Workshop on Modelling and Policy Making
A Workshop on “Modelling and policy making” was held on September 4th,” as part of the Institute of Advanced Studies (IAS) “Tipping Points in Modelling” seminar series. The workshop was supported by Durham Energy Institute and aimed to discuss current and arising modelling challenges in large complex systems, and the use of models in decision and policy making.
The speakers, who came from around the country were ;Professor Michael Goldstein of Durham’s Mathematical Sciences Department, Dr Chris Dent from the School of Engineering and Computing Sciences; Professor Neil Strachan from University College London, Andrew Stiel from Baringa, Dr James Paterson of Edinburgh University and Peter Alexander of Scotland’s Rural College.
Professor Goldstein gave the first presentation outlining the methodology for comprehensive management of uncertainty in the relationship between computer models and real systems, which he was involved in developing as part of the “Managing Uncertainty in Complex Models” consortium and other research projects. He then discussed how Bayesian statistics, in which all uncertainties are represented as probability distributions, provides a framework for decision making under uncertainty providing that the decision maker’s utility arising from different possible outcomes can be specified.
Professor Goldstein was followed by Dr Chris Dent whose presentation on “Statistical modelling challenges in power system analysis” reflected on his experience of working in practical generation adequacy assessment, and the difficulties of communicating statistical modelling results to a broad non-specialist interested audience (ranging from public policymakers to the media and wider populace).
Chris was followed by Professor Neil Strachan of UCL whose presentation on “whole energy systems modelling” described the development and philosophy of systems modelling, the current constraints of energy models and the need to understand the risk associated with the model assumptions. Professor Strachan described the difficulties in integrating all of the systems acting on these models.
Andrew Stiel of Baringa, a consultancy who provide energy modelling advice to government and energy providers, followed Neil with a presentation entitled “ use of energy systems modelling in consultancy”. Andrew described how consultants use established models to understand and analyse the changing market conditions and identify and realise opportunities, manage change and improve performance. He underlined the need for users to understand the principles and limitations of the models in use and to ensure that this communicated to the clients. He was asked whether his company had been blamed for the results of analysis, and was able to reply positively that the companies understanding and communication with Clients has ensured that the results have been robust and accepted.
The discussion then continued over lunch, reconvening for the afternoon with a presentation from Dr James Patterson of Edinburgh University on a European funded project “CLIMSAVE” for which he had been responsible for developing the Scottish biodiversity model “Exploring the future of Scottish land use and ecosystem service provision using the UK NEA scenarios and the CLIMSAVE IAP model” Dr Patterson demonstrated the real time abilities of the software developed to model the results of different scenarios and constraints on Scotland’s land use to 2050. This model is now being discussed with Scottish Government to understand the impact of future policy.
Peter Alexander a PhD student from Scotland’s Rural College then presented his work on modelling the market for biofuel crops within the UK. Basing his assumptions on industry take up on the increase in farmers planting rapeseed oil he argued that these crops take many years to gain supplier confidence within the market. Peter had commenced his research by creating a farm scale economic optimisation model using a normative mathematical programme. Once this optimisation model was calibrated and verified it was run using spatially specific data to generate maps of economic energy crop selection across the UK.
The Workshop ended with a general discussion about the need to ensure appropriate understanding of model constraints and to develop further methodologies for managing risk within model construction.