My PhD project is in collaboration with BT and aims to improve their service delivery models for repair and maintenance of the network. These models are used to understand the impact of various demand scenarios on the queue of maintenance jobs and the resources available such as workforce numbers. The first year of the project consisted of an O.R. approach; hybrid simulation models. Work has now progressed to the use of statistical methods and Bayesian inference techniques to calibrate existing BT models.
- Bayesian inference
- Kalman filtering and dynamic linear models