Ceglarek wins NSF Early Career Development award

Dimensional variation has been a major issue with many U.S. manufacturing industries. High levels of variation can cause expensive product and process design changes after the design stage. It also can lead to long ramp-up time during new product launch, especially for automotive and aerospace assembly lines, and low production yield below design intent expectations. Since some manufacturing- and design-induced variation is inevitable, it is important to have methods for identifying root causes of dimensional variation, as well as a thorough understanding of the sources of variation during a new product launch.

With his $400,000 award, Assistant Professor Darek Ceglarek hopes to develop a generic multistage assembly process model, or stream of variation (SOVA) system, which potentially can be applied widely across discrete manufacturing industries, such as automotive, aerospace, shipbuilding and appliance manufacturing, where product dimensional integrity is of crucial importance. SOVA will take into account key product- and process-control characteristics with varying resolution that can be used during design, launch and full production phases in complex assembly processes . Based on a generic computer-aided design and manufacturing system integrated with statistical analysis to predict early-design-phase process performance, companies can apply the model during a manufacturing system's design, launch and full production phases. The project aims to integrate CAD and CAM models with statistical analysis to predict manufacturing-process dimensional variability during design as it relates to compliant parts. In addition, it will facilitate math-based design and manufacturing by synthesizing product and process variables to expand the classic "part interchangeability" concept into "process interchangeability," and train industrial engineers in statistical methods and their application in CAD and CAM systems.

By integrating research with education, Ceglarek will develop the graduate courses, "Information-Based Design and Manufacturing," and "Reconfigurable/Reusable Manufacturing Systems," and one undergraduate course, "Product and Process Variability Reduction." He also will partner with students and science faculty in Madison-area high schools to develop grade-specific projects and problems by visualizing various geometric phenomena that occur in industrial settings.

Among other efforts, Ceglarek will organize exchange programs and long-distance courses with Berlin University of Technology, Hong Kong University of Science and Technology, and Chalmers University of Technology in Sweden.