On the perturbation of self-organized urban street networks

Abstract

The complexity of urban street networks is well recognised to reside in their information networks. An information network maps roads to nodes and maps junctions to edges. Information networks of self-organized cities typically have a scale-free degree distribution. A recent fluctuating mesoscopic model links their scale-freeness to the preservation of their average amount of surprisal. Here the surprisal measures the astonishment and indecision of city-dwellers. The Metropolis algorithm may in theory allow to generate fluctuating information networks, with the scaling exponent acting as inverse temperature, in thermallike equilibria. In this paper, we show how the Metropolis algorithm can apply to urban street networks along a case study. Our case study, Old Ahmedabad, sustains Metropolis equilibria and offers promising thermodynamic-like results. Our work opens doors to a statistical physics framework for understanding how self-organized urban street networks develop.