Machine learning approaches for uncertainty assessment in reservoirproduction forecasting

A PhD studentship is available in the Institute of PetroleumEngineering at Heriot-Watt University, to work on an industry fundedUncertainty Quantification project.

The work will involve application of data-driven statistical learningmethods such as Support Vector Machines and Artificial Neural Networksto address diversity of geological reservoir models. The challenge ofthe project is to link contemporary machine learning algorithms withthe state-of-the-art Bayesian framework for uncertainty quantificationdeveloped in the group.

The successful candidate should possess strong numerical andanalytical skills and an interest in data driven approaches. A goodknowledge of high level programming is desired. Experience ingeomodelling / geostatistics or reservoir engineering is beneficial.

The candidate will join a dynamically developed team of postdocs andPhDs lead by Prof. Mike Christie. The research carried out by the teamaddresses aspects of uncertainty quantification including stochasticoptimization methods, solution error models, neural nets, and employshigh-level scientific computation (including access to a recentlycommissioned 84 node linux cluster). The research is funded by aconsortium of oil companies and also by UK EPSRC, and the skillsacquired in while studying for the PhD are likely to be applicable toa wide range of areas, including the oil industry.

To apply send a cv to Dr V Demyanov, vasily.demyanov@pet.hw.ac.uk, by31 October 2005.