Urban cellular automata and agent based models for the simulation of urban dynamics: a review of practive and applications

Date
2020-03
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: Current scientific planning instruments and practices are inadequate to address the multidimensional problems and challenges faced by cities as complex dynamic systems. The aim of this research is to provide an international comparative analysis of Cellular Automata (CA) and Agent-based modelling (ABM) techniques and its potential application within spatial planning practices. The research provides explanations on the key considerations for spatial simulation model conceptualization, components, design and construction. Cellular Automata (CA) and Agent-based modelling (ABM) techniques abstract the real-world into a series of layers as a visual representation of complexity and spatial-temporal urban dynamics. The meta-analysis of published spatial simulation research results over the past decade (2009 – 2019) found that urban modelling approaches have grown consistently. Applications of urban simulation models appear to be regionally divergent with the major focus on the global North. Uptake of these urban models is lagging in areas with rapid urbanization and urban growth rates, which are predominantly located in the global South (including South Africa). The comparative analysis found that the development and design of urban models are also now incorporating aspects of strategic planning within their scenarios in order to measure and monitor the appropriateness and effectiveness of policy interventions, such as urban growth boundaries, zoning schemes, sustainable development outcomes and environmental protection zones. The research found that CA and ABM-based urban models improve the understanding of the local and historical contingent factors and how multidimensional and complex problems influence urban systems across time and space.
AFRIKAANSE OPSOMMING: Huidige instrumente en praktyke vir wetenskaplike beplanning is onvoldoende om die multidimensionele probleme en uitdagings wat stede as komplekse dinamiese stelsels in die gesig staar, die hoof te bied. Die doel van hierdie navorsing is om 'n internasionale vergelykende analise van Cellular Automata (CA) en Agent-gebaseerde modellering (ABM) tegnieke te bied en die potensiële toepassing daarvan binne ruimtelike beplanningspraktyke. Die navorsing verskaf verduidelikings oor die sleuteloorwegings vir ruimtelike simulasiemodelkonseptualisering, komponente, ontwerp en konstruksie. Cellular Automata (CA) en Agent-gebaseerde modellering (ABM) tegnieke abstrakteer die werklikikeid in 'n reeks lae as 'n visuele voorstelling van kompleksiteit en ruimtelik-temporele stedelike dinamika. Die meta-analise van gepubliseerde navorsingsresultate vir ruimtelike simulasie oor die afgelope dekade (2009 - 2019) het bevind dat die benaderings vir stedelike modellering konsekwent gegroei het. Toepassings van stedelike simulasiemodelle blyk streeks uiteenlopend te wees, met die grootste fokus op die ontwikkelde wêrel. Die gebruik van hierdie stedelike modelle hou egter nie verband met gebiede wat 'n vinnige verstedeliking en stedelike groeikoers ondervind nie, soos byvoorbeeld die globale Suide (insluitend Suid-Afrika). Die vergelykende ontleding het bevind dat die ontwikkeling en ontwerp van stedelike modelle nou ook aspekte van strategiese beplanning binne hul vooruitbeplanning inkorporeer om die toepaslikheid en doeltreffendheid van beleidsintervensies, soos stedelike groeigrense, soneringskemas, volhoubare ontwikkelingsuitkomste en omgewingsbeskermingsones. Uit die navorsing is bevind dat CA- en ABM-gebaseerde stedelike modelle die begrip van die plaaslike en historiese faktore verbeter en hoe multidimensionele en ingewikkelde probleme stedelike stelsels oor tyd en ruimte beïnvloed.
Description
Thesis (MA)--Stellenbosch University, 2023.
Keywords
Citation