PRIORITY_ENGINE // AHP-OCI

A structured decision-support framework integrating causal modeling with multi-criteria prioritization.

The Priority Engine (formerly GovernanceOS) is a React-based decision intelligence tool designed for public administrations and mission-driven teams. It synthesizes the Outcome–Cause–Intervention (OCI) causal structure with the Analytic Hierarchy Process (AHP) to prioritize interventions based on mathematical utility rather than intuition alone.

// Methodology

The system imposes a strict three-tier ontology to ensure traceability:

1. Outcome (Tier I): The singular strategic objective (e.g., "Reduce Rural Depopulation").
2. Drivers (Tier II): Probabilistic causal factors influencing the Outcome.
3. Interventions (Tier III): Discrete operational programs mapped to specific drivers.

Prioritization is calculated using Eigenvector Centrality derived from pairwise comparison matrices (Saaty, 1980). The final utility score ($S_i$) for an intervention is calculated as:

$$ S_i = Coverage \times Effectiveness \times Feasibility $$

// Features

• Swiss Style Design System: A rigorous, high-contrast UI based on International Typographic Style principles (Inter / JetBrains Mono).
• Gemini AI Integration: Automated discovery of causal drivers and strategic interventions using Google's Gemini 2.5 Flash model.
• AHP Computation Engine: Client-side calculation of Eigenvectors, Consistency Ratios (CR), and geometric mean aggregation.
• Interactive Visualization: Force-directed logic graphs using D3.js and priority matrices using Recharts.
• Privacy Focused: API keys are processed client-side; no data is persisted to external servers.

// Tech Stack

• Core: React 19, TypeScript
• Styling: Tailwind CSS (Custom "Swiss" config)
• Math/Logic: Custom AHP implementation (Power Method / Geometric Mean)
• AI: Google GenAI SDK (@google/genai)
• Viz: D3.js, Recharts

// Setup & Installation

1. Clone the repository

   git clone https://github.com/gbrlpzz/OCI-AHP_Priority_Engine.git
   cd OCI-AHP_Priority_Engine

2. Install dependencies

   npm install

3. Environment Configuration Create a .env file in the root directory and add your Google Gemini API key:

   API_KEY=your_google_api_key_here

4. Run the development server

   npm start

// License

Distributed under the MIT License. See LICENSE for more information.

// Credits

Concept & Engineering: Gabriele Pizzi Studio: Opera Incerta Correspondence: [email protected]

Based on the Analytic Hierarchy Process (AHP) methodology by Thomas L. Saaty.

---

Copyright © 2025 Gabriele Pizzi // Opera Incerta