Different industries utilize statistical prediction models that predict the product properties in process planning, control, and optimization. An important aim is to decrease the number of disqualifications. The model can prevent disqualifications efficiently if the disqualification probability is predicted accurately. This study gives step-by-step instructions for developing, validating, comparing, and visualizing models that predict the disqualification probability with high accuracy. The work summarizes industrially applicable statistical modeling methods that are most suitable for the development of accurate predictors for the disqualification probability. Currently, the information on such statistical methods, e.g. quantile regression, modeling of distribution shape, and joint modeling of mean and deviation, is scattered in the existing literature. The main contribution of this work is that it pulls together this methodology into a unified framework which allows the comparative analysis of probability predictors that are based on the different approaches. The proposed modeling procedure (ProPred) is demonstrated using three manufacturing industry applications. In the case applications, the predictors generated using the ProPred procedure are 10-30% more efficient in avoiding disqualifications by means of process planning and control operations than the baseline predictors.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.