A Summary of Operations Research and Industrial Engineering Tools for Fighting COVID-19

Operations Research and Industrial Engineering (OR & IE) approaches are widely used and play important roles in improving the design and operations of many standard corporate activities such as supply chain management, job/staff scheduling, vehicle routing, facility location, and resource allocation. In the midst of the COVID-19 pandemic, policymakers, companies, community workers and individual households have been designing new systems and procedures to fight the virus. Many problems related to optimizing these systems and their operations can be tackled by extending the traditional OR & IE approaches with new objectives, constraints, and input data. The purpose of this document is to summarize potential scenarios one may encounter during the prevention, disease control, intervention, and recovery phases during COVID-19 outbreaks, and point out the OR & IE models that can be applied for solving the related problems. We are not medical experts and thus will not focus on the drug & vaccine discovery, nor analyzing the disease transmission rate and its spread patterns. Instead, we consider decisions made by multiple stakeholders that can prepare for rare but catastrophic events such as the COVID-19 pandemic, can better inform the public to perform “social distancing”, can better utilize resources and ensure medical supplies during the outbreak, and can improve the quality of life and work to mitigate economic losses.

The remainder of this report is organized as follows. In Section 1, we describe the overall lessons learned from countries at different disease phases and summarize their effective/ineffective procedures when fighting the COVID-19. In Section 2, we focus on the disease prevention phase and describe what models can be applied to set up enough resources. In Section 3, we shift to problems related to controlling the disease during an outbreak and examine different industries that could be impacted due to diverse control policies. In Section 4, we consider the recovery phase and use today’s China as an example to review possible issues and procedures for avoiding the rebound of infection numbers. Lastly, in Section 5, we conclude our report and list useful online resources for keeping up with data, guidelines, status update about the COVID-19 pandemic.

1. Overall Lessons

As of March 24, 2020, the COVID-19 remains a global health crisis of grave and uncertain magnitude. We are at an unforeseen and unprecedented moment, and there is no doubt that fighting the disease requires global cooperation and teamwork. Figure 1 shows different stages (elaborated on the left-hand side of the figure) that major countries are going through and their total confirmed cases as of March 16, 2020.

Figure 1: COVID-19 phases around the world (Source: McKinsey & Company)

Similar to Figure 1, one can find many articles online related to observation and prediction about the COVID-19 spread in different countries, but we found few on how to apply mathematical modeling and analysis to make more responsible and effective decisions. For example, the following two well-disseminated articles published in Medium.com definitely raise public awareness about the severity of COVID-19 spread and about `social distancing’ to hopefully `flatten the curve.’

Up to this point, most of us are probably familiar with this `Flatten the Curve’ picture which shows that intervention decisions can help with decreasing the total patient number and avoiding overwhelmed healthcare systems.

We, on the other hand, aim to provide descriptions about decision-making problems related to disease prevention, intervention, control and recovery and how to solve them using OR & IE tools. In other words, these models will allow policymakers, corporates, and other decision-makers to jump from the observations to solutions, and from data to actions.

Since March 18, 2020, China has zero new local confirmed cases nationwide including in the city of Wuhan. A huge price is taken by shutting down all the economies for two months to fight COVID-19 spread. We summarize a few key lessons learned from China’s lockdown and emergency response as follows, although not all of them can be borrowed by Western countries due to cultural differences.

  1. Enhancing community-based control of self-quarantine; tracking the paths of disease spread; warning people with a potential high risk of infection.

At the beginning of the COVID-19 outbreak, the Chinese suffered (e.g., having 5000+ new confirmed cases in Wuhan and 300+ death tolls daily for many days) from not implementing some of the above lessons learned. As of March 24, 2020, China has 81807 cumulative total confirmed cases, 3283 death tolls, while there are a total of 268,239 number of confirmed cases and 13,899 death tolls outside China. The latter happened quickly around the world since the very first patient was diagnosed in Italy in late February, and several outbreaks meanwhile appeared also in Iran and in South Korea. In the US, the daily newly confirmed cases have been skyrocketing in the past week, ranging from 4000 to 8000 per day. The total cumulative confirmed cases are 46,556 in the US and the death toll is 592 as of March 24, 2020. However, only 303,014 Americans have been tested. (All the figures are based on WHO, CDC and other online resources we later summarize in Section 5.1.)

