FAULT PREDICTION BY USING DATA MINING APPROACH
M.E. Student, Production & industrial engineering Department PEC University of technology, India
ABSTRACT
Data mining is an effective tool which can be used in decision making in the organization. In this research, data mining is applied in the JCB Ballabgarh plant for the prediction of faults in the parts of machinery manufactured at JCB. Data is collected and retrieved from the quality department of the JCB and then it is transformed into the format which can be used in data mining tool i.e. rattle. Many variables were added to the dataset and many steps were done to make data useful for data mining. Data mining algorithms were applied on the data which are decision tree, support vector machine (SVM), Artificial Neural Network (ANN). From the results it has been found that ANN showed very high accuracy in the fault prediction while decision tree was good at predicting but SVM showed poor performance in the fault prediction. It have been showed that data mining is an excellent tool for making prediction than the traditional statistical methods.
