From Cells to Cities

Abstract

Since mathematical models came to be applied to problems of architectural and urban form, new concepts based on predicting large-scale structure from local rules have emerged through insights originating in computation and biology. The clearest of these are computer models based on cellular automata (CA) and their recent generalization in evolutionary biology and artificial life. Here we show how such models can be used to simulate urban growth and form, thus linking our exposition to the longer tradition of ideas in studies of built form emanating from the ‘Cambridge School’. We first review developments of CA in general and then in urban systems in particular. We propose a general class of CA models for urban simulation and illustrate two simple applications, the first a simulation of the development of the historical ‘cell’ city of Savannah, Georgia, the second, a generic hypothetical application. We then show how this generic model can be used to simulate the growth dynamics of a suburban area of a mid-sized North American city, thus illustrating how this approach provides insights into the way microprocesses lead to aggregate development patterns.

Ecological benefits assessment and spatial modeling of urban ecosystem for controlling urban sprawl in Eastern Beijing, China

“…Urban CA models have better performance in simulating urban growth and expansion than conventional urban models because they are much simpler than complex mathematical equations, but produce results that are more meaningful and useful with intuitive results (Deadman et al, 1993;White and Engelen, 1993;Wu, 1998). Its applications differ from the typical natural systems applications in that they tend to have a very limited total number of cells and a small number of temporal iterations (Batty and Xie, 1994). A more substantial difference lies in the definition of transition rules (Wu and Webster, 1998).…”

“…A variety of approaches have been used to model the spatial process of urban expansion, such as Markov chains (Ló pez et al, 2001) and spatial logistic regression (Cheng and Masser, 2003). In recent years, Cellular Automata (CA) techniques have gained attention as modeling tools for urban expansion process simulation (AlAhmadi et al, 2009;Barredo et al, 2003;Batty and Xie, 1994;He et al, 2008;He et al, 2006;Li and Yeh, 2000;Liu et al, 2008;Stevens et al, 2007;White and Engelen, 1993;Wu, 1998). The application of CA in urban expansion modeling can give insights into a wide variety of urban phenomena (Li and Yeh, 2002).…”

Complexity and Dynamic Modeling of Urban System

“…Tobler, from University of Michigan, formally simulated the development of Detroit of Great Lakes for the first time in 1970s [15]. Thereafter, Helen Couclelis [16], Batty [17], Clarke [18], Li and Yeh [19] successively studied geographical and urban cellular automata, who made a good many meaningful achievements.…”

Urban Research and Complexity

“…Portugali and Benenson (1994) applied the method for generating spatial distributions of communities inside a city according to their preferences for a type of social neighbourhood. A model of spatial urban growth using a two-state cellular automata process for the filling of built-up space in peri-urban fringes as well as in an entire urban area in the long term has 26 been developed by Batty and Xie (1994). This last paper, instead of exploring only the theoretical possibilities of the model, also tries to calibrate it with an observed urban evolution.…”