Open AccessApplication of Machine Learning Method to Predict Reliability in Lubricating Oil System Components
Author(s) -
Nurvita Arumsari,
Feby Agung Pamuji,
Bianda Devi Puspita
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1175/1/012012
Subject(s) - support vector machine , reliability (semiconductor) , computer science , series (stratigraphy) , regression , distribution (mathematics) , artificial intelligence , component (thermodynamics) , data mining , machine learning , pattern recognition (psychology) , mathematics , statistics , power (physics) , physics , thermodynamics , paleontology , mathematical analysis , quantum mechanics , biology
In predicting the reliability and failure of components, classical methods are often used by determining the distribution between failure times. Sometimes, the determination of this distribution does not always match the data pattern that is owned because of the limited data records and the many types of distribution that must be chosen. In addition, how much influence the time series has on components cannot be clearly analyzed. Therefore, in this study a prediction will be carried out by combining classical methods with machine learning, namely Support Vector Regression (SVR) and Least Square Support Vector Regression (LSSVR). Both methods are considered capable of improving the prediction accuracy of a series of data. The results showed that the classical combined method with LSSVR had better accuracy than SVR.