Predict and Contain

flame start
crossing the road
spread
burnt road + burnt forest
road block
burnt buildings + crossing the river
uh oh

Inspiration

While brainstorming ideas, we stumbled upon John Conway’s Game of Life and the concept of Cellular Automata. This sparked the idea of using similar principles to model natural phenomena. Given the increasing frequency and severity of wildfires, we were inspired to create a simulation that leverages the Game of Life mechanics to predict and visualize wildfire spread. The goal was to develop a tool that could demonstrate how fires propagate in different environments and help with better disaster response and planning.

What it does

Our project is a real-time fire simulation model that uses a grid-based map to visualize and forecast the spread of fire by adjusting environmental factors like wind speed, fuel type, and terrain, the simulation provides insights into fire behavior, helping users understand where and how fast a fire might move and when it could potentially be contained.

How we built it

We used python to create a 2D grid-based simulation that models fire spread using cellular automata. Each grid cell represents a node with the letter of each node representing its land type (forest, road, etc.) and the color representing its state (not burned, burning, etc.). The simulation is visualized in real-time using Matplotlib, allowing users to see both the spread of fire and affected areas.

Challenges we ran into

We had difficulties setting up the simulation's real-time visualization and managing the spread of fire with different environmental factors.

Accomplishments that we're proud of

We're proud of successfully building a functional firespread model that can be accurately visualized in real time and reacts with the different conditions and environmental factors in place. It was rewarding to see the simulation visually represent how fires might spread under different conditions, showcasing the practical impact of our project.

What we learned

We gained a deeper understanding of cellular automata, real-time data visualization, and the challenges of modelling natural phenomena. Working with various aspects of the program at the same time also strengthed our collaborative and problem-solving skills.

What's next for Predict and Contain

In the future, we plan to enhance the realism of our simulation by incorporating real-world maps and geographic data. This will allow us to model specific terrains and environmental conditions, making the tool more practical for real-world applications. By integrating these elements, Predict and Contain could support emergency response teams in analyzing actual wildfire risks and planning containment strategies more effectively.