SYSE.4:11 - SoTA-Echoing
| Current practice line | What changes in this pattern | Source and use | Adoption status |
|---|---|---|---|
| Current model credibility research treats model use as decision-specific and makes error, uncertainty, extrapolation, technical validity, model history, competence, access, and decision risk visible. | A model or simulation challenge names its applicability, uncertainty, extrapolation, configuration, and decision question; confidence does not become validity or physical truth. | Riedmaier et al. (2021), survey of more than 200 sources; Schwarzburg, Trauer, and Rebentisch (2024), literature review, 40-person survey, and an untested confidence-assessment proposal. | Adopt and bound. Use decision-specific credibility questions; select the VV&UQ Method and warranted reliability claim for the receiving model use. |
| Digital-twin engineering uses heterogeneous model transformation, code generation, and interpretation across design, implementation, and operation, mainly in manufacturing and transport. | A digital-twin arrangement can contribute a named model, automation, observation, or evidence-maintenance relation, but is not itself the challenge, physical evidence, or assurance result. | Lehner et al. (2025), mapping study of 66 included publications and 136 reported applications. | Adapt. Recover the twinned subject, model use, domain, result, and maturity limit; use the arrangement only for its established contribution. |
| Continuous integration, CPS, and SRE practice combine automated feedback with simulation, Hardware-in-the-Loop, physical checks, and risk-sensitive review in bounded technology settings. | Challenges may run frequently and produce results used as evidence during realization and operation, while cadence, pipeline structure, risk rule, permission, and release remain local. | Current DORA capability pages; Thurgood’s 2018 SRE error-budget example; Zampetti et al. (2022), interviews in ten organizations and a 55-practitioner survey. | Adapt narrowly. Use frequent feedback where conditions fit; establish the local automation boundary, cadence, pipeline, and independent-assurance need. |
| Continuing requirements and compliance research keeps collaboration, traceability, monitoring links, legal interpretation, engineering descriptions, and evidence connected as systems change. | A challenge can cite maintained traceability and monitoring, but the account keeps legal interpretation, engineering claim, evidence use, specialist verdict, and decision separate. | Hernández, Moros, and Nicolás (2023); Norheim et al. (2024); Kosenkov et al. (2025). | Adapt within scope. Preserve continuing evidence links; let the receiving legal and engineering practices determine their Methods and Work organization. |
| BDD and test-intent research shows that software scenarios and executable checks can clarify part of intended behaviour, while industry evidence and automation coverage remain limited. | A scenario-to-check link can become one planned challenge for a software claim, but the check does not establish an outside effect, obligation, acceptance, or complete assurance. | Mohanani et al. (2022); Binamungu and Maro (2023); Lahiri et al. (2022); Fakhoury et al. (2024); Wang et al. (2025). | Use as a bounded branch. Retain test-intent clarification for software and qualify transfer of its metrics or result scope for each other engineered System. |
| Early model-based V&V and system-theoretic assurance research remains heterogeneous and profile-specific. | The common claim–challenge–evidence-use–revision method stays a guide-derived synthesis; specialist frameworks contribute only the named analysis, traceability, or argument result used by the decision. | Cederbladh, Cicchetti, and Suryadevara (2024), systematic review, DOI 10.1145/3631976; Ahlbrecht, Sprockhoff, and Durak (2024), aircraft-safety proof of concept, DOI 10.1007/s10270-024-01209-6. | Keep provisional. Use each V&V or safety-assurance framework within its supported profile and transfer only its established contribution. |
These sources support particular challenge, model-use, automation, traceability, and specialist-assurance branches. The common claim–challenge–evidence-use–revision Method remains a cross-source synthesis. Reopen it when later cross-domain comparison changes evidence composition, operating-feedback use, assurance-case applicability, or the boundary with specialist safety, security, ethics, legal, certification, acceptance, permission, or release decisions.