An auxiliary process controller was designed, implemented, and validated for on-line process and quality optimization. The objective function included terms related to the process variation, model uncertainty, and control energy. The controller architecture relied on characterized models including both process transfer functions and principal components analysis to perform online optimization in parallel with the physical molding process. New process and quality observations were input to the controller to update the models and provided new settings for the machine controller. Experimentation included characterization with a D-optimal design of experiments followed by a validation to measure the controller's performance with respect to controller stability, extrinsic material variation, cycle time reduction, and other common manufacturing goals. In every case, the controller was able to reduce the value of the objective function while also improving the part dimensions relative to tight tolerance specifications. While characterization experiments could be costly, the use of the resulting process models greatly speeds convergence and facilitates the consideration of various cost and quality terms in the objective function. POLYM. ENG. SCI., 55:2743-2750, 2015

Manufacturing process automation is often impeded by limitations related to automatic quality assurance. Many plastics manufacturers use univariate statistical process control (SPC) for quality control by charting the critical process states relative to defined control limits. Alternatively, principal component analysis (PCA) and projection to latent stuctures (PLS) are multivariate methods that measure the process variance by the distance to the model (DModX) and the Hotelling t-squared (T2) values. A methodology for robust model development is described to perturb the manufacturing process for process characterization based on a design of experiments; best subset analysis is used to provide an optimal set of regressors for univariate SPC. Four different statistical models were derived from the same data set for a highly instrumented injection molding process. The performance of these models was then assessed with respect to fault diagnosis and defect identification when the molding process was subjected to twelve common process faults. Across two hundred molding cycles, the univariate SPC models correctly diagnosed five of the twelve process faults with one false positive, detecting only eighteen of twenty four defective products while indicating two false positives. With the same molding cycles, PCA and PLS provided nearly identical performance by correctly diagnosing ten of the twelve process faults and detecting twenty three of the twenty four defective products; PCA indicated two false positives while PLS indicated only one false positive.