Cambridge University logo

My contact details:

Dr.I. J. Lewis
Director of Infrastructure Investment
Department of Computer Science and Technology
University of Cambridge
Office phone: 01223 331859
Email: ijl20@cam.ac.uk

Office mailing address:

Department of Computer Science and Technology
Gates Building
JJ Thompson Avenue
Cambridge
CB3 0FD
UK
If you are a University member authenticating via Raven, you can see more contact details including my mobile and home phone numbers via my Lookup page. If it's important, don't hesitate to call me via my mobile or at home.

Research Activity

I manage the Adaptive Cities Programme, researching the digital aspects of Future Cities. This work contains a few major themes:

  1. Crucially, our work is comprehensively in partnership with the local authorities in our local region. The straightforward objectives include:
    • via the University's expertise, the region benefits from real digital infrastructure that is far more capable then might otherwise be the case,
    • the University gains access to infrastructure that is fit for purpose for our research
  2. Current urban digital infrastructure as at an inflection point from typically dozens of sensors (e.g. measuring air quality) to comprehensive coverage with thousands or tens of thousands of sensors, some of which will be stationary while others will be mobile, having in common both spatial and temporal references.
  3. Sensors are being transformed in capability. An excellent example is a typical traffic sensor today is a metal loop embedded in the road which provides a vague 'blip' when a vehicle passes over. It has recently become possible economically to deploy image-based AI sensors that can differentiate and count cars, bicycles, trucks, buses and pedestrians within the scene.
  4. We assume we are measuring things in the region because we are likely to want to do something about it. I.e. when traffic congestion occurs we want intelligent adaptation of the road network.
  5. Any adaptation should be based on a prediction of an issue occurring, rather than waiting for the problem to become visible and attempting to fix it then. Hence we consider the adaptation proceeding both in time and space.
  6. Our work assumes the collection, analysis, predition, and adaptation to be performed in real-time. I.e. while we have billions of real data points stretching back several years, our historical analysis is a means to an end, i.e. we want to inform decisions taken in a timely manner.