Code Against the Flames

Inspiration

Wildfires are a growing threat, and while firefighters are the frontline defense, technology can enhance response efforts. This project uses AI and automation to optimize wildfire response and predict future outbreaks, ensuring efficient resource deployment.

What It Does

Real-Time Firefighter Deployment Optimization (Challenge Part 1)

Assigns firefighting units based on fire severity and resource availability.
Reduces operational and environmental damage costs.
Flags fires that cannot be addressed immediately and estimates potential damage.

AI-Powered Wildfire Prediction (Challenge Part 2)

Analyzes historical wildfire data to forecast fire outbreaks.
Evaluates environmental factors like weather, vegetation, and human activity.
Generates GPS locations of potential fires and visualizes them on a map.

Emergency Email Alerts

Sends fire prediction updates to emergency responders.
Enables quick decision-making for firefighting teams.

How We Built It

Used machine learning (RandomForest, SciPy) for fire risk prediction.
Developed a Streamlit dashboard for real-time visualization.
Integrated an SMTP-based email alert system.
Designed an optimized resource allocation algorithm.
Analyzed wildfire data from 2020 to 2025.

Challenges

Balancing cost constraints while prioritizing high-severity fires.
Merging large datasets for accurate predictions.
Ensuring API key security and preventing hardcoded credentials.
Creating clear and actionable visualizations.

Accomplishments

Built an AI-driven wildfire prediction model.
Learned how the random forest algorithm works and how to implement it in Python.
Developed a system for optimized firefighter deployment.
Implemented real-time emergency email alerts.
Created an interactive fire risk map.