Stakeholder-Weighted Ranking

Rank with weights by stakeholder perspective: identify all relevant parties (researchers, engineers, policymakers, end users, etc.), construct an independent weight vector for each class of party, rank separately, and then take the consensus.

When to Use

• The research involves multiple stakeholders (e.g., medical AI: clinicians + patients + regulators)
• Different parties have fundamentally divergent definitions of "importance"
• A consensus must be built across parties, or the ranking differences across perspectives must be shown
• A funding agency or collaborator needs to see priorities from their own perspective

Thinking Framework

Core principle: there is no objective "most important gap", only "most important to whom".

The process has three layers:

Layer 1: Stakeholder identification List all groups that would be affected by the research results. Each class of party has a different value function — engineers value feasibility, policymakers value impact, academic researchers value novelty.

Layer 2: Within-perspective ranking For each class of party, use the same four-dimensional scoring as multi-criteria-ranking, but with a different weight vector. For example:

• Academic researchers: novelty 0.40, importance 0.30, impact 0.20, feasibility 0.10
• Engineers: feasibility 0.40, impact 0.30, importance 0.20, novelty 0.10
• Policymakers: impact 0.45, importance 0.35, feasibility 0.15, novelty 0.05

Layer 3: Consensus merging Borda count or weighted-average the per-perspective rankings, identifying the "cross-perspective robust top gaps" (deemed important by all parties) and the "perspective-divergent gaps" (highly valued by some parties, ignored by others).

Key insight: perspective divergence is itself information — a gap with large divergence may need interest-alignment first, rather than a direct attack.

Budget Gate

| Tier | Gap count | Party count | Consensus method | Final output | |------|---------|-----------|---------|---------| | S | 5–10 | 2–3 classes | Simple average | Per-perspective rankings + consensus top-3 | | M | 11–20 | 3–5 classes | Borda count | Per-perspective rankings + consensus top-5 + divergence analysis | | L | 20+ | 5+ classes | Weighted Borda + sensitivity | Full perspective matrix + consensus ranking + divergence heatmap |

Default Reference Flow

1. Call the gap-normalization SOP: unify gap format
2. Identify stakeholder classes (CC judges autonomously or the user specifies)
3. For each class of party, call the ahp-weighting SOP: generate that perspective's weight vector
4. Run the four-dimensional scoring in parallel for each class of party (importance-scoring, feasibility-scoring, novelty-scoring, impact-scoring)
5. Call the scoring-matrix-construction tactic: build the gap × party × dimension three-dimensional matrix
6. Call the priority-sensitivity-testing tactic: test the effect of stakeholder weight changes on the consensus ranking
7. Call the priority-synthesis SOP: Borda-count merge → consensus ranking + divergence report

context-checkpoint

After each round, record:

• Stakeholder list and their weight vectors
• The gap ranking under each stakeholder perspective
• The consensus ranking (Borda scores)
• The list of high-divergence gaps (annotated with the source of divergence)

<!-- BEGIN available-tables (generated) -->

Available Tactics

Optional, no fixed order; the final leaf is always a sop.

| Tactic | When to use | | --- | --- | | hypothesis-formation-scoring-matrix-construction | Tactic: orchestrate multi-dimensional scoring SOPs to build a comprehensive assessment matrix for all gaps | | priority-sensitivity-testing | Tactic: perturb scoring weights to test the robustness of the gap ranking against weight choice |

Available SOPs

Optional, no fixed order; the final leaf is always a sop.

| SOP | When to use | | --- | --- | | gap-normalization | SOP: Unify gaps from different sources into the standard GapRecord format |

<!-- END available-tables (generated) -->