Virtual versus physical prototypes : development and testing of a prototyping planning tool
This thesis documents the development of a heuristics-based decision-making tool to guide a designer’s choice between virtual or physical prototypes, based in part upon published prototyping strategies, as well as the design, implementation, and results of a pilot experimental study used to test this virtual-vs-physical decision-making tool for prototypes. Eighty undergraduate mechanical engineering students volunteered for a pilot experiment to test the decision-making tool. They were given the choice of physically or virtually prototyping a four-bar linkage. Forty participants in this pilot study were instructed to use a Likert-scale instrument to choose their prototyping technique, and an additional 40 participants, who did not use the instrument, served as a control group for evaluating the effectiveness of the instrument. Analysis of participants’ performance metrics undeniably shows that virtual prototyping is the optimal technique for this design problem, as virtual prototypers on average across both test groups achieved performance metrics almost five times higher in about half the time compared to physical prototypers. With the aid of a heuristics-based decision-making tool, 10% more participants in the experimental group picked the best technique versus those who did not use the tool in the control group (32 of 40, and 28 of 40, respectively). The prototyping choices of participants among each test group were analyzed using the comparison of two population proportions, and results from a two-tailed z-test yielded p = 0.303, thus the null hypothesis cannot be rejected with statistical significance for the test of two population proportions. Although the difference in choice of the optimal prototyping technique between test groups of this pilot study is not statistically significant, it serves as a preliminary model for a systematic approach that incorporates consideration of type of prototype as a strategic decision. Although the findings of this four-bar linkage study cannot be extrapolated to a generic prototyping process, this work provides a paradigm for thinking critically about virtual vs. physical prototyping decisions using a heuristics-based, structured prototyping strategy. The pilot results provide a template and motivation for conducting a larger scale experiment for generic prototyping applications.