Book Review: Advanced Geosimulation Models , D. Marceau, and I. Benenson (Eds.). Bentham Books
- PDF / 66,193 Bytes
- 3 Pages / 439.37 x 666.142 pts Page_size
- 95 Downloads / 116 Views
Book Review: Advanced Geosimulation Models, D. Marceau, and I. Benenson (Eds.). Bentham Books Alison Heppenstall
Received: 1 August 2012 / Accepted: 2 August 2012 / Published online: 31 August 2012 # Springer Science+Business Media B.V. 2012
In this book, Marceau and Benenson are extending the original concept coined by Benenson and Torrens (2004) in the book “Geosimulation”. There, as Batty points out in the foreword, geosimulation was defined having a focus on systems “whose structure and function are dominated by geographical or spatial representation and processes” (Batty 2012, p. i). This notion is extended in the new book with a focus on “elemental objects” that “embody explicit behaviours that needs to be simulated”; by this of course, they are referring to agent-based modelling (ABM) and cellular automata (CA), models that “constitute the heartland of Geosimulation” (Batty 2012). Marceau and Beneson in the opening of the book set out what they see as the main challenges in geosimulation: spatial representation, model validation, visualisation and dynamics. Through this focus on ABM and CA, the book aims to examine these challenges. While scholars have focused a great deal of attention on ABM in the past decade, the inclusion of a chapter by White, Shahumyan and Uljee on a variable grid for cellular automata (CA) that allows overlapping neighbourhoods for the Dublin region highlights that there is still considerable research ongoing with this methodology. The following chapter by Moore relates to this work by showing how assumptions of a grid can be relaxed to other non-regular shapes. Inevitably, visualisation is always a key issue in any geosimulation. This is reflected in the book through three chapters that deal with representation; in particular these focus on geo-visualisation, construction of geographic virtual environments and on the use of multimedia in both the scientific development and dissemination of outcomes. Crooks, Hudson-Smith and Patel use agents that move within an urban and spatial extent to show how easy it is to both embed these models into virtual spaces and to build such models in environments such as NetLogo. Mekni, Moulin and Paris take a different approach by building semantically-enhanced virtual geographic
A. Heppenstall (*) Centre for Spatial Analysis and Policy, University of Leeds, Leeds, UK e-mail: [email protected]
252
A. Heppenstall
environments where the focus is on visualising the outcomes of models, rather than exploiting visualisation capabilities. The book then moves onto calibration and validation; Heckbert and Bishop focus on empirical calibration of ABMs in various visual environments (presenting three data-based approaches to validating agent models) nicely linking back to the ideas in the preceding two chapters. Heckbert and Bishop advocate that models should be tuned to available data and model specifications should be flexible to the validation process. Hatna and Benenson extend the work on validation through examination of the patterns of reside
Data Loading...
