Analyzing NJ COVID-19 responses with Control theory, extending a study from earlier this year.

Submission Number: 75
Submission ID: 107
Submission UUID: 6c43287b-1d1a-4de4-bc1d-360b517bc3af
Submission URI: /form/project

Created: Mon, 11/02/2020 - 16:51
Completed: Mon, 11/02/2020 - 17:21
Changed: Tue, 08/02/2022 - 15:04

Remote IP address: 73.10.232.250
Submitted by: Sean McQuade
Language: English

Is draft: No
Webform: Project
Analyzing NJ COVID-19 responses with Control theory, extending a study from earlier this year.
CAREERS
image for CAREERS.PNG
matlab (2), programming (5), programming-best-practices (49)
Complete

Project Leader

Benedetto Piccoli
{Empty}
(856) 225-6356

Project Personnel

Galen Collier, Sean McQuade
Ryan Weightman
{Empty}

Project Information

The COVID-19 pandemic has infected every continent except Antarctica. Loss of life and crippling economic effects have been felt across the world. COVID-19 infection rates can be reduced through the use of testing, social distancing, and contact tracing. Each of these strategies requires a range of costs to apply. We use control theory and numerical optimization techniques with data from a variety of NJ-based sources to estimate how much each of each strategy should be applied within counties across the state in order to minimize the total cost. A student facilitator working on this project will help introduce the lab to the use of MATLAB for existing AMPL-based simulations. The student will use general computational skills and mathematical modeling techniques which will provide valuable experience with tools implemented by computational researchers today across many fields.

Information about the research associated with this project: https://rand.camden.rutgers.edu/2020/03/26/timing-county-hospital-bed-shortfall-during-covid-19/

Information about the Piccoli Lab: https://piccoli.camden.rutgers.edu/

Project Information Subsection

A paper will be written and published in early 2021.
{Empty}
Advanced study in mathematics / statistical modeling
AMPL
MATLAB
Interest in computational modeling for urgent, real-world problems
{Empty}
Practical applications
{Empty}
Rutgers-Camden
303 Cooper St, Camden, NJ 08102
Camden, New Jersey. 08102
CR-Rutgers
11/16/2020
No
Already behind3Start date is flexible
6
02/10/2021
{Empty}
08/11/2021
  • Milestone Title: Initial feasibility study and review of work leading to this project
    Milestone Description: Initial feasibility study examining existing models and tools that may be used for this project: AMPL API for MATLAB and the MATLAB – Optimization Toolbox
    Completion Date Goal: 2021-02-10
    Actual Completion Date: 2021-02-10
  • Milestone Title: NJ COVID control model paper
    Milestone Description: We will leverage simulations run by the student to complete a publication that is planned to be submitted to the journal "Networks and Heterogeneous Media" (NHM).
    Completion Date Goal: 2021-03-10
    Actual Completion Date: 2021-03-10
  • Milestone Title: COVID control with Herd Immunity
    Milestone Description: We will leverage our model with recent publications about herd immunity for a combined study
    Completion Date Goal: 2021-04-30
  • Milestone Title: Expand results to US
    Milestone Description: We will do a similar study to the NJ one but consider all US
    Completion Date Goal: 2021-05-15
  • Milestone Title: Model Vaccine Deployment strategy
    Milestone Description: We will study the optimal way to distribute the vaccine with computational tools.
    Completion Date Goal: 2021-05-30
{Empty}
{Empty}
A paper was submitted to Mathematical Models and Methods in Applied Sciences (M3AS) May 26, 2021.
Title: "Control of COVID-19 outbreak using an extended SEIR model."
Authors: Sean T. McQuade, Ryan Weightman, Nathaniel J. Merrill, Aayush Yadav, Emmanuel Trélat, Sarah R. Allred, Benedetto Piccoli
Interface between two different programming languages: AMPL and MATLAB (i.e., solving problems using the AMPL API for MATLAB).
{Empty}
They will see how their support can assist the ongoing COVID-19 struggle.
{Empty}
{Empty}

Final Report

This project builds on the standard SEIR disease spreading model. It shows that we can use control theory techniques with SEIR to estimate some key aspects of the pandemic, and evaluate the various responses used to control the spread.
Our project has many uses, including optimizing for the best vaccine distribution plan. This was not obvious when we started.
No.
Yes, our project shows one way to evaluate the non-pharmaceutical interventions to fight COVID-19. Similar methods will expand upon this in the future.
Yes, these simulations benefit from fast computing, such as using Amarel, or other RU-Camden resources.
Yes, cluster computing may become more valuable as more tools that are similar to our simulators are implemented.
No.
Yes, this has the capability to point us toward the good ways we halted the spread of COVID, and point out the less efficient ways as well. I expect this project, and others that are similar will help us handle the next pandemic more efficiently.
Projects like this are based on many assumptions that are coded into simulators. Other experts in related fields are a valuable resource to make those estimates. A project such as this is ideal for getting many experts of different fields together.
The project was a success, not just in meeting our goals, but also in providing growth to the students, and allowing them to take the initiative to answer research questions.