We refer to the online google doc or this PDF file for the details of the following sections, where we introduce OR & IE models for decision-making problems that can flatten the curve, centering around the above lessons.

2. Disease Prevention Phase

2.1 Testing Facility Location Design and Testing Kit Distribution

2.2 Healthcare and Other Relief Resource Preparation

3. During Disease Outbreak

3.1 Increasing Testing Ability and Information Transparency

3.2 Lockdown and Quarantine

3.3 Hospital System Reform and Operations

3.3.1 Patient Triage and Admission Control

3.3.2 Capacity Allocation in ICUs and Different Types of Wards

3.3.3 Locating Ambulances and Their Dispatch

3.3.4 Telemedicine Service Scheduling

3.3.5 Medical Home Care Delivery

3.4 Supply Chain Management of Food and Essential Items

3.5 Online Retailing and Grocery/Medicine Delivery

3.6 Airline Fleet Rescheduling, Call Center Staffing, and Airport Screening

3.7 Work from Home — Allocating Enough Bandwidth

3.8 Impact on the Low-income and Underserved

4. Recovery and Post-Recovery Planning

5. Conclusion and COVID-19 Resources

In this article, we summarized OR & IE models that have been developed for corporate design and operations management and explained how they can be applied and extended to solve COVID-19 related problems. We believe that with proper use of these models and efficient algorithmic design, we can allocate the most critical resource to the most vulnerable of the overall disease control process. It will allow policymakers to act more quickly and accurately to react to the pandemic.

“It takes 67 days to reach the first 100K confirmed COVID-19 cases worldwide, 13 days to reach the next 100K confirmed cases, and then 4 days to reach another 100K cases.” said by Dr. Tedros Adhanom Ghebreyesus, Director-General of the World Health Organization. Indeed, effective control and international cooperation are the keys to fight the virus and we hope that this document can provide references for guiding the needed system reform, redesign, and operations.

COVID-19 Data and Cases

Below we also summarize repositories of COVID-19 data, parameter estimates, cases, and tools for analyzing the data that we found online while we search for the OR & IE literature.

https://news.qq.com/zt2020/page/feiyan.htm#/global

https://news.ifeng.com/c/special/7tPlDSzDgVk

https://feiyan.wecity.qq.com/wuhan/dist/index.html#/

  • A tool to visualize confirmed cases and their statistics (e.g., distributions by age, gender, nationality, or infection sources). As of March 21, 2020, Singapore only has 432 total confirmed cases and 2 death tolls.

https://co.vid19.sg/dashboard

  • An iterative model to visualize how intervention procedures can alter possible disease spread patterns and the number of infections.

http://gabgoh.github.io/COVID/index.html

It contains information and resources for COVID-19 modeling research, including a list of dashboard and visualization software, COVID-19 data and parameter estimates.

  • Some links of data repositories for COVID-19 related patient data.

https://github.com/CSSEGISandData/COVID-19 (by Johns Hopkins University)

https://github.com/pcm-dpc/COVID-19 (by Italy government)

https://datarepository.wolframcloud.com/resources/Patient-Medical-Data-for-Novel-Coronavirus-COVID-19

https://datarepository.wolframcloud.com/resources/Epidemic-Data-for-Novel-Coronavirus-COVID-19

Acknowledgment: Greatest appreciation to my former and current PhD students (http://www-personal.umich.edu/~siqian/group.html) for discussing and contributing ideas, recommending references.

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Siqian Shen

Mom of two girls, Ironman 70.3 finisher, Associate Professor at the University of Michigan