PSD.3 - Generate Plural Problem Formulations

Type: DPF pattern body Status: Candidate

Primary working result: a plural formulation set for one bounded receiving inquiry: materially different accounts of what warrants intervention, each with its subject, boundary, affected participants or Systems, value tension, evidence need, intervention logic, assumptions, and consequence for the receiving use.

PSD.3:0 - Use This When

Use this pattern when one wording of “the problem” is becoming naturalized even though materially different accounts would change what is included, what evidence matters, which interventions are plausible, whose values are exposed, or what the recipient should do next.

Begin from the bounded engagement question in PSD.1. Use the PSD.2 participation and concern account when affected participants, Systems, duties, interests, or dissent are material. Generate formulations that differ in decision-bearing structure, not merely vocabulary, and return them without selecting a winner.

The practical gain is an inspectable choice of frames before modeling, option generation, or comparison locks in the sponsor’s first account. Do not use PSD.3 for a thesaurus exercise, generic creativity session, alternative generation for an already-stable problem (PSD.8), or archive/front stewardship (C.18). If one formulation is already adequate and no credible rival could change the receiving use, proceed to the actual missing result.

PSD.3:0.1 - Working Distinctions

Working term	Meaning in this pattern
problem formulation	A bounded, source-qualified account of what condition or relation warrants inquiry or intervention for a named receiving use. It proposes a subject, boundary, concern, evidence need, and intervention logic; it is not the situation itself.
plural formulation set	Two or more materially different formulations retained together because each can change a downstream boundary, model, alternative family, value treatment, evidence request, or recommendation.
material difference	A difference that changes at least one decision-support move. Synonyms, rhetorical emphasis, or unexplained labels are not material plurality.
intervention logic	The connection a formulation proposes between the condition treated as problematic and the families of action, inquiry, adaptation, or restraint worth considering. It is a hypothesis, not proof of causation or effectiveness.
framing commitment	A supported or provisional inclusion, exclusion, assumption, value emphasis, or causal hypothesis built into one formulation.

PSD.3:1 - Problem frame

Ill-structured situations do not arrive with one neutral problem object waiting to be described. “Insufficient capacity”, “unequal protection”, “maintenance fragility”, and “unsafe deployment” can concern the same pumping arrangement while selecting different bearers, boundaries, evidence, interventions, and consequences. The formulations may conflict, overlap, or remain jointly useful.

The primary concern is to make those decision-bearing differences recoverable before one model or Method absorbs them. Recognition is cheap: one credible rival intervention logic is enough to test plurality. Assurance remains direct for every factual, causal, evaluative, legal, ethical, safety, authority, or source claim used within a formulation.

PSD.3:2 - Problem

The sponsor’s first sentence often becomes the model title, data request, option space, and final recommendation. Its assumptions disappear because they are encoded as scope. A second failure produces many “perspectives” that differ only in wording and therefore add ceremony without changing any downstream work. A third treats plurality as permanent indecision and refuses to bound the inquiry.

The needed discipline is neither immediate convergence nor unlimited reframing. It is to construct a small set of materially different formulations, expose what each makes visible and suppresses, and stop when additional formulations no longer change the current receiving use.

PSD.3:3 - Forces

Force	Tension
Plurality	Rival formulations can prevent premature closure, while decorative variety wastes effort.
Usability	Each formulation must guide a next move, while early precision can pretend that contested claims are settled.
Participant fidelity	Concerns should shape formulations, while participation does not prove the formulation.
Evidence	A formulation reveals what evidence matters, while it must not smuggle unsupported causes or effects into the case.
Exploration and stopping	New frames can reveal alternatives, while an unbounded set cannot support a decision.
Authority	A recipient may compare formulations, while neither a facilitator nor the set itself owns the later choice.

PSD.3:4 - Solution

Generate a bounded set of problem formulations whose differences are visible in their downstream consequences. Bind every formulation to the same receiving inquiry unless an identity change is explicitly returned to PSD.1. Preserve sources, assumptions, uncertainty, and conflicts; then hand the set to boundary, modeling, alternative, and inquiry patterns without declaring a chosen problem.

