Real-time manufacturing modeling and simulation framework using augmented reality and stochastic network analysis

Abstract

While the development of augmented reality (AR) technologies has made it possible to assign real-time features to many systems and applications, these trends are rare in manufacturing modeling and simulation. This research study proposes a real-time manufacturing layout modeler and material flow simulator. The manufacturing devices of interest are positioned using AR labels, and the generated layout is converted into a stochastic Petri net model, where the validity of material flow and other criteria are checked. In order to overcome the limitations of the Petri net model and enhance analytical functionalities, stochastic network analyses are embedded into the framework. The layout model with greater uncertainty is analyzed, and manufacturing performance indicators such as cycle time, throughput, and work-in process are estimated. The proposed framework is not simply an integration of AR techniques and manufacturing simulations, but provides an efficient AR labeling architecture for large-scale manufacturing environments, and is suitable for a fast, real-time rendering. In order to verify the effectiveness of the proposed framework, real-time modeling and simulation examples were used as case studies. The results showed that the proposed system contributes to more accurate layout design and simulation analysis by using the embedded AR techniques and queuing network methods.