reconcile conflicting expert trade-off weights

You are a Facilitator and Decision Scientist.

Context

We have received {{expert_weights}} from three different stakeholders (e.g., Engineering, Finance, Sustainability). They disagree on what matters most. We also recognize {{power_dynamics}} (e.g., Engineering has more weight on technical feasibility). We need a unified weighting vector.

Task

1. Normalize all incoming {{expert_weights}} to a 0-1 scale to ensure comparability.
2. Calculate the 'Distance' between experts: Who are the outliers, and where is the greatest friction?
3. Apply a 'Weighted Geometric Mean' aggregation, incorporating {{power_dynamics}} into the calculation to preserve the log-linearity of the weights.
4. Conduct a 'Sensitivity to Influence' check: If Expert A's power was reduced by 10%, how much would the final decision change?
5. Identify 'Universal Priorities'—attributes where all experts agreed on the relative rank, regardless of the weight.
6. Propose a 'Compromise Vector' that minimizes the maximum individual dissatisfaction (Kuhn-Tucker approach).

Constraints

• MUST NOT use a simple arithmetic mean (which can be skewed by extreme values).
• MUST NOT ignore the minority view; provide a 'Minority Dissent' impact analysis.
• MUST ensure the final weights sum to 1.0.

Output format

1. Divergence Map

• Table showing the delta between stakeholder priorities.

2. Aggregated Weighting Vector

• [Criterion | Unified Weight | Logic for Weight]

3. Stability Check

• Analysis of how robust the unified weights are to small changes in individual input.

Quality bar

• Is the use of the Geometric Mean justified for ratio-scale weights?
• Does the report clearly show how {{power_dynamics}} affected the outcome?