Abstract
The Industry 4.0 (I4.0) vision is about efficient, flexible production supporting rapidly changing requirements through digitalisation based upon enabling middleware software architectures. This paper presents a proposal for a systematic approach to gather quality attribute requirements by presenting a three-phase process and the application of it in a collaboration project with participants from industry and academia.
The first phase, data collection, creates common understanding and gathers initial requirements in the form of industrial high-level requirements (IHLR). The second phase, analyse architectural requirements, systematically analyses the IHLR through a Quality Attribute Workshop (QAW), and the third phase, evaluate QAS, evaluates the Quality Attribute Scenarios (QAS) and facilitates knowledge sharing.
The process presented resulted in 6 IHLRs and 14 associated requirements that were analysed and subsequently mapped into 10 Quality Attributes. The Quality Attributes were refined and described systematically and finally evaluated in interviews with the project partners.
Our study contributes towards a better understanding of I4.0 by 1) reporting on a systematic approach of collecting I4.0 requirements from the industry, and 2) deriving QASes for I4.0 middleware supporting these requirements.
Why this matters
Quality attributes (security, performance, modifiability, interoperability) are the drivers of architectural decisions, but they are notoriously hard to elicit from industry stakeholders who think in business terms. This paper documents a repeatable process for getting from “we want flexible production” to QAW-style scenarios concrete enough to drive architecture – with empirical evidence from a real industry-academia project group.
Citation
Jepsen, S. C., Worm, T., Christensen, H. B., Hviid, J., & Sandig, L. M. (2021). Experience Report: A Systematic Process For Gathering Quality Attribute Requirements for Industry 4.0 Middleware. In 2021 IEEE 25th International Enterprise Distributed Object Computing Workshop (EDOCW), pp. 166–175. IEEE. https://doi.org/10.1109/edocw52865.2021.00046