This study was initiated from designers´ needs to find the most creative engineering solutions in the early idea stages and convert these into feasible design concepts (Efimov-Soini, 2019; Tsai & Yang, 2017). It also evaluates three commonly used visual representation of ideas , namely prototyping, sketching, and computer-aided design (CAD) modelling, in terms of their respective representation qualities to support different phases of the design process (Howard, Culley, and Dekoninck, 2008; Tsai & Yang, 2017).
Recognised as formal design representations, prototypes, sketches and CAD-models have been accredited as aids for facilitating design research and ideation (Babapour, 2015; Camere & Bordegoni, 2016; Joundi et al., 2020; Kelly & Gero, 2014). Before 3D CAD models, physical prototypes and sketches were the most practical and fastest ways to convey and develop ideas for designers (Charlesworth, 2007; Tsai & Yang, 2017; Babapour, 2015). The introduction of 3D computer modelling software notably replaced, to some extent, traditional visualisation methods, which transformed the difficult, slow, and dirty creation process into a more efficient and clean virtual way of designing and prototyping (Isa & Liem, 2014). However, Charlesworth (2007) and Ekströmer & Wever (2019) suggested that designers may encounter restrictions when using CAD models during the early idea. Indeed, the use of CAD models and prototypes as design tools has been widely discussed concerning how they contributed to the thinking processes, as well as how the design activities were managed (Hallgrimsson, 2012 and Isa & Liem, 2020). On the one hand, several researchers discouraged using prototypes in early idea, because building prototypes early can be expensive and time-consuming and should therefore only be done when needed (Dow et al., 2010; Stowe, 2009; Viswanathan, Atilola, Esposito & Linsey, 2014; Murphy et. al, 2022). On the other hand, other researchers insisted that prototyping should be applied early in design and development activities to better understand problems, encourage faster learning, and increase communication among stakeholders and designers (Charlesworth, 2007; Isa & Liem, 2020; Jensen, Elverum, & Steinert, 2017; Leifer & Steinert, 2012; Sanders & Stappers, 2014).
However, limited research is available on the effectiveness of prototypes, sketches and CAD-models in the early generation of ideas (Hess & Summers, 2013; Babapour, 2015; Tsai & Yang, 2017). There is therefore a need for more research on the progression and transition of representations to explore and experience different design activities experienced by designers during the early stages of the engineering design process. We draw on existing studies by Babapour, Hess, Howard and their associates to designate the flow of design activities to guide creativity as “early idea”, “idea development”, “and finalization” (Babapour, 2015; Camere & Bordegoni, 2016; Hess & Summers, 2013; Howard, Culley, and Dekoninck, 2008). This study addresses one research question: How do prototypes, sketches and CAD models supported designers during the early idea, idea development and finalization stages in the design process?
1. Creativity in the Engineering Design Process
This section introduces the early stages of the engineering design process, how design activities, i.e. prototyping, sketching and CAD-modelling are used to facilitate early-stage analysis and creative development, as well as how designers manage creativity constraints and design fixation.
1.1 Divergence and convergence steps in early stages in the Engineering Design Process
According to a comprehensive literature study of 23 engineering design process frameworks (see Howard, Culley, & Dekoninck, 2008), six essential stages can be distinguished in a creative design process. These comprise (1) establishing a need, (2) early idea development (task analysis), (3) idea development (conceptual design), (4) embodiment design, (5) final idea development (detailed design), and (6) implementation. This study focuses on the representation aspects of the process: stages, (3), (4) and (5). During the “idea development” stage, designers iteratively explored the problem space to understand the context that the product will be a part of. Hereby, design problems were addressed holistically in a structured or unstructured manner through a process of transformative idea generation.
