Date of Award

8-2011

Document Type

Thesis

Degree Name

Master of Science (MS)

Legacy Department

Mechanical Engineering

Committee Chair/Advisor

Mocko, Gregory

Committee Member

Summers , Joshua

Committee Member

Fadel , Georges

Abstract

Design changes are a frequent occurrence over the life of a product that may be initiated by an update to the product functionality, new customer needs, or generational improvements. The costs associated with these changes are undesirable, and are often times greatly inflated by additional, unanticipated changes that result from change propagating throughout the system. Propagation paths occur when an initiating change to a component necessitates subsequent changes to coupled components, as the change continues to propagate throughout the product architecture. The nature of this change propagation is challenging to characterize and accurately predict. To address this issue, a change prediction method is developed that builds upon current change management strategies. The method is comprised of: (1) a design structure matrix (DSM) to model the relationships and connectivity within a system, (2) coupling index (CI) values (ranging from 0 to 1) that assess the likeliness of a change to one component/feature affecting another, and (3) design for manufacturing (DFM) information to provide an estimate of the cost and impact of a change.
The method can either be applied at the component level, or through further decomposition, at the interfacing feature level. Modeling the relationships between interfacing features, as opposed to components, offers a more detailed representation of change, but requires more knowledge of the system that may not be available in the earlier stages of design. When evaluating a propagation path, the coupling index values are multiplied together as the path extends, to produce a decreased probability for higher orders of coupling. The proposed change prediction method is applied on three industry examples: BMW X5 headliner and center console assemblies, and a Ryobi drill assembly. The method is shown to produce viable results that allow for informed decisions during change management. These results show that the objective measures of coupling and manufacturing cost of change are effective approximations. A comparison of the results from the component and feature based methods show that a feature level analysis offers improvements in accuracy, and sensitivity to uncertainty and path representation. Furthermore, the method proves to be a valuable tool during the initial design of a product, as it can be used to identify features, interfaces, and manufacturing types that will lower a product's overall ease of change.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.