Supporting Novice Engineers in Idea Generation using Design Heuristics

How can beginning engineers learn to generate a variety of candidate concepts to consider? Because they likely have little experience with idea generation, training in specific techniques may be especially useful. Design Heuristics are an evidence-based tool developed to help engineers expand diversity of ideas considered during the front-end phase of design. Brainstorming is a common ideation technique in both engineering education and practice. Our research sought to compare the characteristics of ideas generated by incoming engineering freshmen using both techniques. The analyses of this study involve qualitative patterns in a subset of data collected from 94 incoming engineering freshmen. We explored diversity, fixation, and the scope of the concepts generated. Ideas were coded based on a variety tree coding scheme, students’ selfperceptions, and system vs. component scope. We found that concepts generated using Design Heuristics were more likely to focus on the methods for achieving functions and on specific components, while Brainstormed ideas tended towards more holistic systems. The results suggest that alternative tools for idea generation may serve to focus attention on different qualities of design, and that multiple tools may be important for design in engineering education. Introduction Engineering design has been described as a problem-solving process that “searches through a hypothetical space of many possible ideas.”[1, 2] A design process can be viewed as having a frontend and a back-end, where the front-end design is a “fuzzy” process full of ambiguity[3] that involves a series of divergent and convergent thinking episodes in order to come to a conceptual design[4]. The front-end phase includes defining problems, conducting research and design ethnography, identifying design requirements, idea generation, and initial prototyping[3, 4, 5]. Engineering curricula often focus on the “back-end” of design, where processes focus on developing, building, and testing concepts[6], potentially because these processes appear deterministic (such as optimizations). However, many crucial design decisions are made in the front-end phase[7], and skills in front-end design are necessary to successful design. This study focuses on one important stage in the front-end design: idea generation; in particular, what are strategies and tools that can be used to support front-end success in design? One of the challenges for both engineering students and practicing engineers is to develop innovative solutions to problems because innovation is necessary to solve the world’s most pressing issues.[8] As increasing technology allows engineers to learn more about existing and emerging problems, it also provides opportunities to develop solutions like never before.[9] Consequently, engineers must be successful in idea generation[8, 10] by fully exploring solution spaces through the generation of multiple and diverse ideas[11]. Considering a wide spectrum of ideas at the front end of the design process is important in order to consider strong aspects of concepts to develop while filtering out other aspects. Generating diverse ideas to consider requires divergent thinking,[4, 12] while engineering methods focus on converging on a single solution. While experienced designers appear to use strategies in their idea generation process[13, 14], novice designers may have fewer approaches to help them consider a variety of concepts. One way to assist novice engineers in idea generation is through idea generation tools. A common tool is Brainstorming, the technique of generating ideas without evaluating them, encouraging wild and exaggerated ideas, valuing quantity over quality, and considering every idea to be of equal worth[15]. Brainstorming is the most popular tool used across all disciplines as a way to generate ideas[16]. One challenge with Brainstorming, however, is that it fails to provide students direction or inspiration beyond their initial ideas. Another design tool with increasing use in engineering contexts is Design Heuristics, which are a set of 77 strategies to incorporate into product design idea generation[17]. The Design Heuristics ideation tool is derived from evidence of strategy use by engineers, and has been studied across contexts including undergraduate and graduate engineering, industrial designers, high school engineering education, and professional engineering and design practices[13]. For this study of idea generation in beginning engineers, we compare these two ideation methods to examine qualitative differences in outcomes. Background Research indicates the importance of considering multiple concepts in the early stages of design.[3] However, many engineering students do not dedicate adequate time to the idea generation process. This leads to limited consideration of alternative solutions, and can potentially decrease the chance of innovative outcomes. Further, designers often become fixated on their initial ideas, and find it difficult to think of very different concepts[18, 19]. Engineering students have been shown to struggle with fixation to a greater extent when compared to industrial design students[18]. Engineering students in particular may benefit from the use of idea generation tools to help them push beyond initial ideas to solutions outside the obvious[20]. One ubiquitous approach to idea generation across disciplines is Brainstorming, a technique originally proposed for groups to help postpone judgment of ideas, encourage wild ideas, aim for quantity over quality, and value every idea[15]. As originally proposed by Osborn), Brainstorming was developed as a technique to use with groups[15]. In practice, “brainstorming” refers to any method of idea generation where groups or individuals are instructed to generate as many different ideas as possible[21]. Engineering instructors, in many cases, encourage their students to generate ideas using “brainstorming,” but may not provide students with specific instruction on how to execute it (following Osborn). Instead, the term may be used to suggest a “natural” approach to thinking of ideas, pursuing whatever comes to mind in the moment. The expectation is often that ideas should arise without using any cognitive strategy, and without any instructions on how to ideate successfully. This lack of instruction often prevents novice engineers from considering possible strategies as they create and develop concept ideas[14]. As a result, students are less likely to recognize the large space of potential designs possible, and may not seek out tools to help them identify varied candidate designs. The instructions in Brainstorming [15] push designers to continue creating more solutions; however, they do not guide or direct towards potential solutions. This lack of direction also causes students to rely on their initial ideas[34]. Another idea generation approach is Design Heuristics, an evidence-based concept generation tool developed to help designers think systematically and intentionally about variety in idea generation[20, 22]. Collected from protocols of engineers talking aloud while creating designs, analyses of award winning products, and a case study of a long-term design project, Design Heuristics capture the cognitive “rules of thumb” used by designers to intentionally vary their set of candidate designs[23]. These strategies appear to be ones that expert designers employ automatically, without consciously deciding to do so[24]. The heuristics were individually extracted across multiple concepts from multiple designers to reflect a useful level of abstraction in describing how to alter design characteristics to create new ones[25]. The resulting set of Design Heuristics capture 77 different strategies, each of which can be applied independently or in to create new designs[26]. The set of Design Heuristics is packaged as an instructional tool for use in idea generation. A set of 77 cards includes prompts, such as “Add Motion,” as shown in Figure 1. Each prompt is a different strategy to help the designer generate a novel idea. One side of each card has a graphic demonstrating the strategy along with a written description (Figure 1). The other side includes two examples of the heuristic as employed in two existing products (Figure 1). To illustrate that each heuristic may be applied to every design problem, one of the illustrated products is always a type of chair. The other product on each card varies, demonstrating that these strategies apply across a wide range of applications and industries. While initial use is supported by the cards, it is possible that novices will eventually incorporate the heuristics they often use into their everyday design practice as they develop design expertise[11]. There are many ways that a single Design Heuristic can be applied during idea generation. Past studies have documented their use by means of a single application of one heuristic card to generate a new design, repeated use of one card to generate multiple designs, and combining multiple cards to generate an idea[20]. In addition, students have successfully transformed an existing design into a new one by applying a heuristic[27]. Design Heuristics have also been successful when used in conjunction with other idea generation strategies, such as Morphological Analysis[17]. Research on Design Heuristics has shown that engineering students at varying levels of training can learn to use the Design Heuristics cards within a short instructional session, and then go on to successfully create their own novel and diverse concepts[20]. One study of 48 first-year engineering students given different subsets of 12 Design Heuristics used Design Heuristics in over half of their created concepts for a portable solar oven[28]. Further, the concepts resulting from the application of Design Heuristics were rated by blind coders as more creative designs. Studies with more advan