COVID-19 Modeling and Decision Trees for San Diego Schools

About the Project

A key concern amid the COVID-19 pandemic is the safe return of children, teachers, staff to schools.

GeoACT (Geospatial Agent-based Model for COVID-19 Transmission) simulates a range of scenario data to help schools evaluate and improve their reopening plans to prevent super-spreader events and outbreaks.

image

Features

image

Seating Arrangements

image

Bus Routes

image

Transmission Dynamics

image

Testing and Vaccination

Learn More

Models

image

Bus Model

Along with integrating our model into the school ecosystem, GeoACT extends its capability to activities beyond school-hours. By accounting for transmission, ventilation in confined spaces and using real-time case data on specific routes, GeoACT successfully simulates the spread of COVID-19 aboard school buses.

Learn More

School Model

Our school model takes user-input school-specific data such as floor-plans, class schedules, furniture layouts, as well as planned pharmaceutical and non-pharmaceutical interventions to estimate the extent of COVID-19 transmission in specific schools. The GeoACT model then blends this with the best available science governing COVID-19 transmission dynamics to provide users an estimate of the extent of COVID-19 transmission as well as the spatial distributions of case loads in specific schools.

Learn More
image

News and Events

Get to know all things GeoACT

image

Model Overview at Halicioğlu Data Science Institute

View Video
image

How to use GeoACT for your School?

View Video
image

A Deeper Dive into the Science

Read Paper
image

Results

Example Results

Meet Our Team

Project Founders:

Kaushik Ganapathy

Johnny Lei

Eric Yu

Bailey Man

DSC 180 Capstone Project Section (Fall 2020 - Winter 2021,
mentored by Dr. Ilya Zaslavsky, San Diego Supercomputer Center):

Kaushik Ganapathy

Johnny Lei

Eric Yu

Bailey Man

Akshay Bhide

Evan Price

Farhood Ensan

Areeb Syed

Michael Kusnadi

Bernard Wong

Songling Lu

Ziqian Cui

Student interns:

Stephen Gelinas

Laura Diao

Alice Lu

Saarth Shah


NSF support (award 2139740) is gratefully acknowledged