Using support vector machines for recognizing shifts in correlated manufacturing processes

Traditional statistical process control (SPC) techniques of control charting are not applicable in many process industries due to the fact that data from these facilities are auto-correlated. Several attempts have been made in the literature to extend traditional SPC techniques to deal with auto-correlated parameters. However, these extensions pose several serious limitations. This paper demonstrates that support vector machines can be extremely effective in minimizing both type-I errors and type-II errors in these auto-correlated processes