Bayesian Statistical Modelling for Quantitative Risk Analysis

University of Bath / DNV GL – PhD studentship available - Deadline for applications: Wednesday 8th April 2015

Bayesian Statistical Modelling for Quantitative Risk Analysis

In collaboration with the EPSRC Centre for Doctoral Training in Statistical Applied Mathematics at Bath (SAMBa) www.bath.ac.uk/samba

The Department of Mathematical Sciences at the University of Bath and DNV GL, a company who use statistical models to provide risk assessments across a wide range of industries have developed a PhD project which will be co-funded for 3.5 years starting in September 2015.

We are looking for exceptionally qualified students wanting to work at the interface between applied mathematics, statistics, and probability, with a clear industrial focus.

Brief project description: Quantitative Risk Analysis (QRA) has been applied to risk and safety assessment for major accidents in the oil and gas industry. The interaction between system components can be handled by directed network graphs of probabilistic and deterministic sub-models, leading to a computationally challenging problem that requires novel solution techniques. Advances in Bayesian
statistical modelling, estimation, and computational methodology should be exploited to advance the state of the art of practical QRA, and to analyse model sensitivity and uncertainty. The aim of the PhD would be to develop innovative QRA models, e.g. including human factors and temporal dynamics, and computational methods for larger and more realistic models than current techniques allow.

Candidates should have a strong background in mathematical statistics, including statistical modelling, strong probability or Markov processes. Programming, numerical methods and simulation skills are also essential.

The supervisory team will be drawn from a number of potential supervisors at DNV GL and Bath, including Dr David Worthington (DNV GL), and Dr Finn Lindgren, Dr Simon Shaw, Dr Evangelos Evangelou and Dr Karim Anaya-Izquierdo (Bath). Students will be based in Bath but expected to spend time on placements at DNV GL’s offices.

Students taking up the award will benefit fully from the infrastructure of Bath’s EPSRC Centre for Doctoral Training (SAMBa). This includes symposia seminars delivered across a wide range of topics in mathematics, statistics, and their applications, week-long Integrative Think Tanks involving industry, academics and students formulating mathematical problems, transferable skills training and support to attend academic conferences.

Further information about the Department of Mathematical Sciences, and the EPSRC CDT in Statistical Applied Mathematics is available on www.bath.ac.uk/samba.