PSD.3:4.1 - Fix the receiving inquiry, not the problem answer

State the subject, recipient, receiving decision or Work, horizon, and retained choice owner from PSD.1. Import only those PSD.2 participants, affected Systems, concerns, duties, interests, and dissent that can change this inquiry. If the subject or receiving use differs, create or recover the other engagement rather than merging formulations across identities.

Write the current formulation in one sentence. Mark its source, factual support, assumptions, and important exclusions. This makes the default frame inspectable rather than privileged.

PSD.3:4.2 - Generate structural contrasts

Seek contrasts along the smallest productive set of questions:

What condition or relation is treated as problematic, for whom or what?
Which bearer, place, configuration, horizon, or consequence is central?
Which value, duty, interest, or risk makes intervention worth considering?
Which mechanism or dependency is hypothesized, and what evidence could challenge it?
Which action families become plausible, and which disappear?
Who can prepare, receive, authorize, or implement a response?

Do not require one formulation per question. Combine contrasts into coherent candidate formulations. If two formulations imply the same boundary, evidence, action family, and receiving-use consequence, merge them or state that their difference is rhetorical.

PSD.3:4.3 - Give every formulation a usable face

For each retained formulation, record:

Position	Required content
formulation	One ordinary-language account of what warrants inquiry or intervention.
subject and bearers	Exact subject, affected participants or Systems, configuration, place, and horizon needed by the claim.
boundary and exclusions	What the formulation includes, what it suppresses or defers, and why.
value tension	Attributed concern, duty, interest, risk, or desired change that makes the formulation material.
evidence need	Current support, uncertainty, contrary evidence, and the smallest question that could reverse or narrow it.
intervention logic	Action, inquiry, adaptation, or restraint families made plausible, without claiming effectiveness.
receiving-use consequence	Which model, alternative, comparison, scope, recommendation, or decision-support return would change.

Where a durable formulation is used as a claim-bearing episteme, preserve its source and identity through the direct episteme pattern. Ordinary working sentences need no new root kind.

PSD.3:4.4 - Test material plurality without ranking

Compare each pair only far enough to ask whether the difference changes a downstream move. Preserve incompatibility when one formulation requires evidence, values, boundaries, or interventions that another excludes. Record overlap when formulations can be investigated together.

Do not score, rank, vote, synthesize, or select unless another pattern and authority govern that result. C.17 may characterize novelty, diversity, or use of an already-generated formulation set when that characterization is material. C.18 applies only when the project must govern generation records, descriptors, archives, fronts, telemetry, or retained exploration value. Neither pattern supplies the PSM-specific construction of problem formulations.

PSD.3:4.5 - Stop and return the plural formulation set

Stop when every retained formulation changes at least one named downstream move and another credible formulation would not change the current receiving use enough to justify its cost. There is no universal target count. A claim of plurality requires at least two materially different formulations; a case with only one adequate formulation should not manufacture a set.

Return the formulations, their contrasts, shared and disputed premises, evidence gaps, dependent next results, and reopen conditions. If a missing participant, concern, fact, source, or authority relation prevents a responsible formulation, return the exact gap, narrow the claim, request the bounded result, or abstain.

PSD.3:4.6 - Pass plurality forward without creating a workflow

Pass the set to PSD.4 when included and excluded scope must be chosen, to PSD.5 when different models are needed, to PSD.8 when formulations imply different alternative families, and to PSD.16 when engagement activities obstruct a material formulation contribution. These branches may be concurrent. Pattern numbering is not a compulsory stage sequence.

Reopen locally when a new participant, concern, affected System, evidence result, value tension, intervention family, horizon, receiving use, or authority fact creates a materially different formulation or defeats one already retained.

PSD.3:4.7 - What changes in practice

The practitioner stops asking participants to agree on one problem sentence before useful work begins. They can show a small set of genuinely different formulations, the evidence and actions each foregrounds, and the exact next result that plurality changes.

