Tracing DAY-ZERO and Forecasting the Fade out of the COVID-19 Outbreak in Lombardy, Italy: A Compartmental Modelling and Numerical Optimization Approach.

Lucia Russo; Cleo Anastassopoulou; Athanassios Tsakris; Gennaro Nicola Bifulco; Emilio Fortunato Campana; Gerardo Toraldo; Constantinos Siettos

As discussed, for our computations we ran 48 times the numerical opti- For all the near-optimal points obtained using the genetic algorithm optimization, the residual was of the order of ∼ 4,750,000. Regarding the values of the optimal parameters, we fitted their cumulative probability distributions using several function including the Normal, Log-normal, Weibull, Beta, Gamma, Burr, Exponential and Birnbaum-Saunders functions and kept the one resulting the maximum Log-likelihood (see in the SI). Note that the optimal values of DAY-ZERO were between January 5 -January 23 (see Figure S1 ), the optimal values of β were between 0.63 and 0.9 (see Figure S2 ) and the optimal values of were between 0.01 and 0.27 (see Figure S3 ). The best fit to the Based on the derived values of the "effective" per-day disease transmission rate, the basic reproduction number R 0 is 4.51 (95% CI: 4.14, 4.90). 350