About the Project
In the traditional design of experiments, we assume that each run of the experiment involves the application of a treatment consisting of fixed, scalar, settings of the controllable variables, or factors. In many modern experiments, treatments can in fact be varied during each run. Examples include (i) varying the temperature during a chemical reaction and (ii) controlling variables that mimic dynamic wind speed and direction in engineering experiments to develop wind turbines.
This PhD project will extend the usual optimal experimental design paradigm to experiments where the settings for one or more variables are functions, for example, of time. For these new experiments, a design will consist of choices of combinations of functions for each run of the experiment, together with settings for any non-functional variables. Methodology will be developed to specify the optimal designs to answer scientific questions of interest, taking account of all sources of uncertainty. The project will involve researching suitable modelling methods (both empirical and mechanistic) and optimal design strategies, along with associated computational techniques. The work will be motivated by, and applied to, real experiments from engineering, science or industry.
The successful applicant will join the world-leading Design of Experiments group at Southampton, the largest and most active group in its field in Europe. Our enthusiastic and lively group of staff, postdocs and PhD students, along with regular study groups, seminars and external collaborations, make for a stimulating research environment. We are part of the Southampton Statistical Sciences Research Institute, one of the largest and most varied groups of academic statisticians in the UK.
ENQUIRIES: For more information, contact Dr Antony Overstall ([Email Address Removed])
HOW TO APPLY: Visit
http://www.southampton.ac.uk/maths/postgraduate/research_degrees/apply.page?&utm_source=mathspgr&utm_medium=mathspgrCTAapply&utm_campaign=mathspgrCTAapply
Continue with Facebook
