MS Thesis Presentation by Jamal O. Wilson
Monday, August 8, 2005

(Dr. David Rosen, Chair)

"Selection for Rapid Manufacturing under Epistemic Uncertainty"

Abstract

Rapid Pro to typing (RP) is the process of building three-dimensional objects, in layers, using additive manufacturing. Rapid Manufacturing

(RM) is the use of RP technologies to manufacture end-use, or finished, products. At small lot sizes, such as with cus to mized products, traditional manufacturing technologies become infeasible due to the high costs of to oling and setup. RM offers the opportunity to produce these cus to mized products economically. Coupled with the cus to mization opportunities afforded by RM is a certain degree of uncertainty. This uncertainty is mainly attributed to the lack of information known about what the cus to mer's specific requirements and preferences are at the time of production.

In this thesis, I present an overall method for selection of a RM technology under the geometric uncertainty inherent to mass cus to mization. Specifically, I define the types of uncertainty inherent to RM (epistemic), propose a method to account for this uncertainty in a selection process (interval analysis), and propose a method to select a technology under uncertainty (Decision Theory under strict uncertainty). I illustrate this method with examples on the selection of an RM technology to produce cus to m caster wheels and cus to m hearing aid shells.

In order to perform selection, the performance of each alternative must be quantified. In the context of RM, the key attributes of concern are build time and part cost. In the current marketplace, the existing build time and part cost estimation software requires the use of a exact CAD models for analysis. In many cases, these CAD models may not be available. In this thesis, I develop and present parametric models for build time and cost estimation where only overall part geometry, such as part volume and bounding box, is needed. Since these models are parametric, they can be adapted for use across all the different RM technologies.