Abstract
Incorporating human factors engineering guidelines early in design has the potential to reduce the cost and product lead-time to market. Also, products that go through strict ergonomics assessments are associated with better comfort and safety ratings. However, designers are often caught in the dilemma of what prototyping method to use when assessing product ergonomics early in design. This is especially problematic during the conceptual design phase before the physical prototypes are available or built. In this research, we explore the computational prototyping dilemma for early design ergonomics assessments from both fidelity and human-product interaction perspectives. In this paper, three computational prototypes with different fidelity levels (low, medium-, and high-fidelity) are compared in their adequacy for evaluating designs that comprise low- to high-levels of human-product interactions. We used three computational prototyping strategies: (1) Method #1 is a low-fidelity methodology based a digital sketchpad tool; (2) Method #2 is a medium-fidelity methodology consisted of computer-aided design and digital human modeling; and, (3) Method #3 is a high-fidelity methodology composed of computer-aided design, digital human modeling, and surrogate modeling. In order to perform computational ergonomics analyses using above approach, we selected a generic wall mounted cabinet design and a simplified Boeing 767 cockpit model as case studies to illustrate designs that require low- and high-levels of human-product interactions. Our preliminary results show that low-, medium- and high-level prototyping strategies produce similar ergonomics outcomes when evaluating low-level human-computer interaction (e.g., cabinet model). On the other hand, both low- and medium-fidelity (Method #1 and Method #2) prototyping strategies are limited in terms of providing detailed information about human performance when compared to high-fidelity prototyping (Method #3) in evaluating designs with high-level human-computer interaction (e.g., cockpit model).
