Is Biologically Inspired Invention Different?

The paradigm of biologically inspired design views nature as a vast library of robust, efficient and multifunctional designs, and espouses the use nature as a source of analogues for inspiring novel designs in domains of interest such as architecture, computing, engineering, etc. Over the last generation, biologically inspired design has emerged as a major movement in engineering, architectural, and systems design, pulled in part by the need for environmentally sustainable design and pushed partly by the desire for creativity and innovation in design. An important question is whether biologically inspired design is fundamentally different from other kinds of analogybased creative processes. This question is critical because the computational theories, techniques and tools we need to develop to support biologically inspired design depend on the nature of the task itself. In this paper, we first summarize some of our empirical findings about biologically inspired design, then derive a task model for it, and finally posit that biologically inspired design indeed is a novel methodology for multiple reasons. Biologically Inspired Design The paradigm of biologically inspired design (also known as biomimicry, biomimetics and bionics) views nature as a vast library of robust, efficient and multifunctional designs, and espouses the use of nature as an analogue for designing technological systems as well as a standard for evaluating technological designs (Benyus 1997; French 1994; Gleich et. al. 2010; Turner 2007; Vincent & Mann 2002; Vogel 2000). This paradigm has inspired many famous designers in the history of design including Leonardo da Vinci, and in a wide variety of design domains ranging from architecture to computing to engineering to systems. However, over the last generation the paradigm has become a movement in modern design, pulled in part by the growing need for environmentally sustainable development and pushed partly by the desire for creativity and innovation in design. Thus, the study of biologically inspired design is attracting a rapidly growing literature, including patents (Bonser & Vincent 2007), publications (Lepora et al. 2013), and computational tools (Goel, McAdams & Stone 2014). The Biomimicry Institute (2011) provides numerous examples of biologically inspired design. The design of windmill turbine blades mimicking the design of tubercles on the pectoral flippers of humpback whales is one example of biologically inspired design. As Figure 1 illustrates, tubercles are large bumps on the leading edges of humpback whale flippers that create even, fast-moving channels of water flowing over them. The whales thus can move through the water at sharper angles and turn tighter corners than if their flippers were smooth (Fish et al. 2011). When applied to wind turbine blades, they improve lift and reduce drag, improving the energy efficiency of the turbine. Figure 1: Design of windmill turbine blades to increase efficiency inspired by the tubercles on humpback whale flippers. (The Biomimicry Institute 2011) From the perspective of computational creativity, two characteristics of biologically inspired design are especially noteworthy. Firstly, biologically inspired design often is creative: its products, such as the windmill turbine blades illustrated in Figure 1, are novel, valuable, feasible, and non-obvious (even surprising at first). Secondly, the conceptual phase of biologically inspired design engages analogical transfer of knowledge from biological analogues to design problems in the domain of interest. The latter point raises an important question: is biologically inspired design fundamentally different from other kinds of analogy-based creative processes other than the obvious fact the source domain here is biology? This question is important because the computational theories, techniques and tools we need to develop to support biologically inspired design depend on the nature of the task. For example, Nagle (2014) describes an engineering-to-biology thesaurus that maps function terms used in engineering into equivalent function terms used in biology. The (implicit) assumption in the work on the engineering-to-biology thesaurus is that biologically inspired design is not very different from other analogy-based creative processes (e.g., Veale 2003), that if we could only bridge the vocabulary Proceedings of the Sixth International Conference on Computational Creativity June 2015 47 gap between design and biology, we could borrow the rest from extant theories of design, analogy and creativity. In this paper, we first summarize some of our empirical findings about biologically inspired design and then derive a Task Model for it. Finally, we will posit that biologically inspired design is a novel methodology for multiple reasons, and thus requires the development of new computational theories, techniques and tools. Research Methodology Theories of biologically inspired design process can be normative and prescriptive or descriptive and explanatory. Vincent’s et al.’s (2006) BioTRIZ theory, for example, is a normative and prescriptive account of biologically inspired design. In contrast, we have developed a descriptive and explanatory account. Thus, our research methodology consists of three major elements: In situ observations of biologically inspired design practices, task analysis of biologically inspired design, and comparison with current theories of design, analogy and creativity. Observations of Biologically Inspired Design Practices: Given that the professional biologically inspired design community at present is nascent, sparse and diffused, we studied biologically inspired design practices in the Georgia Tech ME/ISyE/MSE/BME/BIOL 4740 course from 2006 through 2013 taken by ~350 students. This a yearly, interdisciplinary, project-based course on biologically inspired design taught jointly by biology and engineering faculty. The class is composed of mostly senior-level undergraduate students from biology, biomedical engineering, industrial design, industrial engineering, mechanical engineering, and a variety of other disciplines. Although it evolves a little every year, the course is consistently structured around lectures, found object exercises, journal entries, and one or more design projects. Some lectures discuss biological systems; some lectures focus on case studies of biologically inspired design; and some lectures formulate, analyze and critique problems for students to solve in small groups. Yen et al. (2011, 2014) provide a detailed account of the teaching and learning in the course. Task Analysis of Biologically Inspired Design: Given our observations in the ME/ISyE/MSE/BME/BIOL 4740 classes from 2006 through 2013, we conducted a task analysis of the macrostructure of biologically inspired design practices. Crandall, Klein & Hoffman (2006) describe the methodology of task analysis in detail. Task analysis helps identify the task decomposition of a complex task, the methods used to accomplish the various subtasks in the task decomposition, and the contents of knowledge used by the different methods. For example, Chandrasekaran (1990) presents a high-level task analysis of the general design task while Goel & Chandrasekaran (1992) present a task analysis of the specific method of case-based design. In general, task analysis may describe the behaviors of an individual designer, the interactions among a team of designers, or the behaviors of a design team viewed as a unit. Although we are interested in all three levels of aggregation, in this work we focus on interdisciplinary design teams of biologists and engineers viewed as the unit of analysis. Our task analysis of biologically inspired design by interdisciplinary design teams generates a task model of biologically inspired design: the task model describes the processes and the knowledge used in biologically inspired design. Comparative Analysis with Theories of Design, Analogy and Creativity: Given our task model of biologically inspired design, we compared it with theories of biologically inspired design such as BioTRIZ (Vincent et al. 2006) and Design Spiral (Baumeister et al. 2012). However, because of space limitations, here we will compare our task model only with BioTRIZ. We also compared our task model with established theories of analogical reasoning such as Gentner (1983), Hofstadter (1996), Holyoak & Thagard (1996), and Kolodner (1993). Again because of space limitations, here we will compare our task model only with Gentner’s structure-mapping theory of analogy. ! Data The ME/ISyE/MSE/BME/BIOL 4740 classes from 2006 through 2013 resulted in 83 extended, open-ended design projects. The 83 case studies of the design projects in the classes were the focal points of our data collection. The projects involved identification of a design problem of interest to the team and conceptualization of a biologically inspired solution to the identified problem. Each design project grouped together an interdisciplinary team of typically 4-5 students. Each team had at least one student with a biology background and a few from different engineering disciplines. Each design team also had at least one faculty member. Each team identified a problem that could be addressed by a biologically inspired solution, explored a number of solution alternatives, and developed a final solution design based on one or more biologically inspired designs. Each design team presented its final design to an interdisciplinary design jury. Goel et al. (2015) describe a digital library, called the Design Study Library (DSL), of all 83 case studies.