COMBINING PROCESS KNOWLEDGE FOR CONTINUOUS QUALITY IMPROVEMENT

Quality improvement efforts often make use of various mathematical mod els that describe the relationships between quality characteristics an d process factors. Such models typically come from a variety of source s: experiments, theory, on-line data analysis, expertise, and other pr ocess documents. These sources of knowledge are often distinct and sep arate, often yielding models with slightly different predictions, havi ng different precision and validity. In this paper we explore alternat ives in which different mathematical models can be integrated together into a single prediction that takes into account both model validity and model variability. Some guidelines for establishing and quantifyin g model validity are presented. The approach is demonstrated within th e context of predicting surface finish in a machining process.