Multi-domain change impact analysis for agile cyber-physical production systems engineering

Abstract

The manufacturing and production industry is on the verge of a transformation to intelligent networking of machines and processes with the help of information and communication technology, known as Industry 4.0. Cyber-Physical Production Systems (CPPSs) play a central role in this transformation as they offer flexibility and resilience. In CPPS engineering, teams from multiple domains work simultaneously and iteratively on a variety of assets to design and build CPPS, or parts of it. The focus of this thesis lies on multi-domain CPPS engineering, where experts from different domains work together. This thesis addresses the challenges of Multi-domain Change Impact Analysis (M-CIA) of technical changes throughout the CPPS lifecycle. The challenges of such change impact analysis include the exchange between stakeholders from different domains, the high complexity of integration, and implicit and scattered production knowledge. The thesis applies the design science methodology and builds on agile change management workflows and existing software engineering approaches. The thesis 1) presents the results of an expert survey, 2) identifies challenges of M-CIA in existing environments 3) designs and validates the M-CIA method and a supporting system design. The results are validated with a feasibility and a case study with the help of a system prototype. The evaluation results of the method and the system design show that the solution approach is feasible. An increase in efficiency and perceived improvement are determined through a case study with an industry partner. The proposed M-CIA method utilizes established Dev- and GitOps practices in software development to facilitate and automate multi-domain change impact analysis in CPPS. The M-CIA method and system design enable practitioners and researchers to improve the coordination of multi-domain change impact analysis in the CPPS environment. In doing so, it facilitates cross-domain stakeholder exchange, centralization of production knowledge, and integration of domain-specific perspectives on their production system, thus preventing costly and late adjustments.