Applications of Experimental Design in Engineering and Food Science

In collaboration with an engineering firm in Santa Barbara County, the department of Statistics at California Polytechnic sought to establish aesthetic performance tracking, control, and capability of a new product line. The new product line required a finish painting operation with stringent customer requirements for coating and aesthetic performance. Through data collection and monitoring, paint operation parameters would be optimized in order to reduce the amount of production parts that do not meet guidelines along the production line. Process optimization should ultimately lead to decreased production costs by reducing the operation's cost of quality. The proposed team would be comprised of one statistics student, five materials engineers and two Cal Poly faculty members; one from each department. While this project is ongoing, a portion of the project was completed and is reported in this document. Additionally, participants of this collaboration are under a nondisclosure agreement and proprietary information may not be revealed or released to anyone not directly involved in the project. Background: The proposed project is broken into six sub projects which include measurement instrumentation system analysis and process quality control (gage study and paint design of experiment), witness panel sampling, film thickness determination, paint process mapping and inspection gate establishment, paint process characterization and finally paint process monitoring and control. Results from the initial stages of this larger project are reported here, including measurement instrumentation system analysis and process quality control, including a gage study and paint process design of experiment to characterize sources of variability for paint thickness and aesthetic performance. The painting process is characterized by measurements of the final painted surfaces. Automated measurements of film thickness, gloss and color are made using several instruments. In order to rely on any results from our study, the instruments used to make measurements must be tested to ensure they are capable of first, measuring at the specification tolerances that the production parts are held to, and second, able to reproduce the measurement. To achieve this task we performed a Gage R&R Study (repeatability and reproducibility). Repeatability refers to the variation in measurements taken by a single person or instrument on the same item under the same conditions while reproducibility refers to the variation induced when different operators, instruments, or laboratories measure the same item. In this study we looked at two instruments, across three appraisers, 24 parts, 3 measurements per part and a total of 20 different measurements …