Degree Type

Thesis

Date of Award

2019

Degree Name

Master of Science

Department

Mechanical Engineering

Major

Mechanical Engineering

First Advisor

Caroline Hayes

Abstract

The study presents two research questions: 1) Can stereoscopic 3-D visualization (like Virtual Reality) improve a design reviewer’s ability to detect errors when reviewing a product model over a traditional CAD system in which 3D product models are viewed on a flat 2D computer screen? 2) Can two collaborating individuals be more effective (faster time or detect more errors) during a product design review than one individual operating alone? The answer to the first question will help provide product design reviewers, managers, and software designers in product design firms decide what features to include, and whether it is worthwhile to purchase VR (Virtual Reality) tools to aid in design review, or whether more traditional CAD tools are just as effective. The second question will help scientists and managers understand whether collaboration really does increase success in product design reviews by increasing error identification or completing tasks faster than working individually. Design review is an important part of the design process: design reviewers aim to identify errors/defects as early as possible in order to save time and money. Various studies have been conducted to measure either user performance or user preference, or both, for VR versus traditional CAD interfaces in a variety of design tasks; however, there has been little quantitative research that has measured performance using VR versus CAD interfaces in a design review. The study used a low-cost desktop VR system as a test bed because of its potential to be used broadly in current work environments in any designer’s office without the need of a special room or large investment. An experiment was conducted to answer the research questions stated above. The investigation involved 16 participants who were asked to use four interfaces: CAD-view with individual interface, CAD-view with collaborative interface, VR-view with individual interface and VR-view with collaborative interface. They were asked to complete design review tasks employing each interface. The tasks involved four model-based design review problems in order to identify errors in product models. The results of the experiment do not provide evidence to support the first question. But, the results from the second question indicate that two collaborating individuals will complete a task faster than an individual working alone. The author believes that the current desktop VR system used in the investigation did not provide sufficient hardware and software support to effectively test the two research questions. There were several issues expressed by participants such as seeing dual images, experiencing eye irritation, and sensing a “jerky” phenomenon. In subsequent investigations it may be worthwhile to build the experimental test bed on a VR-system that will better test the research questions.

Copyright Owner

Mihir Radia

Language

en

File Format

application/pdf

File Size

127 pages

Share

COinS