At the onset of an epidemic outbreak, there are usually no vaccines to prevent the disease; it takes time to develop them. The only resort to keep the infected cases to a manageable level then is to devise efficient non-pharmaceutical interventions (NPIs) such as testing, travel restrictions, social distancing, and sanitary measures. If implemented correctly and on time, such interventions enable the healthcare infrastructure to function without getting overwhelmed. However, strict NPI policies can be detrimental not only to the economy but also to society. It is, therefore, crucial to develop models and methodologies that enable optimal policies for epidemic mitigation with minimum socio-economic consequences.
In this talk, I will present our recent research on epidemic control.
First, I will talk about epidemic suppression through testing. It is well-known that testing is the most effective control mechanism for epidemics as it allows the authorities to detect and isolate the infectious cases, thereby breaking the chains of infection transmission. However, given the epidemic situation and a goal to control it, the problem of computing minimum tests to be performed per day is not well-studied. I will present a simple method for computing the minimum testing rate that is required to stop the epidemic growth at a given time by using a five compartmental epidemic model. This testing rate to “hammer the curve,” also called the Best-Effort Strategy for Testing (BEST), is feasible only if adopted early on during the epidemic. This policy will be evaluated on the COVID -19 data of France.
Second, I will present our model of urban human mobility incorporating the epidemic spread process. At a macroscale, the model captures daily mobility between residential areas and social destinations like industrial areas, business parks, schools, markets, etc. At a microscale, i.e., inside each location, it captures the epidemic spread process depending on the density of people. I will present two optimal control formulations aiming to maximize the socio-economic activity in an urban environment while keeping the number of active infected cases bounded. The first is called Optimal Capacity Control (OCC) policy, which limits the epidemic spread by reducing the capacities of destinations by some percentage. The second is called Optimal Schedule Control (OSC) policy, which limits the epidemic spread by reducing the daily business hours of each destination category. Finally, I will introduce our ongoing work on a demonstrator, Healthy-Mobility, which uses real data of Grenoble to devise optimal mobility policies.
