Category Archives: Entropy

Applied Urban Modelling 2018

This conference is the seventh in a by-now very well established meeting held each June in Cambridge UK, which brings together those working on land use transport interaction/integration models (LUTI models) in the narrower sense and more generally on urban simulation models. Although initially set up to continue work on aggregate land use and transportation models, it has broadened to include other kinds of models and this year there are interesting papers on very large scale models such as UrbanSim, PECAS and TRANUS as well as cellular automata models like SLEUTH. The programme is here and those wanting the papers should contact the authors. During the meeting we have had a session dealing with the contributions of Lionel March to urban modelling. I tweeted about this a couple of weeks ago but here is my contribution – the PDF of my talk about my work with Lionel some 44 years ago when we were both at the University of Waterloo in Engineering. Lionel of course set up the Centre for Land Use and Built Form Studies that morphed into the Martin Centre (at the University of Cambridge’s School of Architecture) in the mid 1970s. Much of what goes on in LUTI modelling can be traced to Lionel and I am not being melodramatic. My talk recounts what we did with probability theory and spatial interaction and how we tried to fashion ideas about priors, posteriors, minimum information and so on. There is much more to say about Lionel’s contributions but readers might be interested in my own thoughts which are in the attached PDF. Enjoy.

GeoComputation 2017

2017 International Conference on GeoComputation: Celebrating 21 Years of GeoComputation 4-7 September at the University of Leeds

Stan Openshaw and his colleagues set up the first meeting 21 years ago in Leeds and it returns there for its coming of age. Some good papers will be presented but we have published some commentaries in Environment and Planning B: Urban Analytics and City Science on the state of the art in GeoComputataion, and you can get these by logging on here. All freely downloadable. Read the words of wisdom and incisive critique and commentary on GeoComputation from

Editorial: GeoComputation: Michael Batty

Commentaries: More bark than bytes? Reflections on 21+ years of GeoComputation: Richard Harris, David O’Sullivan, Mark Gahegan, Martin Charlton, Lex Comber, Paul Longley, Chris Brunsdon, Nick Malleson, Alison Heppenstall, Alex Singleton, Daniel Arribas-Bel and Andy Evans



A picture of the second CBD in Sao Paulo rapidly developing as a major financial-retailing centre in Itaim Bibi some 10 kms south west of the established centre at Paulista.

Our ESRC project with the University of Sao Paulo and INPE which will involve quantitative comparisons of spatial and social segregation with a strong focus on transport accessibility, was launched in Sao Paulo on November 26-27 at the Center for Metropolitan StudiesThe London team is led by Mike Batty (CASA-UCL) and Joana Barros (Geography, Birkbeck) with Chen Zhong (CASA-UCL) as research associate. Duncan Smith (CASA-UCL) has been drafted in to help with the web based mapping and portals. In Sao Paulo , the team is headed up by Eduardo Marques (USP) and Flavia Feitosa, Center for Engineering, Modeling and Applied Social Sciences, Universidade Federal do ABC (CECS) , supported by Miguel Vieira Monteiro (INPE: Brazilian Institute of Space Research, Divisão de Processamento de Imagens). The project is part of the ESRC’s Urban Transformations thematic programme and will draw on cognate material from other relevant projects within that theme.

What we are doing in this project is measuring and modelling social and spatial segregation in the two metro areas which are roughly comparable in size (15-20 millions). RESOLUTION stands for REsilient Systems fOr Land Use TransportatION and our mission is to derive measures that show resilience as well as vulnerability with respect to social and spatial segregation with a focus on transport and movement. As cities get bigger, people travel longer and commute times in Sao Paulo (SP) tend to be longer than London but the quality of the experience differs. In SP, the spatial structure of the city locks people out in ways that are different from areas of poor accessibility in London. We intend to measure this kind of lock out using access of different groups to different kinds of employment and go beyond this dealing with access to other facilities such as hospitals, schools and related welfare or lack of it. Our effort is to see where we might identify pressure points that reveal vulnerability and points where we might engender some resilience.

So far was have not completely worked out the sets of measures that we will work with but we consider local indicators of spatial association, entropy measures, distributions of uneven patterns ranging from income distributions to distributions of density – all these will be employed, following some of those we worked with in our previous modelling projects (accessibility indices) and diversity indices worked on by Chen Zhong in our recent Urban Studies paper. (You can download a copy here). We have also pencilled in that we will explore new ideas about accessibility and segregation from agent based model of transport and our aim is still to build simple ABMs reflecting household choices with respect to travel for SP and London. We do not intend to develop MATSIM for London any further in this project but something somewhat simpler that this and more focussed on segregation is on the cards. Flavia has done a lot of work already on SP with respect to modelling segregation using ABMs and we will follow her lead. Worth looking at her recent papers here.

Last but not least we are beginning by establishing two web portals – one for London and then once this is up and running an equivalent one for SP – so that we can compare data but also provide a resource that others might use to explore the same data hat we will assemble and work with. This also provides an initial focus on how we might begin our comparisons. Although we think there is reasonable comparability of data between the two cities at similar levels of spatial aggregation, until we assemble the data and develop the portals we will not know of the problems that are likely to arise.

The meeting in SP raised all these issues and more and in a two day intensive session we developed an agenda for these next six months. This will be placed on a Wiki-like page once we have absorbed all the material from this opening workshop.