Aiming at the problem that the current bottleneck identification method is challenging to respond to the discrete manufacturing process in a timely,dynamic,and accurate manner,combined with the graphical deduction modeling method,an event-data hybrid-driven resource fine-grained condition monitoring method(EDH)was proposed to identify dynamic bottlenecks in the discrete manufacturing process in real-time.Firstly,the manufacturing process's complex and changeable real-time information was obtained and processed through manufacturing IoT and complex event technology.On this basis,the state data of manufacturing resources were clustered in fine-grained using an improved data clustering algorithm(IFGCM).Finally,the bottleneck in the workshop was recognized in real-time by combining the resource fine-grained state time-series flow diagram model and the dynamic bottleneck perception method.This method's effectiveness is verified through the practical application in an elevator parts manufacturing workshop.
维普
万方数据