PSD.3:5 - Archetypal Grounding

PSD.3:5.1 - Four formulations of the flood-pump situation

For the board’s pre-season decision, the PSD.2 account supports four material formulations:

Capacity shortage: peak inflow can exceed available pumping capacity; flow, reliability, and deployment evidence matter; permanent or mobile capacity becomes salient.
Unequal protection: current arrangements shift flood exposure toward particular districts and mobility-constrained residents; distribution, access, and representation evidence matter; allocation and protection rules become salient.
Maintenance fragility: failures arise from maintenance, spares, staffing, and coordination dependencies rather than nominal pump count; readiness, contracting, and staged maintenance alternatives become salient.
Unsafe deployment: emergency placement and operation can create worker, resident, traffic, or downstream risks; safe operating envelopes and restraint branches become salient.

These are not synonyms. They imply different scope, evidence, model, alternative, and value work. The team keeps overlaps and conflicts visible and sends the set to PSD.4; it does not vote for “the real problem” or infer that any intervention is effective.

PSD.3:5.2 - Development-direction advice

For an AI-assisted incident-response arrangement, “model accuracy is inadequate”, “operator coordination is fragile”, and “authority escalation is unclear” imply different holders, evidence, intervention families, and safety questions. PSD.3 retains all three only if each can change the two-quarter development recommendation. Generic slogans such as “improve AI capability” and “make the system better” fail the material-difference test.

PSD.3:5.3 - Cheap non-use

A repair team has a verified broken seal, one affected assembly, one authorized repair decision, and no rival account that changes evidence or action. The live question is which qualified replacement to use. Generating plural problem formulations would add ceremony; proceed directly to comparison or choice.

PSD.3:6 - Bias-Annotation

Scope: plural problem formulation for one bounded decision-support inquiry. Lenses: Gov keeps framing influence and later choice authority distinct; Arch exposes how formulations change dependent results; Onto/Epist separates the situation, formulation, claims, evidence, and causal hypotheses; Prag requires material difference and a stopping rule; Did provides ordinary-language contrasts before formal models.

Recurring bias	Likely drift	Repair
sponsor-frame bias	The opening brief becomes the only formulation.	Externalize it with sources and exclusions, then seek structural contrasts.
vocabulary bias	Synonyms are counted as plurality.	Require a changed boundary, evidence need, intervention family, value, or receiving-use consequence.
consensus bias	Participants must agree on one wording before inquiry.	Retain incompatible formulations and name their downstream differences.
causality bias	A framing hypothesis is reported as the cause.	State mechanism, support, uncertainty, and falsifying or narrowing evidence.
option leakage	Problem formulations become preselected solutions.	State intervention families as consequences of a frame, not proven answers.
endless-reframing bias	Every possible perspective stays live.	Stop when added formulations no longer change the named use enough to justify cost.

PSD.3:7 - Conformance Checklist

Every formulation concerns the same bounded subject and receiving use, or the identity change is returned to PSD.1.
Material participants and concerns from PSD.2 are used as inputs without being treated as proof or settlement.
Each retained formulation changes at least one boundary, evidence need, model, alternative family, value treatment, or receiving-use consequence.
Sources, assumptions, exclusions, uncertainty, and intervention logic are visible.
Causal, evaluative, legal, ethical, safety, and authority claims retain their direct assurance.
The set is not ranked, selected, synthesized, or converted into a decision by this pattern.
The stopping rule prevents both decorative plurality and endless reframing.
Missing premises produce a gap, narrowed claim, request, blocker, or abstention.
The result can feed PSD.4, PSD.5, PSD.8, or PSD.16 without imposing a universal traversal order.

PSD.3:8 - Common Anti-Patterns and How to Avoid Them

Anti-pattern	Repair
“Agree on the problem first.”	Agree on the receiving use; retain materially different problem formulations.
“Brainstorm ten frames.”	Keep only formulations that change a named downstream move.
“Every stakeholder gets a frame.”	Build coherent formulations from material differences; participation identity does not define one frame each.
“The most popular frame is true.”	Popularity establishes neither truth, causal support, value priority, nor authority.
“A novel wording is a new problem.”	Use the material-difference test; use `C.17` only if novelty characterization itself matters.
“The formulation set is an option archive.”	Use `PSD.8` for decision alternatives and `C.18` for governed archives or fronts.

