Método para análise de performance de linhas produtivas baseado no modelo Digital Twin alimentado por dados em tempo-quase-real

Abstract

Reality shows that despite promises to facilitate the analysis of manufacturing systems, using state-of-the-art tools and techniques can become a challenging endeavor. Using data can generate meaningful analytics to help companies understand underlying issues and plan actions to improve processes. Technological advances help to improve yields, operations, decision making and cost reduction when adopted in your daily processes. Digital Twins (DT) are models that can be used to monitor production parameters, possibly perform cause and effect investigations and analyze the performance of production lines, as well as project future events, once fed with reliable and real-time data. Assembly lines are constantly monitoring their procedures, but the detection of adversities that may occur is still considered complex. There is no formula to be followed for the implementation of the DT with information about data, software, integration between the physical and virtual environments and subsequent analysis. In an attempt to shed light on the recurring issues present in the daily lives of those who work in the implementation of Industry 4.0 projects in production lines, this work presents a method for developing the DT powered by near-real-time data, with insights obtained from the application case in the automobile industry. It also proposes a minimal structure, necessary for capturing, reading and sending data as well as subsequent analysis and interpretation of the results obtained. It presents user insights from different functional units within a given company, to build and test the model developed. The results obtained through this study demonstrate that when the steps of the proposed method are followed, the analysis of yields and performance of production lines as well as production follow-ups become more accessible and favorable. The fact that the data is ingested in near-real time implements the DT a great competitive advantage for companies from different fields that wish to use this technology. Learning and absorbing information in near-real time from line events help decision-making and favor predictive actions. Another point noted is that companies must prepare for unexpected problems and limitations ranging from the inadequacy of legacy hardware to obstacles related to human behavior in real-life implementation projects.