面向制造装配过程的大数据分析与调度优化研究

Big Data Analysis for Manufacturing Assembly Processes and Scheduling Optimization Research

针对复杂产品的装配服务执行决策问题,分析了装配大数据和装配调度优化的关键技术,并提出了一个大数据分析和调度优化模型。通过引入局部线性嵌入、自适应Boosting、支持向量机和D-S证据理论方法对装配大数据进行分析,输出不确定信息的决策。与此同时,提出了一种基于生理大数据的人为误差预测方法来探索装配任务和内外部环境对装配效率的影响。最后,通过采用装配任务的可变度量聚类方法,以保证装配效率和生产平衡的最大化。

In order to solve the problem of assembly service execution decision of complex products,the key techniques of assembly big data and assembly scheduling optimization are analyzed,and a big data analysis and scheduling optimization model is proposed.By introducing local linear embedding,adaptive Boosting,support vector machine and D-S evidence theory method,the assembly big data is analyzed,and the decision of uncertain information is output.At the same time,a human error prediction method based on physiological big data is proposed to explore the impact of assembly tasks and internal and external environments on assembly efficiency.Finally,the variable metric clustering method of the assembly task is adopted to ensure the assembly efficiency and the production balance are maximized.