Data Mining Application in Process Control of Smart Material Manufacturing

Abstract

According to Industry 4.0 concept, huge amounts of data is generated in Smart Material Manufacturing and this data needs to be collated, stored, organised and analysed in order to develop a more efficient manufacturing system. This study focuses on the prediction of smart material manufacturing process, based on the current production data. It presents a route of knowledge gain about predicting future manufacturing systems, using data mining. The model proposed for the actual manufacturing process is made to acquire the necessary data for process control. Implementing particular methods of data mining, and by altering the input parameters, we can predict the behaviour of the manufacturing processes. This prediction is then verified by the use of a simulation model. After analysing various methods, the method using neural networks is chosen for deployment of the latest data in the concluding phases. This research aims at designing and verifying the tools for mining data for supporting system control in manufacturing. It aims at improvement of the process of decision making. The practical control strategies can be accurately modified, depending on the predictions made and the targeted results of production. These strategies can then be used in real time manufacturing, without a chance of failure.