Traceability for Automated Production Systems: A Position Paper
Mounifah Alenazi, Nan Niu, Wentao Wang, Arushi Gupta · 2017
Automated production systems are design-to-order, custom-built mechatronic systems that are intended to deliver automation capabilities to satisfy the stakeholder requirements in the manufacturing/production domain. Traceability has its research root in requirements engineering (RE) and is defined as "the ability to describe and follow the life of a requirement". Such descriptions and followings have mainly been scoped within a single software project's process and artifacts. We argue that, in the context of automated production systems, the scope of traceability shall go beyond the software engineering boundary into the environments which the software is operated. We outline in this position paper our aim to define a new form of the trace links and to explore the role of these links in the RE of automated production systems. In particular, we adapt a formal interaction-oriented RE framework that focuses on specifying the commitments of participants rather than the goals of each individual participant. We then integrate model checking into the framework to elaborate when, why, and how much traceability should be instrumented. We demonstrate our approach with a bench-scale automated production system where model-driven engineering practices such as constructing and evolving SysML models are adopted.