PSD.15:4 - Solution

Develop the repertoire around recurring decision-support results and the conditions that make them possible. Recover the actual Methods, compare live alternatives, and retain or change the offering with its source and evidence return. Trial a consequential change only when further evidence can change that decision and its attainable contribution warrants the burden. These are dependencies in reasoning, not a compulsory lifecycle for every engagement.

PSD.15:4.1 - Bound the repertoire question by the work it must serve

Name the practitioner group or service using the repertoire, the recurring situation, the receiving decision-support result, and the horizon over which the offering matters. The starting question might be: “How can our engagements preserve a late but material participant correction without losing the ability to produce a timely qualified recommendation?”

Recover only conditions that can change the answer: disputed or already formed question; participants and power; access to affected parties; values and protected conditions; alternatives; uncertainty; available evidence; technical and facilitation capability; decision authority; receiving Work; and follow-up need.

Use a PSD.14 result only when its evidence can change this repertoire decision under the relevant configuration and horizon. A changed mobile-deployment premise may show that the existing follow-up Method worked by returning the correct blocker. It does not by itself demonstrate a defective Method. If a supplied result is missing, stale or incompatible, use a qualified direct account or name the precise gap.

Distinguish a missing input, a failure to enact the stated Method, a support or capability problem, and a defect in the reusable way itself. A practice cannot repair absent stakeholder access simply by adding another modelling technique. Conversely, more training cannot repair a Method rule that systematically excludes the required correction.

PSD.15:4.2 - Recover candidate ways of working and their contribution boundaries

State each serious candidate as a way of doing: its entry conditions, participant meanings, actions, input/result commitments, required joins, allowed variation, and stop. Use A.3.1 for Method identification. Keep an unresolved proposal as a candidate account; an attractive name, complete document or parent Method does not admit it.

When several grounded occurrences or direct sources leave the reusable way genuinely unclear, A.3.1.MR can help recover candidate accounts. It is not the route for filling a missing design choice in a prospective recipe. Ask the designer to state the proposed action and its conditions.

Decompose a heavyweight school or package only far enough to compare contributions. The following contrasts are useful starting questions, not an exhaustive taxonomy or automatic assignment rule.

Soft Systems Methodology, SODA/cognitive mapping, Strategic Choice, MCDA, scenario and robust-decision approaches can contribute to different rows under their actual source conditions. A software platform may support several such contributions. None supplies a universal Method merely by appearing in the repertoire.

PSD.6 supplies the engagement-level comparison and combination discipline. Preserve its joins: an attributed concern may become a candidate objective for examination, not automatically a numerical weight; a scenario may constrain a calculation without supplying a probability. If a proposed whole cannot recover its reusable action and joins, retain the bounded candidate or arrangement description rather than calling co-use a composite Method.

PSD.15:4.3 - Compare the incumbent with a serious alternative before designing a variant

Name the current offering, a plausible competing Method, and a bounded adaptation or combination only where each is live. Compare them against the same recurring result and the actual conditions, at comparable effort where possible.

Ask what each lets participants and practitioners do, what it leaves unresolved, and where it fails. In addition to analytical adequacy, consider meaningful challenge, unequal power, facilitation burden, interpretation skill, source qualification, time to a usable return and cost of maintaining the offering. State a deliberately accepted trade-off when the richer result costs more.

Do not infer that one Method is universally better because it produces a precise score, a satisfying workshop or more alternatives. A single well-qualified Method can remain the best offering for a narrow use. Plural formulations can be necessary elsewhere; a numerical comparison can still be necessary after them.

When a change is needed, state its mechanism: which reusable action, input condition, dependency, decision rule or stop changes, what remains, and why that difference could repair the failure. Distinguish this from a change in wording, display, software, local staffing or performance. The latter may require maintenance or a new fit check without identifying another Method.

For example, replacing a paper map with a digital map while preserving attribution, challenge and unresolved-claim rules may change support and access. Replacing the challenge rule with automatic aggregation of all statements changes the reusable way and can lose dissent. The same interface change can therefore have different repertoire consequences depending on what practitioners actually do.

Use current ME.15 guidance when generic variant identity, derivation or non-variant maintenance is the live subquestion and its exact claims fit. The remaining PSD question is whether that way serves the professional decision-support use: how interpretation, participation, values, uncertainty, authority and follow-up contribute to a usable return. Generic variant maintenance does not answer that field judgement.

PSD.15:4.4 - Build a source-to-claim basis, not a popularity ranking

For every repertoire claim that can change later use, retain the source contribution and its actual limit. An instruction can describe a Method; a reported application can show that it was used; an observed interaction can illuminate a failure mechanism; a comparative study can support a bounded advantage. These are not interchangeable evidence.

A useful basis line says: this source or qualified experience supports this action or applicability claim for these conditions; this uncertainty remains; this new observation would change the offering. Include source edition, relevant locus, evidence window and original meaning where they matter. Recover contrary cases and the serious alternative, not only supportive citations.

The 2019 Smith and Shaw review provides an exploratory way to compare PSM characteristics; its classification is not an exhaustive admission rule. Kogetsidis’s 2025 application review and 2026 online review of Europe and beyond provide leads about reported applications. Their journal and search boundaries limit what they establish about practice outside that corpus. Do not rank Methods by the number or location of publications.

A recent decision-analysis synthesis such as Borgonovo et al. (2026) can identify developments in value-focused thinking, graphical modelling, uncertainty, sensitivity and information acquisition. It does not substitute for the source of a particular Method or cover every adjacent decision-support family.