In the “embodiment design” stage, these initial ideas were then further developed, refined and more frequently verified against design requirements while addressing localised-, /sub-problems. During this stage, quantity was emphasised to generate as many solutions as possible. Especially, during the initial divergent phases, the problem-space was continuously evaluated and redefined from different perspectives so as to pre-empt too early fixation (Council, 2005; Gray, Brown & Macanufo, 2010; Liu et al., 2003; Meinel, Leifer & Plattner, 2011; Cross, 2021). Dym et al. (2005) defined the divergent process as an activity, where questions dominate facts to enlarge the creative space.
During the final ideation phase, mechanisms, components, user interfaces, aesthetics, and ergonomics of the product were further detailed once the final concept has been selected. Every component was designed according to the most suitable production technologies, with detailed specifications on the materials, sizes, and standard requirements (Cross, 2021)
The engineering design process is characterised as convergent, with alternating divergent and convergent moves, focussing on quantity at the beginning of the process, and consecutively narrowing down to potential and a final solution, as illustrated in Figure 1 (Cropley, 2010; Howard et al., 2008; Leifer & Steinert, 2012; Liu, Chakrabarti & Bligh, 2003; Leifer and Steinert; 2012).
For divergent phases, the argument for valuing quantity over quality is based on studies showing that designers are more likely to acknowledge the uncertainties of the “so-called unknown knowns”, the more they explore and experiment (Bernal, Haymaker & Eastman, 2015; Jensen et al., 2017; Sutcliffe & Sawyer, 2013). For convergent phases, the contrary argument applies, where the possibilities were narrowed down and future actions are chosen (de Oliveira, Rozenfeld, Phaal & Probert, 2014). Dym et al. (2005) defined the convergent processes as an operating knowledge domain. In convergent processes, knowledge holds “true” values and is used to specify the actual task or refine and finalize the concept/product (Cropley, 2010).
1.2 Design Activities during early stages of the engineering design process
During the design process, several methods can be utilized to define the future requirements of the final product (Howard et al., 2008; Sutcliffe & Sawyer, 2013). In this paper, we focus on prototyping, sketching, and CAD modelling because they are among the most commonly used design representation tools to communicate design ideas during the concept development stage (Babapour, 2015; Tsai & Yang, 2017; Menold, et. al, 2019). Each of these design activities is briefly described below.
1.2.1 Sketching
As one of the most significant activities during the design process, sketching refers to producing visual images to externalise the process of visual thinking and assist in the creation and development of visual ideas (Mao et. al, 2020 and Ibrahim & Pour Rahimian, 2010; Bao, Faas, & Yang, 2018). Sketches signify a variety of information and are used in diverse ways (Babapour, 2015). A sketch is a preliminary and unpolished visual representation of something without emphasising details. More importantly, it is usually executed rapidly and presents only the key elements of the design. Sketches comprise informal freehand marks without the use of instruments and may include draft lines, text, dimensions, and calculations that help explain the meaning, context, and scale of the design (Pei, Campbell & Evans, 2011 and Bao, Faas, & Yang, 2018). The ambiguity of sketches distinctively creates assumptions, imaginations and reinterpretations, especially of those elements and views that are not explicitly shown. In other words, by omitting details in sketches, the creative space for follow-up ideas will be enlarged (Menezes & Lawson, 2006). However, Schütze, Sachse, & Römer (2003) and Bao, Faas, & Yang, (2018) suggested that sketches have a positive impact on the quality of the designed solution and facilitate individual experience during the designing process. As such, sketches can be perceived as a valuable tool for facilitating visual interactions among stakeholders in the early idea stages, because of their ambiguity (Menezes & Lawson, 2006; van der Lugt, 2005). According to the taxonomy by Pei at al. (2011), a diversity of visual representations can be distinguished for developing industrial and engineering design work.
