Strathclyde Business School
Deadline: 1 September 2013
Funding: Engineering and Physical Science Research Council (EPSRC) funded for 3.5 years; EPSRC eligibility criteria apply)
Research: Strategic Dynamic Bayesian Networks (SDBN) represent a new form of modelling motivated by the need to address decisions problems, such as the design of a resilient supply. SDBN models formally incorporate the optimal decisions of multiple decision-makers as well as modelling the temporal dynamics of the interactions between these decisions and the realisation of uncertainties. A major challenge in operationalizing SDBN lies in the computational complexities required if these models are to be applied to large complex networks. This projects aims to address two gaps in knowledge.
- Research Challenge 1: To reduce the cognitive elicitation burden placed on experts in populating SDBN models, we seek to develop an algorithm that iterates between eliciting expert judgment and a mathematical evaluation of such data.
- Research Challenge 2: To provide optimal decision support for large, complex networks, we seek to develop computationally efficient methods of obtaining accurate inference for the SDBN.
This research is novel and exciting because SDBN are in their infancy and show great potential to be applied within a cooperative game theory framework. The development of more efficient algorithms requires use of novel methods within Graph Theory which are ripe for translation to this new context.
Student Eligibility: This is an EPSRC funded studentship and as such the EPSRC eligibility criteria apply. The studentship covers UK/EU tuition fees and a stipend of approximately Â£13,726pa for 3.5 years. Candidates are required to have an excellent Honours (Undergraduate) degree in a numerical subject, including but not limited to Operational Research, Mathematics, Computer Science, Engineering or Economics. Candidates with a Masterâs degree (or equivalent) will be strongly preferred. For more information about the Department: http://www.strath.ac.uk/mansci/
To apply click on the âApplyâ button below.