow: Logistic Regression Algorithms and Random Forest Algorithms are achieved by many networks and parameters and Failure Prediction is automated. Each group was given a sample of 5 which is given by the power calculation test used known as Gpower and the other setting parameters include: Homoscedasticity: The value of α equals 0. 05 and power=0. 85. was slightly lower as compared to Random Forest Algorithm which was 29. 67% The extent of accuracy of the achieved results proves improved efficiency of the applied 91% Logistic Regression algorithm. 40%. The results of the analysis showed that there is a statistically significant differences in the mean accuracy values estimated by Logistic Regression Algorithm and Random Forest Algorithms; the estimate of the p-value was equal to p=0. ,p> 0. 05) 063 (Independent Sample T Test This research shows the usefulness of the paper in understanding that the Logistic Regression algorithm can provide more efficient and effective results regarding the escalation of Prediction of Escalator Failure in an ITSM system.