Economic and Social Research Council
This website will look much better in a web browser that supports web standards, but it is accessible to any browser or Internet device. Go to main content.

Improving Evidence Based Policy Decisions: Piloting the Application of Advanced Computer Modelling Techniques to Real Life Policy Problems

Organiser

Organised jointly by the University of Birmingham's School of Computer Science, Centre for Urban and Regional Studies (CURS) and Institute for Local Government Studies (INLOGOV).

Date and Location

Friday 9th February 2007, 10am - 4pm
Professional Development Centre, University of Birmingham

Workshop Summary

Computer models may be used to manage complexity and to predict policy impacts. Such models may be in the form of simulations. However, simulations on their own are often based on assumptions that may not be true about the real life situation and are therefore not a reliable guide to policy decisions.

To improve the reliability of such models, a project at Birmingham University is developing a semi-automated approach to assist users in finding evidence to support or refute a models predictions using data mining and analysis tools. The system should help the user to make changes to the simulation according to the evidence. In technical terminology the approach is called "data-driven"simulation.

This workshop reported on how this approach has been piloted on housing policy problems, and considered its development into other areas - especially community safety and crime reduction.

The prototype decision tool has been developed by a team from the University's School of Computer Science, Centre for Urban and Regional Studies (CURS), and the Institute for Local Government Studies (INLOGOV) as part of the AIMSS project.

The workshop brought together managers from organistations concerned with housing and community safety, as well as specialists in the computer science field.

Aim

  • Explain in lay language the principles of evidence-based model development and the way in which it can assist policy making.
  • Demonstrate the pilot software developed by the Birmingham team, using the example of housing policy
  • Illustrate the way in which this approach builds on the existing technologies of agent based simulations and data mining, exploiting the advantages of both and overcome their respective limitations.
  • Review the relevance and applicability of this pilot to the public and not-for-profit organisations concerned with housing and community safety, and identify the implementation problems that may need to be addressed.
  • Review the technical aspects of the approach, and consider ways in which it could be strengthened and prototyped.

The morning session had an emphasis on application to policy makers and managers whilst the afternoon dealt with the technical aspects of the work.

Feedback

Feedback from these Agenda Setting Workshops is extremely important to us. If you attended this workshop, please could you take 5 minutes to complete the feedback form (64kb). Once completed please email to Katy Middlebrough, or post to Katy at the address on the form.

Programme

Improving Evidence Based Policy Decisions: Piloting the Application of Advanced Computer Modelling Techniques to Real Life Policy Problems (535KB)
G. Theodoropoulos, P. Lee, C. Kennedy & E. Ferrari, University of Birmingham

Data Driven Simulations for Evidence Based Modelling: why it is relevant to public policy problems (922KB)
Peter Lee, CURS, School of Public Policy, & Ed Ferrari, University of Sheffield

Piloting Data Driven Simulation in Housing Policy (885KB)
G. Theodoropoulos, University of Birmingham, & Catriona Kennedy, University of Birmingham / NCeSS

Understanding the Opportunities for Developing the Approach for Public Policy; added values of DDDAS for social scientists and policy makers (541KB)
C. Kennedy, University of Birmingham / NCeSS, & G. Theodoropoulos, University of Birmingham

Validating and Verifying Agent-Based Models For Planning and Public Analysis (4.68MB)
Mike Batty & Andrew Crooks, University College London

Issues in the Construction and Validation of a Large-Scale Social Simulation Model 362KB)
Mark Birkin, University of Leeds

Mapping Concepts in Social Surveys: the experience of the Data Chronicles project (229KB)
Karen Clarke (with Judith Aldridge & Phil Edwards), University of Manchester

Investigating the Role of Ontologies in Computer Simulation (249KB)
Simon J E Taylor, Sergio DeCesare & David Bell, School of Information Systems, Computing and Mathematics, Brunel University

Simulating People Being Arrested (150KB)
Andrew Greasley, Anton Business School, University of Birmingham