Complex Data Modelling

Complex Data Modelling


Using innovative techniques, we develop mathematical, statistical and computational methodology to support engineering projects. Our research focus is on the challenges of model-building, in the face of engineering data big or small. We collaborate closely with the Big Data Processing and Mining research group.

Further information


Problems that we solve

The modelling of complex systems and processes underpins all challenges in Engineering for Remote Operations (ERO). Examples include estimating and managing the remaining useful life of assets in a mine; foundational and structural mechanics in offshore platforms; stability and management of distributed and heterogeneous power grids; spread of infection or wildfire; and image recognition and signal processing in robotic systems.

The following are examples of the areas where our modelling expertise is applied to solve complex problems.


The availability of ever growing, large and varied data sets offers the potential to improve our understanding of many complex processes  including engineering processes. We address the challenges in applying statistical or deterministic model-based methods across the range of problems that generate big data.

Reliability and asset management

This work is the result of statisticians in collaboration with mechanical engineers and industry.  We develop statistical methods to inform the decision making of reliability engineers and asset managers, when the decision process is exposed to uncertainty. The primary statistical focus is on longitudinal and time to event methodology

Complex Systems

Complex systems are made up of a network of interconnected components that influence each other, often through complicated and diverse interactions that could not be predicted from the study of individual parts in isolation. This has immediate relevance to many complex engineering projects – in the resource sector and beyond. Our work addresses the growing realisation that many complex organisational and infrastructural systems need to be studied holistically – the effect of renewable energy generation in distributed power networks is a prime example.

Computational Simulation

Our experts are able to interpret and model huge amounts of of data using computational simulation. Examples include the spread of disease among remote communities, and bushfire modelling/prediction through vast and varied terrain.

Computer vision and pattern recognition, tomography and geometric methods

Our research capability in computer vision and pattern recognition allows us to extract useful information from images and video. Pattern recognition, combined with machine learning, is a powerful tool for analysing complex data sets  – especially images. Many engineering applications take place in situations where information is incomplete or indirect, for instance in seismographic studies of earthquakes. Our mathematical research expertise can be applied in these situations to map out and identify data patterns for interpretation.