The evidence from computational models is gaining is use and legitimacy in the field of policy and strategy making, as data availability, computational power, and modelling techniques develop, bring model based decision making into the public eye. For over 100 years mathematical and computational models have been used generate evidence to inform policy and the public - weather forecasting, disease outbreaks, natural disasters, economic forecasting, etc. They are not only use to help prepare for natural events, but are used to inform choices in policy and business, and increasingly as the basis for realtime operations - "automated decision making".

While mathematical models maybe used in deterministic prediction, many contemporary computational models help us deal with the unknown and the uncertain - they help estimate gaps in data, or provide probabilistic scenarios of the future. However, the results from running models can always be contested, and probabilistic evidence can be hard to incorporate in to policy and public debate. It is also hard to interpret and trust the expert's jargon and graphs when compared with political certainties or established heuristics.

The climate change threat saw the rise of modelling in political and public consciousness, and the COVID crisis doubled this, with the whole public struggling to analyse predictions, and life changing decisions being made by political leaders on the basis of hastily repurposed flu models and debatable graphics. Different choices of and in models, often understood by relatively few people can have profound impact on decisions. In this session we welcome two mathematicians to present and discuss modelling in COVID, and invite you to think about the future role of models across public policy.

Chris and Ben are collaborating on a report to  be published by the end of 2021 by the Newton Insitute, on mathematical sciences in the COVID pandemic