Micro-Simulation Modelling And Predictive Policy Analysis System For Leeds

School of Geography, University of Leeds


Principal Investigator:

Dr R Kingston*

Dates:

1st February 2003 – 31st July 2004

Grant:

Leeds Initiative

Summary:

The Leeds Initiative, the local strategic partnership for the Metropolitan District of Leeds, wishes to develop a computer based system which will undertake geographical modelling to enable the description of current conditions and issues in neighbourhoods. They require a facility with will allow them to predict future trends in the composition and health of neighbourhoods and conduct modelling and predictive analysis to measure the likely impact of policy interventions.

The aim is to use the Micro-MaPPAS simulation tool to help policy-makers across the city to explore the implications of different policy scenarios and increase our capacity to use intelligently the information that is held across different organisations. Examples of policy scenarios in which the micro-simulation modelling tool is likely to be useful include:

Moreover, the simulation will be able to offer some clues as to the implications of what might happen if established trends continue and no new policies are introduced.

So what is microsimulation?

Microsimulation is a methodology that builds large-scale data sets containing information about the characteristics (attributes) of individual people or households. The idea is to simulate the entire population of Leeds numbering 715,400 individuals or 301,600 households (according to the 2001 Census of Population).

Traditionally, confidentiality constraints mean that demographic and socio-economic data on individuals, despite being collected from censuses and surveys, are not available for researchers. Instead data gets published for areas of different geographical magnitude.

Microsimulation is particularly valuable because it makes use of the data from these different sources to provide estimates of the probabilities that individuals or households will have particular characteristics…. e.g. that an individual is aged 18, is a lone parent, unemployed and lives in private accommodation in an area prone to high levels of crime…. or that a household in the outer suburbs contains 5 persons, has a head of household who is a professional who works in another city and who earns £50,000+ a year.

Once the long list of individuals and their attributes has been simulated, the individuals (and the attributes which they possess) can be aggregated to any geographical scale which is deemed appropriate using GIS technology.…..census output areas, wards or postal sectors for example,….. or more specific areas designed for policy implementation….. e.g. regeneration areas.

The software tool is being written in Java by the research team bringing together open source micro-simulation and Geographical Information System software.

* Richard Kingston is now based in the School of Planning and Landscape at Manchester University.


[School of Geography homepage] [Leeds University homepage]