Capstone Project
TECHNOLOGY CATEGORY
Machine Learning / AIINDUSTRY SPONSOR
MicrosoftPROBLEM SPACE
Enviromental ScienceDATE COMPLETED
December 15, 2023
Back to ProjectsForeFlame
Problem
Wildfires have an important role in nourishing, and enriching life in the forest. However, when uncontrolled, wildfires can cause significant and irreversible loss in terms of land, people, and the economy.
Approach
The team started by interviewing 6 experts and performing a field study with fire managers and wildfire firefighters. After narrowing their scope and identifying a set of product requirements they then created a prototype that was evaluated by 7 participants including a meteorologist, a RAWS program manager, and a fire regulation program manager.
Solution
ForeFlame is a cloud-based Wildfire Occurrence Analysis and Modeling system that is powered by Machine Learning. This product integrates data from NOAA, NDVI, LANDFIRE, and historical wildfire data to create a DNN model that provides five days prediction results, accessible through a web based dashboard.
