Intelligent Discovery And Response Method Of Workshop Abnormal Production-events In Big Data Environment

Cloud manufacturing and industrial big data and other emerging technologies brings new opportunities for manufacturing enterprises,but in the meantime it also brings new challenges to the update and upgrading of manufacturing execution system(Manufacturing Execution System,MES).As the internal process of manufacturing execution system,job shop scheduling is the main linking MES with external system.job shop scheduling optimization is the technical core of MES's advancement.Firstly,to solve the issues of job shop scheduling in MES,this paper based on the design thought of software architecture optimization analyzes the key points of architecture.We design a complete architecture that is from static scheduling to real-time monitoring and active perception of manufacturing execution,then to intelligent response of abnormal production-events,and finally,to dynamic scheduling,given the global analysis of scheduling optimization problem,which ensures the high fit between theoretical study and practical application.Secondly,for solving two key issues,i.e.discovery of abnormal production-events,intelligent processing of abnormal production-events,we analyze them with the expansion of large data environment,to create a production system with abnormal discovery and processing capabilities by using the forecasting analysis capability of data mining.Aiming at the discovery of production abnormal problem,after taking time series and causality into account,build a multi-decision tree workshop event forecasting model based on the time series,which ensures the accuracy and reliability of forecast.Direct at the intelligent processing of abnormal production-events problem,we propose asolution that is referred to the process from manufacturing status assessment to early diagnosis of abnormity,then to infer future failure time and proactive maintenance.The cutting tool component of key equipment is monitored and quantified,building a multi-neural network decision model based on the time series.Achieve the goal of multilayered life prediction about cutting tool life,which provide a way of thinking for the prognostic and health management in MES.Finally,taking an electrical machinery plant as a case,and combining IEC/ISO 62264 standard,data mining,and the cloud computing method consisting of virtualization,servitization and SOA together,we develop an integrated job shop scheduling optimization system of single and small batch MES,and the aforementioned theory and method are validated.