Our interactive app


When the most critical phase of the pandemic passes, the main issue will be deciding how to ease lockdown and allow people to get back to their activities. The case of Hong Kong suggests that the release cannot be immediate and a policy of progressive release of the lockdown is required. This scenario is unprecedented and decision-makers lack tools to quantitatively assess the consequences of policies.

What it does

A partial lockdown means we can send only part of the population out at a given time, reopening progressively businesses and institutions, so that the epidemics stay under control, keeping the transmission at levels that avoid getting our hospitals overwhelmed. So how, do we do put that in place?

We are developing a web app tool to : 1) to test and visualize the effects of different lockdown release strategies,  2) calculate the best release strategy depending on different economic and sanitary objectives

The tool is powered by a study that builds upon existing and proven epidemic models in order to take into account the economic and sanitary consequences of different policies to release the lockdown.  Very concretely, the web app allows users to interactively select different dates for re-opening schools, businesses, quarantines, etc; and observe the prediction of the evolution of the number of patients infected, in intensive care units, or deceased.

How we built it

With Python, some epidemiological models from the literature, math, control theory and a lot of creativity.

Challenges we ran into

The major challenges were raised in estimating how different policies impact model parameters, and thus the final outcome of the simulation will be sensible to assumptions. The trading-off, quantifying and comparing the economic and health implications was a difficult task.

Accomplishments that we are proud of

We not only successfully predict pandemic spread, but more than that we suggest optimal policy and decision processes such that we let people out of lockdown as early as possible, reducing the impact on the economy while making sure our healthcare system is not overwhelmed. This analysis is absolutely crucial and useful to policymakers. Our work enables policymakers to make the best decision for a variety of possible scenarios.

What we learned

First, having a precise model is absolutely difficult to obtain in the absence of reliable data. Second, indeed, how to let as many people out without making healthcare systems overloaded is a very delicate and challenging decision. As a team, we notice, communication, proper task distribution and dedication were very crucial for the success of this project.

What's next for See You Outside

We will happy to help policymakers to give the insight to tackle the lockdown situations and come up with the best policies. Furthermore, our current model is macroscopic and is done for entire Switzerland, This can easily be divided into region and canton wise analysis. Since our analysis is for Switzerland, however, it is not limited to Switzerland, and therefore, many countries and regions across the globe can use it in this challenging time.

Try It out



control, markov-chain, python

Devpost Software Identifier