The research objective of this project is to support and enhance engineering innovation through identifying the cognitive benefits of physical representations (such as prototypes) in the design process. The research on idea generation with physical models and on the use of prototypes in the design process is very limited. The benefits and limitations of physical representations to support the early phases of design will be identified. Two key hypotheses will be evaluated. The first is that physical models cause design fixation. The second is that physical models assist in overcoming inaccuracies in designers' mental models resulting in a larger number of functional, feasible ideas. These key research questions will be investigated through a combination of qualitative and quantitative studies of undergraduate mechanical and practicing engineers. Answering the key research question will provide a basis for design methods development for industry and engineering education.
If successful, this research will improve the methods and tools available to engineers thus increasing their ability to solve critical design problems. The results of this research will enhance the understanding of how physical representation effect engineering cognition, provide guidelines for their use within idea generation and develop novel design approaches to support innovation. Results of the research will provide new design principles and methods of use in industry and education. Novel design methods based on the fundamental cognitive and design principles defined in this project will be directly integrated into undergraduate and graduate design classes. The new methods for innovation will be applied in developing sustainable, appropriate technology while working with a community service group and workshops on teaching innovation in engineering education are planned. In addition, class modules for enhancing student innovation skills will be created.
This project investigates the role of physical representations and other representations in the engineering idea generation process. These studies explore the effects of physical models on designers’ mental models, design fixation and design fixation mitigation. These studies include controlled laboratory and triangulated qualitative studies. A set of studies addressing the research questions were completed. Studies ranged from highly controlled lab experiments to observational and interview data collected from practicing engineers. The first set of controlled and qualitative studies tested two hypotheses that were the main focus of this work: physical models supplemented designers’ erroneous mental models and physical models led to design fixation. The results from set of studies showed that physical models supplemented designers’ mental models, but failed to show design fixation, which occurs when a designer’s ideas are limited by their initial idea or by presented examples. The difference in results on design fixation for the current experiments as compared to prior studies is due to the Sunk Cost Effect. In the prior observational studies in literature, the design problems are complex and contain more components. When the designers solve design problems that require more resources in the form of time, effort, or money, they tend to fixate more. In the initial controlled studies from this grant, designers used minimum resources and hence fixation is not evident. In a follow-up experiment, a condition was added which required significantly more time to build the prototypes. The results provided support to the presence of Sunk Cost Effect, showing that design fixation was not inherent with physical modeling. The results confirm that the sunk cost of building plays a vital role in the observed fixation; thus, physical models do not inherently cause fixation. Moreover, results also show that physical models supplement designers’ erroneous mental models, leading to higher quality ideas. To minimize sunk costs very early in the design process, models should be built with materials requiring minimal time, cost, and effort for the designers. This project also had the opportunity to collect data on the impact of physical prototypes for a group of practicing engineers on a large project to reduce the cost of solar panel installation. It was found that a large amount of testing occurs with prototypes that are not articulated by the engineers. This was true for both practicing engineers and for graduate student industry sponsored class projects. Engineers were interviewed and asked about aspects of the design they were planning to test with each prototype. This result could be a reason that the transition to mostly virtual prototypes has presented challenges. A large amount of testing is either unarticulated or unintentional and thus left out when designers transition to only virtual prototypes. Another controlled experiment investigated how the design fixation of expert designers and novices differed in the idea generation process. The expert data from the study by Linsey, et al., 2010 were compared with the data collected from a controlled experiment with senior undergraduate students. The results from this study showed that both novices and experts fixated to the presented example. The experts generated more non-redundant ideas as their set of mental models is larger. At the same time, the defixation materials benefited the experts more than the novices. This study was also repeated with a physical model of the example solution given to the novices. The results showed that the physical model of the example fixated the students to the same extent as the pictorial example. To follow up on these result, further work explored the impact of different example representations on fixation. A series of three controlled experiments sought to further identify the impact of design representation on fixation. It was found that pictures, sketches and CAD models induce fixation to a similar degree, but that CAD models helped the designers to identify the effective design features more easily. It was also found that a hierarchical, tree-based functional model of a device did not induce a significant amount of design fixation when it was presented without a concrete example. While the building and testing of prototypes has significant risks for inducing fixation through the Sunk Cost Effecter, it also has large potential benefits. Novice engineers can overcome negative design fixation by having the opportunity to build and test their designs.
