Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Distance to the CBD and neighboring commercial employment (land-use) have been the core determinants of spatial-related production externalities for firms. In these models, travel time to work (firms) is the single most important factor for residential land-use allocation. New theories of complex urban systems (CUS), however, have begun to cast some doubt on the efficacy of the “distance to CBD” model. There is some evidence for example, that large urban systems might evolve with scale-free transportation networks. In this paper, we examine urban land-use data from the Chicago Metropolitan Statistical Area (MSA) to argue for a theoretical shift from a “distance to CBD” based prototype to one that considers the complexity inherent in urban systems structure. We use a Stochastic Greedy Algorithm to quantify connectivity and attractions to every land cell (30 × 30 m) to existing population and employment centers, Points of Interests (POIs), and highway and major roads. We measure the frequency of commercial and resident land-uses relative to these found attraction levels and develop algorithms that help explain the relations. Using these methods, we find that both CBD-driven and network-driven approaches are empirically valid for explaining current urban structures. We also find, however, that these relations change when temporal variables are considered. For example, we found that the land-use change in Chicago from 2001 to 2011 is an obvious deviation from the “distance to CBD” based urban growth assumption. Our results suggest that we should re-examine the core urban structure assumptions of spatial equilibrium models.