Multi-city modeling of epidemics using spatial networks: Application to 2019-nCov (COVID-19) coronavirus in India

Bhalchandra S Pujari; Snehal M Shekatkar

We investigate the infected population of all the cities in time for both the networks separately, by first infecting Delhi with x Delhi (0) = 0.0001φ Delhi as it is well connected internationally. Fig. 1 shows the time series of infected population of several cities for D = 0.01. We note several key features of these time series. First, as a result of our modification of the SIR model that allows us to handle heterogeneous population sizes, the values of the maxima are proportional to the populations of the corresponding cities. Second, because β and γ are same for all the cities, bigger cities take longer to achieve their maxima. Also, for both the networks, most of the maxima occur in close proximity with each other implying the necessity of the preparedness against simultaneous large-scale outbreaks. Since the train network includes delay, the maxima tend to occur much later relative to that of Delhi. Because the train transport is the dominant means of transport in India, this shift of the peaks is more relevant for the response against coronavirus.