This thesis presents an automated method for assessing conceptual designs with respect to manufacturing and supply chain, using geometric data mining and machine learning algorithms. It is important for designers to understand how design decisions will impact downstream manufacturing and sourcing. Many critical decisions are made during conceptual design that impact production cost even before detailed design is finalized; however, the effects of these decisions are not known until later. Design for manufacturing and design for supply chain are methods that provide feedback to the user in a way that enables proactive design changes.

A conceptual design is largely defined by the geometry found in CAD files. In this work, feature-free geometric algorithms were used to extract meaningful manufacturability metrics from 3D models, which were classified as either castings or machined parts. The developed metrics serve as useful attributes for a machine learning model that can help select the manufacturing process of a conceptual design. A classification accuracy of 86% was achieved using a random forest algorithm, which is comparable to other approaches in the literature, while only using geometry as input. The work in this thesis provides methods for using geometry to evaluate a design for manufacturability and supply chain, enabling proactive design decisions early during new product development.