In the past 10 years the area of Design for Manufacturing (DfM) has come under intense investigation. A large number of methods and tools have been developed for manufacturability evaluation in various domains. DfM today spans a vast spectrum, from simple handwritten scorecards and check sheets to sophisticated Knowledge-based Design advisory systems. Most of the research in this area has focused on building automated tools which are often based on ad hoc methods and measures of manufacturability. In fact, it is difficult to make a clear distinction between DfM metric, DfM method (for search, evaluation, or optimization), and DfM tool (embodiment of method) when reviewing reports and software.
Our work is motivated by a desire to put DfM on solid theoretical foundations. We evaluated measures of manufacturability and classes of DfM methods and frameworks independent of the specific manufacturing processes. Criteria used in evaluation include theoretical foundation, accuracy, flexibility in choosing utility/objective function, domain independence, ease of use, level and extent of information required, computational cost, ability to incorporate uncertainty and market factors. An implicit assumption in DfM studies seems to be that one DfM metric fits all. However, product development objectives vary a great deal with product type, batch size, discrete or continuous manufacturing, and market conditions. Also, the amount of information available and uncertainty in design variables at different phases of the design process varies. It is not possible to use the same metric in all situations.

The project produced a normative DfM system. On the ASU side there were three components:

• Parametric Design Shell (Design Rating System)
• Manufacturability evaluation system (DfM Testbed)
• Theoretical basis for Cost-Benefit Analysis