Postgraduate project opportunities: Prof Colin Fidge

The following are some project ideas for future Honours, Masters or PhD students. They are meant as a starting point for discussion - other suggestions are welcome.

• Information Security Analysis of Network Topologies
  The Secure Information Flow Analyser (SIFA) is a tool for analysing potential information flow through systems that can be represented as graphs. It was originally designed for analysing electronic circuit diagrams. However, the tool is general enough that it could also be used for analysing communications network layouts, or other such complex systems. This project will explore and develop SIFA's potential as a tool for evaluating network security. To do so, it will be necessary to devise models of not only the network's topology, but also its protocols and services. In particular, the project will consider ways in which the tool could be extended to deal with mobile, wireless networks, in which security-critical events may be separated both in time and geographically.
• Modelling and Analysis of Information Security Evaluation Procedures (Honours or Masters project)
  Communications devices intended for use in high-integrity applications are subject to international standards such as the Common Criteria for Information Technology Security Evaluation. The processes involved in performing such evaluations are highly technical, time consuming and costly. This project will explore how modelling the information security evaluation process in a workflow engine can help manage and streamline it.
• Identifying Risky Business Processes
  Process models are graphs that define the sequences of tasks involved in business activities. As such models become larger and more complex it becomes increasingly difficult to recognise weaknesses in them that may lead to unacceptably risky situations. We can guard against this by defining patterns for safe business practices (e.g., independent approvals of expensive transactions, staff rotation to avoid overtiredness, etc) and by then ensuring that a given process model is consistent with these patterns. This project will develop a software tool capable of showing consistency between two business process models, both represented as graphs. This will be done by proving that one graph is subsumed within the other.
• Modelling and Mining Product Distribution Processes
  The way products move through distribution networks depends on the workflow processes enacted at each node. Characterising such processes accurately is vital for optimising the network and understanding its overall behaviour. This project will show how distribution network processes can be modelled formally and implemented in a workflow engine. It will then show how logs of the network's past behaviour can be mined in order to extract performance data that can be used to predict future behaviours. As a practical case study, the project will focus on the problem of predicting the location of suspect items during safety-critical product recalls.
• Tools for Asset Management Decision Support
  Efficiently and effectively managing large-scale engineering assets, such as power stations, rail networks, etc, is a challenging task. This project will consider how tool support can make the job easier. A variety of different tools could be applied to the problem. One approach would be to use probabilistic model checking to explore maintenance strategies. Challenges include the need to satisfy multiple business and technical objectives, over a wide range of time scales, based on historical maintenance data which may be incomplete or inaccurate. Another approach would be to use a workflow engine to model and enact the overall maintenance decision lifecycle.

Back to home page