Obtain missing source interpretation, domain evidence or representative evaluation through A.15.9, preserving supplier authority, and qualify actual reliance through A.10. If the decisive source contribution cannot be recovered, keep that offering conditional or return the source gap. Neither a famous school nor a new source date repairs the missing claim.

PSD.15:4.5 - Trial the claimed improvement in representative engagements

For a selected trial, turn the proposed improvement into a question that can fail. Specify the Method or status-preserved candidate, the situation represented, participant and power conditions, required capability and support, the serious comparator, the expected result, the burden accepted, and the observation that would retain, narrow or reject the proposal.

Select evidence proportionate to the claim. A walkthrough can expose an impossible join. A constructed exercise can test whether practitioners preserve an objection through a calculation. An authorized field trial can test feasibility and use in a real engagement. Broader transfer or causal-effectiveness claims need stronger and appropriately designed evidence. Do not generalize the last two from the first two.

Observe the working mechanism as well as the output: who could contribute or challenge; which material statement entered the model; where meaning changed; what was withheld; how uncertainty and protected conditions reached the recommendation; and whether the receiver could use it. The distinction between technically valid models and what people do with them is central to Franco et al.’s behavioural OR review (2021). It supports attention to intervention configuration and interaction, not one universal trial design.

Preserve competing explanations. Better results may reflect a more skilled facilitator, a different participant group, more time, an easier problem or changed support rather than the Method difference. Where those cannot be separated, report what the trial establishes and what remains unresolved; a useful feasibility result need not pretend to be a causal result.

Do not use a consequential live decision as an uncontrolled Method experiment. Establish the authority, consent, protection, observation and fallback conditions appropriate to the engagement. A prospective candidate can be investigated through a separate trial arrangement without claiming that an unadmitted candidate whole has already been enacted as a U.Method.

Retain disconfirming results. A richer Method that cannot run within the actual access or competence conditions may be a poor offering for that use even when its idealized output is better. Conversely, a failed local performance does not disprove the Method when a required condition was absent; it may instead defeat the offered applicability claim.

PSD.15:4.6 - Return a small current repertoire with differentiated dispositions

Decide what the practice can now offer, for which use and with which limits. Useful dispositions include retaining an established offering; narrowing its applicability; adding a separately identified variant for a qualified use; retaining a proposal for a bounded trial; requesting missing evidence; and withdrawing an offering from one use while preserving its history.

Keep identification separate from validation. An independently identified Method may still have unproved practical fit. A candidate account may have useful trial evidence without yet resolving its Method identity. State both rather than using one maturity label for every question.

For each action-changing entry, make the following recoverable in the simplest form that serves its next user.

A repertoire may retain alternatives without ranking them. When the actual question is declaring a selector-facing set result over already identified members, use G.5 with the exact members, outcome, ordering and basis it requires. Distinguish alternatives retained for later choice from members all included in one named joint use. A Method-family registry or selector declaration supplies neither field fit nor actual enactment and need not be rebuilt for every repertoire update.

Make the next use easy to find: which Method a practitioner should inspect for the named situation, which claim still needs evidence, or which use has been withdrawn. A large source map that gives none of these is not the promised repertoire.

PSD.15:4.7 - Preserve cultural lineage without inventing cultural continuation

Record where a reusable change came from when that relation matters: an identified source Method, a documented adaptation, a changed stopping rule, a taught practice, or another qualified source. Similar names, chronological succession and a shared diagram do not establish derivation.

Keep proposed derivation, observed local enactment, transmission to another practitioner, recognition, selection, retention and loss distinct. A source publication can make a description available without proving that anyone learned or enacted the Method. A successful trial can justify a restricted offering without showing that a professional community adopted it.

Use C.36 for an actual cultural-evolution question. Supply PSD.17 with the repertoire’s qualified variant or local evidence only when it matters to that cultural-continuation decision. Name the practitioner population, engagement/community, place and period that the evidence actually covers, and mark the unobserved links. The receiving cultural decision is not entailed or authorized by the repertoire.

This boundary applies equally to new software support and AI-assisted practice. A tool release is not cultural uptake; widespread use of a tool is not evidence that the same Method was enacted by everyone. Teaching, service change and deliberate cultural intervention require their own work and authority.

PSD.15:4.8 - Refresh the affected offering and preserve an honest stop

Maintain a responsible source and use return for the current repertoire. Reopen an entry when new evidence, a source claim, changed Method semantics, participation or authority conditions, support, qualification window or receiving need can change its offering. A new publication date or a file rename alone is not such a change.

For a changed claim, identify its exact repertoire use and the consequence: preserved, narrowed, reopened for a new trial or source result, or withdrawn from that use. Preserve independent entries and the earlier evidence for its original conditions. If applicability remains unresolved, state the uncertainty rather than assigning an unsupported negative or positive status.

A changed service rule might restrict the human career-advising branch without changing organization or AI-domain truth. A new AI configuration may invalidate evidence for a software-assisted offering without changing the manual Method. If the reusable decision-support action itself changes, identify and assess that semantic change rather than hiding it as maintenance.

Use A.10 for a known bounded reliance, and A.10.1 when a changed source requires discovering and revalidating several actual receiving uses. Invoke G.11 for an admitted currentness or refresh-planning/reporting object when its conditions obtain, such as a relied-on evidence or selected set; it does not supply the new repertoire judgement or make every local entry a refresh-orchestration problem.

Close this repertoire question when the relevant offerings, conditions, source/evidence gaps and next uses are clear. Withdraw only the defeated use, not recoverable history or other supported uses. A still-unqualified new proposal can remain visible without being offered as established practice.