Variability detection and robustness design in complex production system

This paper will study the detection of variability and the techniques of robust design in complex production system consisted of many processes with close correlativity. First, the background of variability detection in production system is introduced, and then a weighted network corresponding to correlation matrix of all processes is set up. Based on the network and clustering criterion of maximum spanning tree, a classification of all processes is derived. Further, the variation of each process in a class is determined by residual analysis. Finally, the methods of robust design that can be effectively conducted for the processes with large variability are discussed.