PSD.3:9 - Consequences

Later models, boundaries, alternatives, and recommendations inherit visible framing assumptions instead of one hidden sponsor frame. Practitioners can investigate incompatible accounts without forcing premature consensus. The cost is explicit source and contrast work, and some engagements will stop because a framing premise cannot yet be supported.

PSD.3:10 - Rationale

Problem formulation is a field-specific generative move: it connects contested descriptions, concerns, values, evidence needs, and intervention logics to a receiving decision-support use. Generic candidate-set patterns can characterize or steward a set after generation, but they do not construct problem formulations. PSD.3 therefore owns the smallest plural-formulation result while reusing direct FPF patterns for evidence, episteme identity, relations, set characterization, and authority.

PSD.3:11 - SoTA-Echoing

Practice question	Best-known line	Serious alternative or default	Defect overcome and pattern mutation	Source roles and limits	Reopen condition
How should a practitioner preserve plural framing without losing a usable inquiry?	Treat framing as selection and organization of what matters, externalize materially different formulations, compare their consequences, and revise them as contextual evidence changes.	A fixed sponsor brief is the serious default; unconstrained perspective collection is the opposite default.	The first hides framing commitments and the second cannot stop. Adapt: `PSD.3:4.1`–`4.5` requires structural contrasts, downstream materiality, sources, and a receiving-use stop.	Kelly and Gero’s 2022 framing review supports multiple non-equivalent meanings of design framing; Litster, Cardoso, and Hurst’s 2024 study supports observation of changing externalized team framing; Nickel, Hurst, and Duimering’s 2024 study supports contextual trade-offs. These sources establish neither one framing ontology, universal Method, shared team frame, prevalence, causality, nor superiority.	Reopen if later comparative evidence changes the best-known framing line or if contextual evidence materially changes a retained formulation.
What remains specifically PSM work after generic candidate handling is subtracted?	PSM practice retains the construction and facilitation of rival problem formulations for ill-structured situations; FPF supplies direct evidence, relation, identity, and later-choice governors around that result.	Treating formulation as generic brainstorming, alternative generation, or archive management is the serious default.	It loses the relation between a formulation and the concerns, boundaries, evidence, and intervention logic of the receiving inquiry. Adopt: the formulation face in `PSD.3:4.3` and the pass-forward boundary in `4.6`.	Smith and Shaw’s 2019 review and Kogetsidis’s 2025 review support PSM as a continuing, diverse, facilitated branch for ill-structured situations. Current `C.17` characterizes novelty/diversity/use of a named set; current `C.18` governs open-ended generation, archives, fronts, descriptors, and telemetry. None proves that one PSM, formulation count, facilitation arrangement, or archive policy is universally sufficient or superior.	Reopen if a direct generic pattern acquires the full PSM-specific construction at equal or lower effort, or if PSM evidence changes the material plurality test.

PSD.3:12 - Relations

PSD.1 supplies the bounded subject, receiving use, horizon, recipient, and authority boundary. PSD.2 supplies material participants, affected Systems, concerns, duties, interests, and dissent.
PSD.4 receives the plural set when the current included/excluded problem boundary must be chosen. PSD.5 receives formulation-dependent model needs.
PSD.8 receives formulation-dependent alternative families; it does not retroactively define the problem. PSD.16 can use the plural formulation set when deciding how to resolve material interference in the engagement.
C.17 may characterize novelty, diversity, or use of the generated set. C.18 governs archive/front/generation-record stewardship only when those objects are actually needed.
Direct episteme, evidence, relation, value, causal, legal, ethical, safety, governance, and choice patterns govern their own claims. PSD.3 neither ranks the set nor authorizes the later choice.