Complex systems research seeks to understand how collective behaviour, self-organisation, and critical phenomena emerge from interactions between many individual components in large systems. Our research develops and integrates methodologies spanning agent-based modelling and simulation, information theory, dynamical systems, complex networks, and distributed AI. These approaches enable the analysis and modelling of complex systems across a wide range of domains, including brain dynamics, pandemics, systems biology, social systems, swarm behaviour, smart cities, energy grids, and supply chains.
A rapidly emerging area of research is the development of digital twins: high-fidelity computational models that replicate the behaviour of real-world systems. These technologies allow designers and decision-makers to simulate, test, and optimise systems before deployment, supporting applications such as cyber-physical systems, robotic swarms, sensor networks, intelligent transportation, and distributed energy grids.
Our research spans across multidisciplinary research
Agent-based modelling and simulation provides a powerful framework for studying complex systems composed of many interacting, autonomous components. In this approach, individual agents – representing entities such as people, biological organisms, or computational units – are modelled with simple behavioural rules, and their interactions are simulated to observe how collective behaviour emerges at the system level. Our research places particular emphasis on modelling the spread of infectious diseases, where agent-based approaches capture how individual behaviours, contact patterns, and mobility shape epidemic dynamics. More broadly, we develop and apply agent-based and multi-agent simulation techniques to investigate complex adaptive systems across social, biological, and technological domains, integrating data-driven methods and examining how agent behaviours, network structures, and environmental factors jointly influence system dynamics.
This research has significant impact in improving our understanding and management of epidemic disease. Agent-based simulations enable detailed exploration of how diseases spread through populations, supporting the evaluation of interventions such as vaccination strategies, social distancing policies, and contact tracing. These tools provide policymakers and public health authorities with actionable insights for mitigating outbreaks and enhancing preparedness for future pandemics. Beyond epidemiology, our work contributes to the design and analysis of resilient and adaptive systems more broadly, including infrastructure, financial systems, and distributed technologies, by revealing how local interactions give rise to system-level outcomes and how these can be guided through targeted interventions.
Professir Mikhail Prokopenko, Associate Professor Mahendra Piraveenan
Critical phenomena and phase transitions describe how complex systems can undergo abrupt changes in behaviour as underlying conditions vary. This includes transitions to crashes in financial markets as market senitment changes, and to epidemic spread as diseases mutate. Our research applies these ideas to complex systems by investigating how collective behaviour arises near critical points, where systems exhibit heightened sensitivity, large-scale correlations, and rich information dynamics. In particular, we study how phase transitions manifest in networked and adaptive systems, including neural systems and socio-technical systems, and how information-theoretic measures can be used to detect and characterise these regimes.
This research provides important insights into the mechanisms underlying sudden transitions and emergent behaviour in real-world systems. Identifying and characterising critical transitions helps anticipate tipping points in systems such as financial markets and infrastructure networks, enabling earlier detection of instability and improved risk management. In neuroscience, it contributes to understanding how brain dynamics may operate near critical regimes that support flexible and efficient information processing, and how deviations from such regimes may be linked to dysfunction. These advances also inform the design of adaptive and resilient technologies, by leveraging criticality to enhance responsiveness, robustness, and computational capability in complex engineered systems.
Professor Mikhail Prokopenko, Dr Michael Harre, Associate Professor Joseph Lizier
Complex networks are all around us, arising in systems such as neural connections in the brain, interactions between genes in regulatory systems, and relationships in real and online social networks. Complex network science uses tools from graph theory to analyse the structure of these systems, which are typically large-scale and exhibit non-trivial patterns of connectivity. Our research investigates how network structure shapes system behaviour, including the visualisation of complex networks, the development of new measures to characterise structural features such as node importance, methods to infer network structure from time-series data, and approaches to link structure with functional properties such as stability and synchronisation.
Our research enables deeper understanding of how real-world networks are organised and how their structure influences function and behaviour. By applying complex network analysis to domains such as neuroscience, we uncover how patterns of connectivity contribute to brain function and dysfunction, including conditions such as epilepsy. More broadly, these approaches provide tools to analyse, predict, and potentially control complex systems across biological, technological, and social domains, supporting advances in areas ranging from healthcare to infrastructure and communication systems.
Associate Professor Mahendra Piraveenan, Associate Professor Joseph Lizier, Professor Seokhee Hong, Professor Peter Eades, Professor Albert Zomaya
Collective behaviour in complex systems arises from interactions between many individual components. Information theory, originally developed to analyse communication systems, is increasingly used as a rigorous mathematical framework to quantify how information is stored, transferred, and transformed within such systems. Our research develops new theoretical approaches for applying information theory to complex systems, enabling the analysis of how information flows shape the state updates of individual components and, in turn, system-level behaviour. We apply these methods across diverse domains, including the brain and financial markets, to uncover the mechanisms by which distributed information processing gives rise to collective dynamics.
Our research provides fundamental insights into how information processing underpins function in real-world complex systems. In neuroscience, it has revealed how patterns of information flow support cognitive tasks and how these processes are altered in conditions such as autism spectrum disorder. More broadly, these approaches establish principled ways to analyse and design distributed systems, contributing to the development of artificial systems, including collective robotics and adaptive technologies, that exhibit robust, efficient, and nature-inspired information processing capabilities.
Associate Professor Joseph Lizier, Professor Mikhail Prokopenko, Dr Michael Harre, Professor Albert Zomaya
