Modelling and predicting the effect of social distancing and travel restrictions on COVID-19 spreading

The candidate solution faithfully reproduces the paper’s calibrated ADN–SEINR meta-population framework, parameter values (including Table 1), NPI schedules, and the four core findings: (i) early, severe mobility restrictions would have reduced deaths by >12% in the Islands and can be slightly detrimental in the North-West, (ii) spatial dependence of mobility benefits, (iii) during reopening, outcomes depend mainly on activity reduction (α), with negligible contribution from mobility (β), and (iv) targeting only the low-activity/vulnerable cohort is insufficient; outcomes are driven by the high-activity class. All of these appear verbatim in the PDF, including numbers and scenarios, e.g., Islands >12% and negligible/negative effect in the North-West, and β’s negligible role during relaxation (see the Implementation and Results sections and Fig. 4–6 narratives) . The candidate adds a short analytic sensitivity (∂Π/∂β at fixed prevalence) consistent with the Methods’ contagion function but not required for the paper’s conclusions; otherwise, the proof strategy (calibration + simulation) is substantially the same.