Boundary Objects in Agile Practices

Continuous Management of Systems Engineering Artifacts in the Automotive Domain.

We are honored to have received a best paper award at ICSSP conference for this collaborative work with Rebekka Wohlrab, Patrizio Pelliccione, Eric Knauss and Mats Larsson.

In this paper, we derive guidelines for managing requirements, models, and other systems engineering artifacts continuously. This becomes an increasingly important capability, especially as more and more complex system development organizations transition towards agile methods. This can for example be observed in the automotive domain.

When reasoning on how to manage such artifacts, we found that the most important criterion is the artifacts’ context of use: Is it locally relevant to a (cross-functional) team? Or is it a boundary object relevant to two or more teams, potentially on different abstraction levels?

Locally relevant artifacts

Since such documents are only used by the team itself, it can (and likely: should) be retired when it is no longer useful. The other way around: if the document is supposed to be kept alive, we recommend to keep it useful by using it for example to generate (parts of) the output of the team, e.g. code. Thus, our recommendation is to focus on prescriptive models and similar artifacts.

Boundary objects

These artifacts are used to coordinate between teams and represent an interface between these teams. Examples include architectural models, high-level requirements, and variability information.

Our guidelines

Our first two guidelines are targeting system management artifacts in general, to create the sufficient overview to systematically manage artifacts.

  • (G1) Involve stakeholders from different areas to identify what artifacts are boundary objects and locally relevant artifacts. Find out how and by whom they are used, who should be responsible at what point in time, and how traceability can be established.
  • (G2) Make sure that you evaluate artifacts’ relevance and usage at frequent intervals. Depending on how the system evolves throughout its lifecycle, artifacts might become more relevant or should be discarded. Boundary objects could be converted into locally relevant artifacts and vice versa.

For boundary objects, we offer the following guidelines:

  • (G3) For each boundary object within your organization, establish
    a group of representatives from affected teams to discuss issues and later propagate that information. Store information in an accessible tool.
  • (G4) Find a lightweight and flexible approach to defining high- level artifacts upfront. Exploit this information to make impact analysis and changes, to avoid suboptimal decisions. With time, the artifacts should be refined and become more mature.

Finally, we also offer guidelines for locally relevant artifacts:

  • (G5) Produce locally relevant artifacts, especially those for documentation, as late as possible and only when they are actually needed. If possible this documentation should be automatically generated from other artifacts and code.
  • (G6) Aim to make locally relevant artifacts reusable (as with pre-scriptive models) and convey their relevance and use. Establish traceability to boundary objects.

Please refer to our paper for details, including our reasoning on why we provide these guidelines.


Reference: Wohlrab, R.; Pelliccione, P.; Knauss, E. & Larsson, M.: Boundary Objects in Agile Practices: Continuous Management of Systems Engineering Artifacts in the Automotive Domain (PDF). In: Proc. of Int. Conf. on Software and System Process (ICSSP), pp. 31-40, Gothenburg, Sweden, DOI: 10.1145/3202710.3203155, 2018

Abstract: Automotive companies increasingly include proven agile methods in their plan-driven system development. The adoption of agile methods impacts not only the way individuals collaborate, but also the management of artifacts like requirements, test cases, safety documentation, and models. While practitioners aim to reduce unnecessary documentation, there is a lack of guidance for automotive companies with respect to what artifacts are needed and how to manage them. To close this knowledge gap and create practical guidelines, we conducted a design-science study together with 53 practitioners from six automotive companies. Using interviews, surveys, and focus groups, we analyzed categories of artifacts and practical challenges to create applicable guidelines to collaboratively manage artifacts in agile automotive contexts. Our findings indicate that different practices are required to manage artifacts that are shared among different teams within the company (boundary objects) and those that are relevant within a specific team (locally relevant artifacts).

Advertisements