Introduction | Our Goal | Features | Projections | Data Sources | Simulation Model of COVID-19 with System Dynamics | Our model | What the model can do | What the model cannot do | Media Coverage | General Disclaimers
We are engineers and researchers from Greece.
We decided to make a good use of our expertise in Computer Science, Data Analysis and Scientific Modelling in order to contribute to the ongoing research about COVID-19.
We are not doctors neither epidemiologists.
You can find more information about our team members on our social media accounts. Feel free to contact us directly.
Alkaios Sakellaris, Konstantina Miteloudi, Dimitris Sakellaris, Nikolaos Tampouratzis
Our goal is to provide insights and projections about the progression of coronavirus pandemic in a fast, clean and understandable way. We have developed a publicly available tool that helps everyone to comprehend the huge amount of information and numbers. This is necessary because people’s behavior is the main lever for controlling the outcome of the disease.
The main features of the website, are:
The data for analysis and visualization are based on multiple sources and they are classified in two main categories: live data and time series data.
The main sources for live data are: worldometer, BNO and Johns Hopkins University.
The main sources for time series data are: WHO, CDC, ECDC and Johns Hopkins University.
For the analysis, it is also used governments' responses to the coronavirus, per country, from Oxford COVID-19 Government Response Tracker (OxCGRT).
Data may contain errors and inconsistencies as pandemic is in progress.
System Dynamics (SD) is a methodology and mathematical modeling technique to frame, understand, and discuss complex issues and problems.
SARS-CoV-2 pandemic is such a complex problem that makes SD tool a very good fit for modelling the progression.
SD methodology can model not only disease-related variables, but also other variables such as policies, human behavior and economy.
As the scope of this paragraph is to give insights about the framework that is used to create the projections, we will not go deeper into the methodology. We suggest further reading in the dedicated page of System Dynamics Society for COVID-19. Our approach is based on Community Coronavirus Model and discussion about it can be found here.
Every model is wrong. We make this statement firstly because we want to make clear that nobody can model reality. The results of all models must be considered as scenarios.
Every outcome of a model is based on the assumptions of the modeller and the quality of data that are used to derive these assumptions.
The core of our system is based on Susceptible-Exposed-Infected-Recovered (SEIR) epidemiology model. As the following picture depicts, we have made some changes in the basic structure of SEIR. We have split the stock "Infected" in two new stocks, "Infected Symptomatic" and "Infected Sick". Also, we assume that there is a fraction of asymptomatic cases inside the stock "Exposed". These changes were mandatory due to the behavior of SARS-CoV-2.
There are three main loops that change the behavior of the dynamic system.
Red color shows ths reinforcement loop, where the number of past infected people infect others.
Orange color shows two loops:
- the first loop simulates the public isolation mechanism that is established by the Public Health System.
- the second loop refers to the hospital capacity. If the serious cases are more than the hospital capacity of the country that is being examined, then the fatality rate is increased due to overwhelmed, chaotic health care.
Blue color shows the loop that changes the outcomes of imposed mitigation measures. Assuming that the government's measures (social distancing, closed schools, lockdown etc.) changes the relative behavioral risk to a steady number after a specific time, a mechanism of human behavior is activated that is relative to the numbers of Confirmed Observed Cases and Confirmed Deaths that are daily reported. In other words:
- if people see the curves going up, they strict their behavior and they are more alarmed.
- if people see the curves going down, they loosen up their behavior and they are more careless about self-protection measures.
The variables that control simulation are also colored:
Red color shows the variables that characterize COVID-19 or are driven by the behavior of the disease. In order to have consistent simulations among different countries, the same numbers for every country were used based on the research by CDC.
Blue color shows the variables that are country-depended.
Green color shows the variables that control the transmission of the disease.
Our approach is generic and can be applied to many countries, independently of their data.
There are many things left out of the analysis. Some are:
- Immunity of recovered people: we don’t know if recovered people have immunity to COVID-19 or the population percentage that may have immunity. There is ongoing research in this field. Future models may address this issue, but for now, there is no loop-back from the stock "Recovered" to the stock "Susceptible".
- Economy: the effects and the pressure from downgrade economies has been left out. These are political decisions that can be incorporated in the model implicitly with the variables that include "Measures Time".
- New drugs that lower the fatality rate: at the time of writing this report, there is no drug that has solid effects on the disease.
- Rates of the disease based on age cohort.
Also, we have not included in our results any uncertainty of the projections. We want to emphasize that the results are one simulation run and they are not prediction neither forecast. To assess uncertainty bounds we must do Monte Carlo Simulation that generates thousands of scenarios. This type of analysis needs time and equipment that we don't possess for the time being.
We want to thank the following news agencies for sharing our project with the world:
This site contains Work in Progress (WIP) results. No rights can be derived from any of the results.
Information on this web site is provided "as is" without warranty of any kind, either express or implied about the completeness, accuracy, reliability, suitability or availability with respect to the web site or the information, products, services, or graphics contained on the web site for any purpose including, but not limited to, the implied warranties of merchantability, fitness for a particular purpose, or non-infringement. Some jurisdictions do not allow the exclusion of implied warranties, so the above exclusion may not apply to you.
You expressly understand and agree that COVID-Analysis shall not be liable to you for:
a) any direct, indirect, incidental, special, consequential or exemplary damages which may be incurred by you, however caused and under any theory of liability. This shall include, but not be limited to, any loss of profit or revenue (whether incurred directly or indirectly), any loss of goodwill or business reputation, any loss of data suffered, cost of procurement of substitute goods or services, or other intangible loss;
b) any direct, indirect, incidental, special, consequential or exemplary damages arising out of or related to COVID-Analysis web site or the information contained in it, whether such damages arise in contract, negligence, tort, under statute, in equity, at law or otherwise;
c) any loss or damage which may be incurred by you, including but not limited to loss or damage as a result of:
any reliance placed by you on the completeness, accuracy, reliability, suitability or availability with respect to the information on the web site and the services provided;
any changes which COVID-Analysis may make to the information or services, or for any permanent or temporary cessation in the provision of the information or services (or any related features);
COVID-Analysis takes no responsibility for and will not be liable for the web site being temporarily unavailable.
Links provided on this site are provided solely as a convenience to you and the provision of any such link does not constitute our endorsement of that site or its provider or of any of the content, products, or services contained or offered therein. Υour use of each of those sites is subject to the conditions, if any, that each of those sites has posted. Υou agree that we are not responsible for the accuracy, copyright compliance, legality, decency, or any other aspect of the contents, products, services or any transmissions received through such sites. you further agree that we have no liability whatsoever from such third party sites and your usage of them.
COVID-Analysis is not directly or indirectly owned or managed by any governmental organization, public charity, non-profit, or private foundation. COVID-Analysis is not affiliated, associated with, or in any way has any official status with the United Nations or any other official organization.
Information on this web site may contain inaccuracies or errors. Information may be changed or updated without notice. COVID-Analysis may also make improvements and/or changes in the content, the products, and/or the services at any time without notice. COVID-Analysis takes absolutely no responsibility for any errors on the site. Any reliance you place on the information is therefore strictly at your own risk.