Figure 3
From: Activity of vehicles in the bus rapid transit system Metrobús in Mexico City

Statistical analysis of the distances \({{\mathscr {D}}}_{\mathrm {KL}}\) for speed densities. (a) Average Kullback–Leibler distances \(\langle {\mathscr {D}}_{\mathrm {KL}}\rangle\) in each segment, error bars represent the standard deviation of the values. (b) Average values \(\langle {\mathscr {D}}_{\mathrm {KL}}\rangle\) represented in the segments map. (c) Probability densities of the values \({\mathscr {D}}_{\mathrm {KL}}/\langle {\mathscr {D}}_{\mathrm {KL}}\rangle\) in each segment codified in the color bar. Probability densities \(\rho _{{\mathrm {month}}}(v)\) of the speed v at the scale of months for: (d) segment 10 with \(\langle {\mathscr {D}}_{\mathrm {KL}}\rangle =0.296\) and (e) segment 212 with \(\langle {\mathscr {D}}_{\mathrm {KL}}\rangle =0.050\), dashed lines represent \(\rho _{{\mathrm {total}}}(v)\) for the segment. All figures were created using python 3.8 and the matplotlib (3.4.3) package (https://matplotlib.org). The map in panel (b) was created using the geopandas (0.10.2) package (https://geopandas.org).