CNC Cutting Tools` Life Prediction Using Data Mining Approach

The failure of CNC machine tools has always been a negative impact on the manufacturing environment. The consequences of the failure will influence the production control, which further increases the duration of unplanned maintenance. To avoid such situations, it is required to predict the tools’ behaviours based on the raw data collected from machines. Hence, the objective of this paper is to obtain the machine tool life using the machining parameters including cutting speed, feed rate, and depth of cut which may affect the tool life in the prediction. All the data is collected by using different types of machine tools material against different types of workpieces. In this paper, classification is chosen to be the data mining approach with two algorithms to build the model for prediction, which are linear regression and multilayer perceptron. The data collected was being split into training and testing data. There are 40% of the data used for training data to build the predictive models while 60% of the data collected is used as testing data. The result of predicted tool life is then validated with the Taylor’s Extended Tool Life equation according to the ISO standard 3685 and ISO 8688-2. The results show that our proposed method is on par with the tool life predicted by Taylor’s method.