How do demand shocks caused by natural disasters impact store inventories? In order to prepare for such events, retailers must stock up on the appropriate products that will be in high demand while distributing less-contested items to stores that remain unaffected. HurrEcon is an application that allows its users to visualize and predict the data needed for such preparedness.

What it does

HurrEcon displays the data of Tractor Supply Co. stores around the country, including each of the store's most popular items and how a hurricane would affect its sales. It pulls live hurricane data form the NOAA (National Oceanic and Atmospheric Administration) servers and combines this with data from past hurricanes to model their impacts on consumer demand for the specific products that would become popular within a store. We can then use this information to predict the effect of future hurricanes on nearby regions in a store-by-store basis.

How I built it

Using data from the 2018 Atlantic hurricane season and the dataset provided by Tractor Supply Co., we were able to track how the items bought by consumers shifted during the course of a natural disaster. After parsing the data to a readable (and more explicit) format through Python scripts, we used OpenLayers to depict the map data of each store as well as reported hurricane location throughout the year.

Share this project: