Modeling resource consumption in the US air transportation system via minimum-cost percolation

This repository contains the code for

• Minsuk Kim, C. Tyler Diggans, and Filippo Radicchi, Modeling resource consumption in the US air transportation system via minimum-cost percolation, (Journal Information to be added)

• Preprint (arXiv)

• BibTex entry:

  @article{kim2025modeling,
  title={Modeling resource consumption in the US air transportation system via minimum-cost percolation},
  author={Kim, Minsuk and Diggans, C Tyler and Radicchi, Filippo},
  journal={arXiv preprint arXiv:2504.04245},
  year={2025}
  }
  

How to retrieve the publicly available data

Our work rely on publicly available data from various sources:

• Airline On-Time Performance Data (flight schedule data)
• DB1B Data (sold tickets data)
• FAA Aircraft Database
• OpenFlights Database
• Census Population Data

Installation and Requirements

Environment Setup

There are two ways to set up the environment:

Option 1: Using environment.yml (Recommended)

# Create and activate conda environment with all dependencies
conda env create -f environment.yml
conda activate mcp-us-airline

# Install the project package
pip install -e .

Option 2: Manual Installation

# Create and activate a new conda environment
conda create -n mcp-us-airline python=3.10
conda activate mcp-us-airline

# Install the project package with all dependencies
pip install -e .

System Requirements

• Python 3.10 or higher
• C compiler (gcc recommended) for MCP model compilation
• At least 32GB RAM recommended for large dataset processing

How to generate the input files for the minimum-cost-percolation (MCP) model

The raw data retrieved from public sources should be preprocessed before running the MCP model. We provide a Jupyter notebook file generate-input-files.ipynb which does the job. To properly run this notebook, you need to install this library mcp_us_airline from this repository.

How to run the MCP model?

We provide a makefile that compiles the source code of the MCP model, which is written in C language. You can compile it by typing:

make

To execute the model:

./min_cost_percolation.out <DEMAND> <FLIGHTS_INFO> <FCN> <OUTPUT_FILE> <COST>

Required arguments:

• <DEMAND>: File containing origin-destination demand data
• <FLIGHTS_INFO>: File containing flight details for the flight-connection-network (FCN)
• <FCN>: Adjacency list representation of the FCN
• <OUTPUT_FILE>: Name of the file where results will be saved
• <COST>: Cost function to minimize (integer 1-3)
  • 1: Length
  • 2: Duration
  • 3: Seat-availability-cost

Example usage:

./min_cost_percolation.out demand.txt flights.txt network.txt results.txt 1

Misc.

We also provide Jupyter notebook files which allow you to generate the figures in the main text.