Implementing shared function modelling in practice: experiences in six companies developing mechatronic products and PSS

ABSTRACT

Function modelling is proposed in textbooks to support solution finding early in a development process. Although it has great potential to support the comprehensiveness and systematisation of conceptual design, it has been taken up in design practice only to a limited extent. Based on a literature review, potential barriers hampering its widespread application are discussed. A central problem revolves around the abstract nature of function modelling, which makes it difficult to apprehend the concrete benefits from its implementation. The article contributes to related research by showcasing success stories in the implementation of function modelling in six companies. The implemented models include morphological charts, use case modelling and two approaches developed in-house. The paper presents an interview study with managers and design leaders investigating the motivations and reasons leading up to the decision to newly implement function modelling in the companies. Furthermore, the paper explores the interviewees’ reflections on experiences gained while (gradually) implementing function models, the specific advantages attained, and also the challenges faced in doing so. The findings show that shared function modelling has great potential to support interdisciplinary collaboration and comprehension and to lead to a significant increase in the efficiency of the design process.

KEYWORDS:

• Functional modelling
• method introduction
• interdisciplinary design
• product-service systems
• mechatronics

Acknowledgements

The author would like to thank all participants of the study for their engagement and valuable insights. Further thanks go to Prof Anja Maier and Prof Emeritus Mogens Andreasen for helping with the development of the used questionnaire as well as Dr Kilian Gericke from the and University of Luxembourg and Prof Lucienne Blessing from the Singapore University of Technology and Design for their support. Finally, we would like to thank the reviewers whose comments have helped us improve this manuscript during the review process.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Other prominent approaches include the Function-Behaviour-Structure framework (Gero 1990), Structure-Behaviour-Function model (Iwasaki et al. 1993), the Function-Behaviour-State model (Umeda and Tomiyama 1997) and the conglomerate approaches by Tjalve (1978) or Hubka and Eder (1988; Eder and Hosnedl 2010).

2 Unified Modeling Language (UML, ISO 2012) and System Modeling Language (SysML, OMG 2012).

3 Comprehensive overviews and discussions on these challenges are provided e.g. by Warell 1999; Chandrasekaran and Josephson 2000; Maier and Fadel 2001; Chiang, Pennathur, and Mital 2001; Deng 2002; Chandrasekaran 2005; Vermaas 2013; Crilly 2010; Carrara, Garbacz, and Vermaas 2011; Aurisicchio et al. 2011, 2012 or Goel 2013.

4 There are exceptions, as Gericke, Roschuni, and Kramer (2015) found that some designers are generally curious and intrinsically motivated to explore new methods that might be useful to them.

5 It must be noted that both companies represent very large organisations with big design teams in the automotive industry, which have vast data management systems in place that are vital for their internal quality assurance.

6 Though the model was developed independently in-house by the company, interestingly, the basic principle of mapping functions and solution elements has similarly surfaced elsewhere as well, e.g. in the Quality Function Deployment matrix (QFD2) after Aurisicchio, Bracewell, and Armstrong (2012) and in other companies (see Eisenbart, Gericke, and Blessing 2016a).

7 He later added that this is because designers, subcontractors and clients simply have a hard time linking the function specification with the solution elements that are proposed and are already fixed in the final product.

8 Arguably, mapping functions and solution elements is not completely new to engineering design research. We already discussed the advancement of the Allocation Matrix from morphological charts and its resemblance to the QFD2 matrix after Aurisicchio, Bracewell, and Armstrong (2012). Similarly, Function Means Trees (see Hansen and Andreasen 2002; Moehringer and Gausemeier 2002; Robotham 2010), for instance, entail a one-directional allocation from the functions to the components that are to fulfil them. Typically, each function is mapped onto one component from top to bottom. Here, however, the concept of top-down and bottom-up traceability is newly introduced. This means that not only are the designers able to visualise which subsystems/components – alone or in combination – realise each function, but they are also able to trace which functions and functionally linked subsystems/components may be affected if a change is introduced to one or more subsystems/components.

Additional information

Funding

This work was part of a PhD research project supported by the Fonds National de la Recherche Luxembourg under grant number AFR PHD-09-186.