Developing Predictive Engineering Analytics to Formulate the Closed-Loop Management for Achieving Re-Industrialisation

Abstract

Regarding the process of printed circuit board assembly (PCBA), existing failure location methods are reactive in nature, while process parameters and performance cannot be predicted to achieve a high level of operational excellence. Designated PCB designs are not customized for specific manufacturing sites, while process performance becomes uncertain to clients and manufacturers. In this paper, an intelligent manufacturing performance predictive framework (IMPPF) is proposed in this paper, which structures the predictive engineering analytics for the smart manufacturing. First, the data collection from the PCBA process is structured by means of multi-responses Taguchi method, which guarantees the data reliability and quality. Second, the artificial neural network is adopted to learn from the existing operational data so as to provide the prediction on machine settings and process performance at the Gerber drawing stage. The contribution of this study is mainly to establish a closed-loop framework to facilitate the predictive engineering analytics for achieving re-industrialization.