1.2.2 Computer-Aided Design (CAD) Modelling
CAD modelling and other forms of digital modeling can be used to define the geometry and visual appearance as well as to facilitate simulations, variations, and appearance testing of future products (Tovey & Owen, 2000). Since their usage became more prominent in the early 2000s, it has been discussed whether they should be applied for only 2-D or to be extended to 3-D representation, especially in early-stage concept development (Ibrahim & Pour Rahimian, 2010; Ekströmer & Wever, 2019; Coyne et al. , 2002). Furthermore, studies conducted by Ibrahim & Pour Rahimian (2010) and Robertson and Radcliffe (2009) claimed that CAD should not be used during the early stages of product development as it can constraint exploration. Other studies present counter arguments outlining major advantages typically ascribed to CAD, such as improved solutions through broader explorations of the solution space as well as cost reduction in project work because of frequent and early iterations (Fixson and Marion, 2012). CAD has therefore been included in this study to further explore these tensions in conjunction with the design process.
1.2.3 Physical Prototyping
The progression in the construction of a physical representation of an idea is called physical prototyping (Houde & Hill, 1997; Lim, Stolterman & Tenenberg, 2008). According to Häggman, Tsai, Elsen, Honda, and Yang, (2015); Menold, et. Al, 2019), Neeley, Lim, Zhu, and Yang, (2013), prototyped designs are generated more quickly than those created using sketches or CAD. However, depending on their purpose, manifestations can be made through the use of different materials, and throughout the engineering design process, to facilitate exploration, evaluation, and communication (Beaudouin-Lafon & Mackay, 2007; Camere & Bordegoni, 2016; Houde & Hill, 1997; Jensen, Balters & Steinert, 2015; Isa and Liem, 2020; Murphy et. al, 2022).
During the initial stages of the engineering design process, low-fidelity physical prototypes can be produced from all types of materials and consistencies, including cardboard, foam, or wood, as well as through role-playing to facilitate exploration and defining requirements rather than concept testing (Ulrich & Eppinger, 2008; Yang, 2005). This indicates that a prototype shares a low resemblance with the final product (Sanders & Stappers, 2014; Yang, 2005), allowing accurately scaled and high-fidelity physical models to be constructed in the later stages for hypothesis and requirement testing. (Jensen et al., 2017; Sutcliffe & Sawyer, 2013; Ulrich & Eppinger, 2008).
1.3.Creativity Constraints and Design Fixation during the Design Process
The overall aim of the design process is in the end to generate a well thought-out solution containing some degree of creativity (Howard et al., 2008). To reach this goal, designers consciously guide themselves through divergent and convergent processes to develop the most creative and innovative solutions (Liu et al., 2003). However, creativity constraints and design fixation can affect the process (Crilly & Cardoso, 2017; Jensen, Birkeland & Steinert, 2016; Onarheim, 2012). Creativity constraints stem from the theory of priming effects which argues that the tightness of these constraints determines a designer´s creative space (Colombo, Bartesaghi, Simonelli & Antonietti, 2015; Onarheim, 2012). In this context, the professional expectations to use certain design tools, such as sketches, prototypes, or CAD-models, can be perceived as creatively constraining or inspiring. How constraining or inspiring certain tools are, is dependent upon the number of ideas generated or the functionalities the participants end up with during prototyping sessions (Friedman, Fishbach, Förster & Werth, 2003; Onarheim, 2012). Nevertheless, Vasantha, Chakrabarti, Rout, and Corney, (2014) mentioned that irrespective of the design tool used, novice designers generated a lower number of redesigned concepts. Therefore, designers might require training for reinterpreting and extracting the necessary information from originally captured concepts, rather than working with poor understanding, ambiguity, and assumptions about the original designer's intent. This requires designers to balance a fine line between being constrained and inspired to enter the so-called sweet spot of creativity (Onarheim and Biskjaer,2015). To avoid over-constraining and under-constraining, Crilly and Cardoso (2017) described the concept of ‘design fixation’ as ‘a blind adherence to a set of ideas or concepts limiting the output of conceptual design’, which designers can mitigate by applying different methods. However, no specific claims were made on how prototypes, sketches, and CAD models support or undermine creativity, because of other underlying motives during different idea generation stages.