EPSRC Reference: 
EP/H007377/1 
Title: 
MUCM2 
Principal Investigator: 
O'hagan, Professor A 
Other Investigators: 

Researcher CoInvestigators: 

Project Partners: 

Department: 
Probability and Statistics 
Organisation: 
University of Sheffield 
Scheme: 
Standard Research 
Starts: 
01 October 2010 
Ends: 
31 December 2012 
Value (£): 
942,100

EPSRC Research Topic Classifications: 
Statistics & Appl. Probability 


EPSRC Industrial Sector Classifications: 

Related Grants: 

Panel History: 
Panel Date  Panel Name  Outcome 
09 Jun 2009

Basic Technology Translation Grants Call 4

Announced


Summary on Grant Application Form 
This project concerns uncertainties in the predictions made by models. A model is a description of a real process, using mathematical equations. Usually, a computer is used to compute or solve these equations to produce the model predictions, and we call this a simulator. We think of the model predictions as the outputs of the simulator. The simulator also has inputs of various kinds, which are numbers to put into the equations. For example, a model to forecast the weather is based on very complex equations describing the movement of the air at various altitudes, the formation of clouds, and so on. The numbers to be put into the simulator include the current state of the atmosphere, the temperature of the air at different locations and altitudes, physical constants used in the equations, and so on. Any model is an imperfect representation of reality, and its predictions are therefore imperfect. The predictions can be wrong because the equations are wrong, they have the wrong numbers in them, or the computer program (the simulator) is solving them inaccurately. In practice, all of these imperfections are present to some degree. As a result, we may expect the true realworld value corresponding to the simulator output to be close to the model prediction, but there is uncertainty about its precise value.The basic objective of the MUCM project was to create the technology to allow modellers and model users to manage the uncertainty in simulator outputs, through developing tools to quantify the uncertainty, to analyse the sources of output uncertainty, and to employ observations of the realworld process to calibrate the model and thereby to reduce uncertainty. Key components of the MUCM approach are (a) an efficient way to manage model uncertainties through building an emulator of the simulator, and (b) a formal recognition and statistical modelling of the difference between the simulator output and reality, which we call the discrepancy.The basic science for building emulators and modelling discrepancy was already developed for some common types of simulator, and had been tested in relatively simple applications. The focus of the MUCM project was twofold. One part of the project was concerned with pushing the limits of the technology, for instance to deal with more complex simulators with many inputs and/or outputs, to take advantage of the availability of multiple simulators (or a simulator that can be run at different resolutions) or to handle dynamic simulators. The other part was directed to delivering a robust, welldocumented technology through the creation of a toolkit and some case studies.MUCM2 seeks to explore new directions that will facilitate taking the MUCM technology to the next stage. Three workpackages will address three important new domains for MUCM methods, where new science is needed. Workpackage 1 is concerned with simulators that are intrinsically random, so that running the simulator a second time with the same inputs will typically produce different outputs. Workpackage 2 considers how to support decision making through analysis of uncertainty. Workpackage 3 deals with simulators for which the output may have discontinuities or other forms of heterogeneous behaviour as the inputs are varied. All of these have been identified as important for taking MUCM methods to new user communities, and Workpackage 4 is focused on methods to build and serve the communities of users and researchers.

Key Findings 
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk

Potential use in nonacademic contexts 
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk

Impacts 
Description 
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk 
Summary 

Date Materialised 


Sectors submitted by the Researcher 
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk

Project URL: 

Further Information: 

Organisation Website: 
http://www.shef.ac.uk 