NH25_Flux_Atlas is a sophisticated economic simulation platform that models resource trading between countries using psychologically-informed bidding behaviors and cluster-based auction mechanisms.

• Dynamic resource trading simulation
• Cluster-based auction system
• Psychologically modeled bidding behavior
• Real-time API integration
• Interactive bidding interface
• Running a python based simulation model

The simulation uses a modified Laplace distribution to model bidding behavior based on supply-demand dynamics. Our bidding function combines market dynamics with psychological distance decay:

$$ v_{value} = base_{price} \cdot (1.0 + max_{increase} \cdot w) $$

where:

$$ max_{increase} = max(0.0, min(1.0, \frac{demand}{supply} - 1.0)) $$ $$ w = e^{-\frac{|quantity - demand|}{b}} $$ $$ b = max(|demand| \cdot k, min_b) $$

The key components are:

• w: Laplace decay factor (psychological willingness to deviate from ideal quantity)
• b: Scale parameter controlling bid sensitivity
• distance: Absolute difference between offered quantity and demanded quantity
• k: Sensitivity multiplier for demand-based scaling
• min_b: Minimum scale parameter to prevent division by zero

v_value = base_price * (1.0 + max_increase * w)
where:
- max_increase = max(0.0, min(1.0, (demand/supply) - 1.0))
- w = exp(-distance/b)  # Laplace decay
- b = max(abs(demand) * k, min_b)  # Sensitivity
- distance = |quantity - demand|

After countries compute their bid values using the Laplace-based psychological model, the auction follows a Vickrey (second-price sealed-bid) mechanism:

1. Each country submits their sealed bid v_value computed from the Laplace function
2. The highest bidder wins the auction
3. The winner pays the second-highest bid price (or base price if single bidder)

This mechanism is chosen because:

• It incentivizes truthful bidding - countries' best strategy is to bid their true valuation
• Reduces strategic manipulation - winning price depends on others' bids
• Balances market efficiency with fair pricing
• Prevents winner's curse and aggressive overbidding

Mathematically: $$p_{final} = \begin{cases} v_{2nd\text{-}highest} & \text{if multiple bids} \ base_{price} & \text{if single bidder} \end{cases}$$

Where $p_{final}$ is the final price per unit that the winning country pays.

Our simulation implements a sophisticated cluster-based distribution system that ensures fair market access across different economic tiers:

Initial Quantity: Q
Round 1: Q/2
Round 2: Q/4
Round 3: Q/8
Round 4: Q/8 (remainder)

This halving mechanism serves multiple purposes:

• Allows countries with smaller demands to participate in later rounds
• Prevents market monopolization by larger economies
• Creates multiple opportunities for price discovery
• Ensures efficient resource distribution across all economic clusters

Countries are grouped into economic clusters (e.g., Emerging Markets, Developing Nations, High-Income Nations). Each cluster:

• Receives a proportional allocation of the auctioned resource
• Participates in independent parallel auctions
• Competes within their economic peer group
• Has equal access to each batch round

Countries decide to sell based on several factors:

1. Supply-Driven Probability: Higher resource supply increases likelihood of initiating auction

   auction_probability = min(0.3, supply_amount / total_world_supply)

2. Conservative Quantity Selection: Countries offer 10-30% of their supply

   auction_quantity = supply * random.uniform(0.1, 0.3)

3. Multi-Round Strategy: Resources not sold in one round cascade to next batch

The bidding strategy incorporates demand-matching and quantity optimization:

1. Demand Alignment: $$bid_urgency = \frac{|quantity - demand|}{demand}$$

   • Lower difference → Higher bid probability
   • Perfect quantity match maximizes bid value

2. Supply-Demand Ratio Impact: $$bid_modifier = \frac{demand}{available_supply}$$

   • High demand/low supply → Aggressive bidding
   • Low demand/high supply → Conservative bidding

3. Quantity Sensitivity:

   • Large quantity mismatch reduces bid probability
   • Small batches attract countries with precise demands
   • Batch size reduction increases participation from smaller economies

This multi-faceted approach creates a dynamic market where:

• Resources are distributed efficiently across economic tiers
• Countries participate based on realistic economic behaviors
• Market access is democratized through the batch system
• Price discovery occurs at multiple levels

Future enhancement planned for incorporating additional economic indicators:

• GDP influence
• Political relations
• Transportation costs

• Python 3.12+
• Node.js 18+
• PostgreSQL 14+

# Clone repository
git clone https://github.com/your-repo/NH25_flux_Atlas.git
cd NH25_flux_Atlas

# Create virtual environment
python -m venv env

# Activate virtual environment
# Windows:
env\Scripts\activate
# Linux/Mac:
source env/bin/activate

# Install dependencies
pip install -r requirements.txt

# Setup environment
cp api/.env.example api/.env
# Edit .env with your database credentials

# Start API server
cd api
python3 run.py

cd frontend
npm install
npm run dev

NH25_flux_Atlas/
├── api/                  # FastAPI backend
│   ├── app/             # API implementation
│   └── run.py           # API entry point
├── auction/             # Auction logic
│   ├── auction.py       # Base auction classes
│   └── turn_based_simulation.py  # Turn-based logic
├── frontend/           # React frontend
├── models/             # Data models
└── requirements.txt    # Python dependencies

Once running, API documentation is available at:

• Swagger UI: http://localhost:8000/docs
• ReDoc: http://localhost:8000/redoc

python3 auction_manager.py

from auction.auction_manager import run_bidding_simulation

run_bidding_simulation(
    bidder_country=japan,
    seller_country=russia,
    resource_name="PETROLEUM",
    total_quantity=50.0,
    base_price=0.5
)

Please read CONTRIBUTING.md for details on our code of conduct and the process for submitting pull requests.

• Aarush Bhat
• Cong Thinh Nguyen
• Sashreek Addanki
• Neamur
• Hamse
• Jacob
• Cherie

See the LICENSE file for details.

• NeurAlbertaTech Hackathon 2025