Jump to text
On this page
Copy and source
Published source confirmed at last check

Source changed 2026-11-18 21:11:39 UTC · snapshot created 2026-11-18 21:12:35 UTC · last check 2026-11-18 22:25:20 UTC

SYSE.7:11 - SoTA and Source Use

Working question. Once engineering descriptions are available and controlled, what further work makes their claims usable together for one decision? The selected answer is an authored combination: retain adequate data-management arrangements, then establish only the missing claim-level interpretation and applicability. Keep that result available to the decision and to the changes that can defeat it.

NASA’s Technical Data Management treatment, §§6.6.1.2–6.6.1.3 is a serious alternative. It selects needed data, addresses adequacy and consistent use, defines exchange conventions and responsibilities, and weighs acquisition burden. It also permits existing contractor arrangements when they meet project requirements. Adopt that selection and reuse principle. A tailored implementation may already provide all the content review this decision needs. Retain an applicable result whose claim interpretation, configuration basis and receiving use remain recoverable.

For a concrete comparison, give both approaches the HP-2 inputs in :5 and engineers able to interpret the sensor specifications and control-model assumptions. Either can discover the conflict between the assumed two-hertz, 0.1-kelvin bound and the selected variant’s half-hertz, 0.3-kelvin bound. Controlled exchange alone leaves that conflict unresolved. Where the existing review already connects those claims to the selected variant and the architecture choice, :4 and :4.2 reuse it. Where that connection is missing, :4.1 moves 4–8 recover the interpretation, test the correspondence and return the unsupported architecture claim. The additional useful result is a recoverable reason for limiting the storage branch and reconsidering it after sensor substitution, as :5 demonstrates.

Both arrangements need the specialist interpretation. This pattern additionally asks that the material correspondence, applicability and return remain recoverable. Accept the work of recording and maintaining them when dependent decisions or changes would otherwise require reconstruction or risk reusing the wrong result. For a one-off decision, a short explanation in its existing record may suffice. A maintained shared account earns its cost through Work that relies on it; stop maintaining entries under :9 when that work is no longer needed. Total effort depends on how often the result is reused and how often its basis changes.

A shared semantic model is another serious alternative when recurring exchanges justify the conversion and maintenance. The landing-gear case in Wu et al. (2025) combines shared model semantics, conflict handling, traceability and versioning. Adapt that joint treatment in :4.1 moves 4, 6 and 10. Use an existing common model when it retains the distinctions needed by the decision; use explicit correspondences between specialist descriptions when conversion would lose a needed distinction or cost more than maintaining those correspondences. For HP-2, either arrangement must preserve the sensor variant and its timing and error bounds. Reject shared names or storage co-location as substitutes for that relation. The case supplies a feasible arrangement, not evidence that one architecture wins across domains.

Bound the reliance on each comparison. Riedmaier et al. (2021), §§2.5 and 7 distinguish validated conditions from the intended model application and examine uncertainty: adopt that distinction in :4.1 moves 5 and 8. Schwarzburg, Trauer and Rebentisch (2024), §§5.2–5.3 examine model history, competence, access and decision risk: adapt these as questions for a consequential unresolved reliance under SYSE.10, without importing an exploratory confidence score as an acceptance threshold. The study’s small non-probability sample cannot establish comparative industrial effectiveness. These sources limit what either data arrangement can warrant: internally consistent claims need a qualified basis for the physical use.

For optional AI assistance, adapt the task-specific allocation proposed by Pradas Gomez et al. (2025), §§4–5. In :5, the AI System proposes correspondences; the assigned engineers check their sources and implications before accepting reliance. Tools remain resources under :4.3. The source offers a conceptual model for choosing contributions. Its survey and literature review do not establish how well the proposed HP-2 arrangement works. Decision authority must be established separately from that source and from assigning the work.

Reconsider the choice when a material relation cannot be recovered before use, a changed configuration or source defeats its applicability, or a different arrangement supplies the same needed distinctions with less total work. Reopen the affected claims and return them under :4.1 move 11 and :4.3. A credibility gap goes to SYSE.10; a configuration-identity gap goes to SYSE.13; a changed description used by Work returns through SYSE.19. Preserve the usable remainder. This is also the limit on :12’s return to SYSE.6: only an account applicable to the same architecture decision, configuration and horizon can inform it.