A methodology and modeling for robust parameter control using observable noise factors

Robust parameter design (RPD) has been widely used as the primary technique to reduce process and product variability. But this offline methodology only can be used in process without further on-line adjustments. With the development of sensing technology, high power computing and signal processing, observable noise factors (ONF) are available in real-time so that the online parameters can be set according to variation of ONE So we can obtain a superior online control performance for reducing process variation compared with the traditional RPD. This paper aims to develop a new methodology and modeling which is called robust parameter control (RPC) to integrate the offline RPD approach and the online automatic process control (APC) theory. Some research in terms of this new regression model based on design of experiment (DOE) is also given. The DOE theory is to be expanded to obtain a regression model for automatic process control purpose and application of RPD approach. An example of a leaf spring experiment helps to illustrate the new methodology.