Scale-free points-of-interest distribution in a city emerging from homogeneous Poissonian-point processes.

Area of Science:

• Urban studies
• Complex systems
• Spatial analysis

Background:

• Urban systems often display scale-invariant properties, such as power-law distributions in human behavior patterns.
• The distribution of commercial activities and points of interest (POIs) across cities is a key example of these scale-invariant phenomena.
• The underlying mechanisms driving these heavy-tailed distributions from local urban dynamics are not well understood.

Purpose of the Study:

• To demonstrate how global spatial inhomogeneity in POI distribution can emerge from locally homogeneous processes.
• To provide a generative explanation for scale-free patterns in urban commercial structures.
• To develop a theoretical framework linking local urban dynamics to observed global distributions.

Main Methods:

• Analysis of Foursquare data from Bologna to identify power-law scaling in POI distributions.
• Development of a theoretical framework based on spatial clusters with shared intensity levels in disjoint areas.
• Introduction of a hybrid hierarchical approach combining spatial clustering with statistical heterogeneity (Poisson mixtures) to model real-world deviations.

Main Results:

• POI distributions in Bologna exhibit clear power-law scaling at the city scale.
• The theoretical framework successfully explains the emergence of scale-free patterns from local dynamics.
• The hybrid model captures real-world deviations from local regularity while maintaining interpretability.

Conclusions:

• Complex global urban phenomena can arise from the spatial superposition of simple, locally uniform dynamics.
• This research provides a generative explanation for scale-free patterns in urban commercial structures.
• The findings offer tools for interpreting, modeling, and classifying urban space by connecting micro-level homogeneity to macro-scale complexity.