Left: The cumulative number of survived activities, N(t), i.e. the number of activities present at time t in Rome, is plotted as a function of time t (expressed in calendar years). The inset shows the number n(t) of new registered activities at each year. Right: The number n^α(t) of new activities is reported as a function of the year t, for a selection of the commercial categories reported in Table 1, i.e. Food, Medicals, Goods and Repair.

https://doi.org/10.1371/journal.pone.0154180.g001

Spatial density of commercial activities (left/top panels), and distribution of the local category entropy H_ij(t) (left/bottom panels) evaluated for the years 1984, 1994, and 2004 (1^st, 2^nd, and 3^rd column, respectively. We have considered a grid of m × m cells, with m = 100, corresponding to cells of linear size of 350 m. The two upper/right panels show the square root of the mean square displacement from the city centre of all the activities, respectively in the registered (d(t)) and cumulative (D(t)) cases. Panels on the right/bottom show the scatter plots of density, P_ij(t), vs entropy, H_ij(t) for the same three years.

https://doi.org/10.1371/journal.pone.0154180.g002

The three panels show the values of the attraction coefficients respectively for α = 4 (“Goods”), α = 5 (“2^nd hand”), and α = 7 (“Repair”), with respect to all the other categories, namely β = 1, …, 8. The radius R was set equal to 200 meters.

https://doi.org/10.1371/journal.pone.0154180.g003