Research Fellow in Calibration and Data Assimilation for Agent-Based Models

You will work on a new project, funded by the European Research Council, called Data Assimilation for Agent-Based Models: Applications to Civil Emergencies (DUST). The utilmate aim of the project is to develop a comprehensive simulation that can be used to model the current state of an urban area and provide valuable information to policy makers with regards to how to manage events that disrupt the ‘normal’ behaviour of cities. Agent-based modelling is an ideal methodology for this type of simulation but one that suffers from a serious drawback: models are not able to incorporate up-to-date data to reduce uncertainty. There is a wealth of new data being generated in ‘smart’ cities that could inform a model of urban dynamics (e.g. from social media contributions, mobile telephone use, public transport records, vehicle traffic counters, etc.) but we lack the tools to incorporate these streams of data into agent-based models. Instead models are typically initialised with historical data and therefore their estimates of the current state of a system diverge rapidly from reality.

The research team is lead by Dr Nicolas Malleson and will be located within the Leeds Institute for Data Analytics (LIDA), which has been established with more than £20 million of funding from the University and four major research councils. The city of Leeds is already recognised as a hub for big data analytics in business, health care and academic research. LIDA’s goals are to move the University even further to the forefront by combining projects in consumer data research, medical bioinformatics, digital humanities and monitoring environmental change to provide a truly multi-disciplinary approach to this exciting and challenging field.

You should have a PhD (or be very close to obtaining a PhD) in Geography, Computer Science, Mathematics/Statistics, Physics – or a related discipline with a significant component of programming and/or data science – and be able to demonstrate a combination of enthusiasm and expertise in computational modelling and data analysis. Evidence of relvant statistical and/or mathematical expertise that could be applied to the challenge of data assimilation is essential. Direct experience in developing and using agent-based models would be ideal but is not compulsory.

Further information on the project and The Leeds Institute for Data Analytics (LIDA) can be found on the candidate brief:

To explore the post further or for any queries you may have, please contact:

Dr Nick Malleson, Associate Professor

Tel: +44 (0)113 343 5248, email: [email protected]

This is a companion discussion topic for the original entry at