Decision-making under Uncertainty for COmplex systems (DUCO) team

To build uncertainty-aware digital twins that make risk and complexity in interconnected systems tractable, at scale.

We pursue this vision through:

Developing scalable risk methods for complex, interconnected systems.
Unifying isolated scientific models within a shared computational framework, enabling them to complement, challenge, and reconcile each other.

To this end, our research foucses on addressing scalability and validation challenges. We study Bayesian network, systems engineering, uncertainty quantification (UQ), reliability and risk assessment, and risk-based optimisation.
We aim to develop software tools applicable across diverse systems and risks. Our interests include, but are not limited to, transport networks, structural systems, energy grids, and process plants, subjected to risks such as earthquakes, floods, wildfires, localised incidents, and deterioration.

Research

Bayesian network for
system resilience Optimisation and
decision-making
under uncertainty System simulation functions Data collection and
Validation

Toolkits

Bayesian network

System risk assessment

RSR: Tensor-based reliability calc. of coherent systems

Data

Network datasets repo: Open datasets for network systems and functions
RDMpy: Reconstructing UK Rail Delay Propagation records

Training & Community Engagement

See posts.

Applications

Infrastructure networks Structural systems