Corresponding Author: K. L. Wood, wood@mail.utexas.edu, (512) 471-0095. Submitted to the 1997 ASME DTM Conference. Functional Analysis: A Fundamental Empirical Study for Reverse Engineering, Benchmarking, and Redesign Aaron D. Little, Graduate Research Assistant and Kristin L. Wood, Associate Professor Department of Mechanical Engineering The University of Texas ETC 5.160 Austin, TX 78712-1063 Abstract At a basic level, product design composes three primary tasks: specification development, conceptual and configuration design, and product refinement. Customer needs analysis, product benchmarking, and business case analysis are fundamental to specification development, whereas conceptual design and product refinement entail functional analysis, the generation of solution principles and product geometry, concept selection, mathematical modeling, prototyping, and Taguchi analysis for variability. In this paper, we focus on an advanced method for functional analysis, a critical component of this process. This method demonstrates clear ties to customer needs, and is based on an empirical study of approximately 60 household consumer products. As part of this study, a common vocabulary for product functions and flows is developed and applied to the consumer products. House of Quality results are then used to correlate customer importance to the product functions. Data from these correlations provide a basis for determining critical functions and flows across all products and within important product domains, such as material processors and beverage brewers. From this study, we discuss how the results can be applied to product testing and benchmarking, design by analogy, the identification of functional groups and dependencies, and design education. 1. Introduction: Background and Motivation Many research, educational, and industrial applications require a basic understanding of functionality. The concept of transforming function to form has been discussed and improved widely among researchers through conceptual design methods (Paul and Beitz, 1984; Ullman, 1992; Ulrich and Eppinger, 1995; Cross, 1994; Hubka et al., 1988; Wood and Otto, 1997); however, the link between customer needs and functional analysis has been unclear. Current methods for linking customer needs, such as “task listing” (Otto and Wood, 1996), simply assure that the needs are expressed in the function structure. This approach maps customer needs to material, energy, and signal flows to generate sub-function sequences; however, the finished functional analysis shows little or no preference from one sub-function to the next. This deficiency results because design methodologies, such as Paul and Beitz’s and Ullman’s, require