Grant Allocation System

This project simulates the process of evaluating organizations for grant funding and fairly distributes funds based on their expected utility scores. It consists of two main scripts:

1. generate_csv.py - Generates randomized grant requests, evaluator budgets, and evaluation data.
2. allocate_grants.py - Allocates funds to organizations based on their computed utility scores.

How It Works

1. Generate Data

Run this command to generate sample data:

python generate_csv.py

This creates three CSV files:

• evaluations.csv: Scores given to organizations by evaluators.
• organizations.csv: Organizations and the amount of funding they request.
• evaluators.csv: Evaluators and their available budget.

2. Allocate Grants

Once the data is generated, run:

python allocate_grants.py

This script:

• Computes an Expected Utility (EU) score for each organization based on evaluator ratings.
• Distributes total available funds proportionally based on EU scores.
• Caps grants at the requested amount to prevent overfunding.

The final grant allocations are saved in grant_allocation.csv.

Theory Behind the Allocation

Grants are allocated using a weighted evaluation system, where:

• Evaluators rate organizations on impact, feasibility, and cost-effectiveness.
• Evaluators also rate their confidence in these scores.
• A weighted sum is calculated, with weights:
  • Impact: 40%
  • Feasibility: 30%
  • Cost-effectiveness: 30%
• Scores are adjusted for evaluator confidence, so confident evaluators have more influence on EU scores.
• The total available funds are distributed proportionally based on the EU scores.
• A cap is applied to ensure no organization receives more than it requested.

This method ensures that organizations with the highest evaluated utility receive funding while preventing excessive allocation.