Ideation is core to the innovation process, and has been the subject of study across a range of fields, from psychology to engineering. However, despite substantial progress in outcome--based descriptions of idea generation, research has often resulted in more questions than answers. For example, open questions remain with respect to the differences in behaviour related to ideation between novices and experts, the change in rates of ideation over time in different design teams, and the changing role of ideation from conceptual to detailed design. In each of these cases robust explanation has proved elusive due to difficulties in characterising the ideation process itself. This paper discusses a major new approach for elucidating ideation and its related design processes through direct observation. A novel network visualization approach is demonstrated in practice for the first time. This uses network analysis to link ideas dynamically to both the engineering context and the wider design process. This linking analysis gives substantial new insight into what drives ideation and how previously inscrutable results can potentially be explained by linking ideation into other design processes.Keywords: creativity and innovation; human creativity; psychology of creativity; descriptive models of the design process; empirical study To cite this article : Philip Cash & Mario Štorga (2015): Multifaceted assessment of ideation: using networks to link ideation and design activity, Journal of Engineering Design, DOI: 10.1080DOI: 10. /09544828.2015 To link to this article: http://dx.doi.org/10.1080/09544828.2015.1070813 2
IntroductionIdea generation is the foundation of innovation and forms the basis for much of new product development and business growth (Chulvi et al. 2012; López--Mesa et al. 2011). This key role has driven an on--going focus on understanding and supporting ideation in practice (Lopez--Mesa et al. 2009;Kudrowitz and Wallace 2013). There have been numerous studies of ideation processes in both cognitive science, and design and innovation (Dorst and Cross, 2001). However, models in cognition have typically been derived from controlled laboratory experiments using simple, isolated tasks (Kirk, 2009). These bear little similarity to ideation in the real world context where there are multiple interacting processes (Spitas 2011).Conversely, more realistic tasks have difficulty discriminating the numerous processes at work due to limitations in measurement approaches (Vargas--Hernandez, Shah, and Smith, 2010). These more 'real world' studies can struggle to generalise their results across contexts due to difficulties in drawing relations between interacting process elements (Kitchenham, 1996). This is highlighted by works such as Schlecht and Yang (2014) where ideation is a key element, but the mechanisms by which it occurs are difficult to assess due to a reliance on outcome--based metrics, such as, number of ideas or quality. In both cases, the reliance on outcome measures means that suggestions for process improv...