PDF Basket
The emerging discipline of ‘complex systems’ applies the laws of statistical physics to make sense of the behaviour of large systems made up of relatively simple parts.
Interest in such problems now extends beyond physics, spurred by the economic incentive to understand the sometimes surprising behaviour of systems such as power grids, transport networks and financial markets. And the growth of computing power and artificial intelligence is finally making them easier to deal with.
‘Some things have always been complex but we couldn’t make much sense of them before because we didn’t have the tools for doing it,’ says Martin Weigel of Coventry University, who coordinated the EU-funded DIONICOS project. ‘So that opens up the possibility of analysing systems quantitatively that we couldn’t analyse before.’
The project developed new tools for studying complex systems and explored how to apply them to novel areas of investigation. ‘A key element of DIONICOS was the idea of methodological cross-fertilisation between research in the physical sciences and applications in socio-economic disciplines and the humanities,’ Weigel says.
Iliad, Beowulf and Harry Potter
Just as it has been possible to understand complex, cooperative and dynamic behaviour within assemblies of atoms or molecules, it is now becoming feasible to apply similar methods to interactions between large numbers of human beings.
One intriguing example is the application of these tools to literature, part of the growing interest in ‘digital humanities’. Researchers in DIONICOS, led by Ralph Kenna from Coventry University, compared networks of characters in mythological texts to purely fictional networks like those in the Harry Potter stories and real-world networks such as scientific collaborations. This work led to the publication of a book in 2016.
‘The Iliad and Beowulf and so on have been around for centuries, if not millennia, but now researchers have the technology to analyse them in a slightly different way,’ Weigel notes. ‘That could be used as a means of deciding how fictional or how real some of the mythological texts are, as this is something that is disputed among historians.’
In another strand, researchers (led by Stefan Thurner from the Medical University in Vienna) analysed a popular online space-trading game, Pardus, to find out how large networks of real people interact with each other. ‘Studying human behaviour in virtual environments provides extraordinary opportunities for a quantitative analysis of social phenomena with levels of accuracy that approach those of the natural sciences,’ he explains.
Exciting research was also performed in the physical sciences. A project led by Jürgen Horbach (University of Düsseldorf), for example, provided the first-ever explanation of rigidity in solids. Other high-profile work included progress on understanding graphene, crumpled globules, knots in polymers, and the design of hard problems for quantum computers, to name but a few of the applications.
Global network
As well as holding more than a dozen scientific workshops and conferences, the four-year project organised talks for the general public and events to encourage researchers in other disciplines to explore how these techniques could help them.
The core project partners comprised eight universities in five countries while the wider DIONICOS network encompassed 19 research groups in 12 countries, including non-European states such as India, Mexico, the US and Venezuela.
Although the project, funded by the EU’s Marie Skłodowska-Curie actions programme, ended early in 2018, some bilateral collaborations are continuing.
Weigel explains that as the work of DIONICOS was at a fundamental level there are no immediate industrial applications. But outside the conventional areas of physics and engineering the project has demonstrated the potential for applying the methods of complexity science to socio-economic systems.
‘Examples include applications of social network analysis in the context of child exploitation networks, research into the effect of group sizes on research quality, and work on the optimisation of urban transport networks,’ he says.