Open Access

Software is, in its essence, an inherently invisible digital construct, and thus its comprehension and its visualization remain a challenge. All software involves some underlying structure(s), and Software Architecture (SA) comprises the (intended) conceptual abstractions and structuring principles across this invisible construct. Agile development methods, DevOps, and continuous development results in a changing implementation and associated SA that is evolving and continually in flux. Any presumed SA understanding and (perhaps outdated or inconsistent) associated SA documentation may also diverge from the reality, while any shared SA concept across stakeholder minds may vary or differ, potentially resulting in a lack of conceptual integrity. In contrast, an Informed Software Architecture (ISA) is grounded in reality based on actual data and evidence, rather than being influenced by out-of-sync models, documentation, misconceptions, or assumptions. Yet the challenge remains of how best to visually convey ISA aspects, such as internal static software structures and behavioral and operational dynamics, to support evidence-based design, comprehension, and insights in an accessible way for a wider stakeholder spectrum. This paper contributes VR-ISA, a Virtual Reality (VR) solution concept to immersively support an ISA with the visualization of structural, behavioral, and operational aspects. To exemplify our solution concept, three VR-based viewpoints, framing different concerns for different stakeholder groups, are used to illustrate the potential of VR to support ISA: 1) components and connectors, for depicting dynamic distributed event and data streams, 2) modules and dependencies, for depicting static internal module composition and their dependencies, and 3) execution observability, for depicting operational execution, tracing, and observability aspects. Our realization shows its feasibility, while a case-based evaluation provides insights into its capabilities and potential.

As systems grow in complexity, so does their associated lifecycle and with it the need to manage the various elements, relations, and activities involved in the Software (or Systems) Development Life Cycle (SDLC). Various notations for system, software, or process modeling have been specified such as the Systems Modeling Language (SysML), Unified Modeling Language (UML), Business Process Model and Notation (BPMN) respectively, yet due to their two-dimensional (2D) diagram focus, they are ill suited for visualizing a comprehensive contextualized view of the entire systems engineering or software engineering lifecycle. To address this need, the Lifecycle Modeling Language (LML) utilizes a relatively simple ontology and three primary diagrams while supporting extensibility. Yet lifecycle comprehension, analysis, collaboration, and contextual insights remain constrained by current 2D limitations. This paper contributes our Virtual Reality (VR) solution concept VR-SDLC for holistic visualization of SDLC elements, relations, and diagrams. Our prototype implementation utilizing LML demonstrates its feasibility, while a case study exhibits its potential.

As systems grow in complexity, the interdisciplinary nature of systems engineering makes the visualization and comprehension of the underlying system models challenging for the various stakeholders. This, in turn, can affect validation and realization correctness. Furthermore, stakeholder collaboration is often hindered due to the lack of a common medium to access and convey these models, which are often partitioned across multiple 2D diagrams. This paper contributes VR-SysML, a solution concept for visualizing and interacting with Systems Modeling Language (SysML) models in Virtual Reality (VR). Our prototype realization shows its feasibility, and our evaluation results based on a case study shows its support for the various SysML diagram types in VR, cross-diagram element recognition via our Backplane Followers concept, and depicting further related (SysML and non-SysML) models side-by-side in VR.

Software design patterns and the abstractions they offer can support developers and maintainers with program code comprehension. Yet manually-created pattern documentation within code or code-related assets, such as documents or models, can be unreliable, incomplete, and labor-intensive. While various Design Pattern Detection (DPD) techniques have been proposed, industrial adoption of automated DPD remains limited. This paper contributes a hybrid DPD solution approach that leverages a Bayesian network integrating developer expertise via rule-based micropatterns with our machine learning subsystem that utilizes graph embeddings. The prototype shows its feasibility, and the evaluation using three design patterns shows its potential for detecting both design patterns and variations.

Today’s Industry 4.0 Smart Factories involve complicated and highly automated processes. Nevertheless, certain crucial activities such as machine maintenance remain that require human involvement. For such activities, many factors have to be taken into account, like worker safety or worker qualification. This adds to the complexity of selection and assignment of optimal human resources to the processes and overall coordination. Contemporary Business Process Management (BPM) Systems only provide limited facilities regarding activity resource assignment. To overcome these, this contribution pro- poses a BPM-integrated approach that applies fuzzy sets and rule processing for activity assignment. Our findings suggest that our approach has the potential for improved work distribution and cost savings for Industry 4.0 production processes. Furthermore, the scalability of the approach provides efficient performance even with a large number of concurrent activity assignment requests and can be applied to complex production scenarios with minimal effort.