Macroscopic Dynamics in Complex Systems: Methods on collective change in social, physical and transportation systems

Abstract

Epidemics, fake news, train disruptions and the functioning of the brain: at first sight, these phenomena seem rather different from each other, but in reality, they have much in common. In these systems, the interaction of only a few parts can lead to much larger and possibly unexpected dynamics. Such behaviour evolves on so-called networks, such as webs of social interactions or highways. These systems are referred to as 'complex systems'. In this thesis, the dynamics of complex systems is studied using three core questions. The first question addresses how we can distinguish and predict states and transitions in complex systems. Using data of the Dutch railway system, brains of rodents and climate models, mathematical techniques are applied to show how railway disruptions, behavioural change and tipping points can be recognized. The second question concerns the impact of a system's network structure on its dynamics. The thesis describes how one can subdivide railway networks in Europe, how techniques from physics and epidemiology can be used to predict railway delays, and how local disruptions can evolve to national problems in the Netherlands. The third question addresses the quantification of vulnerability and resilience of systems against unwanted dynamics such as epidemics. Using high resolution data of social interactions at, among other systems, an art fair in Amsterdam, these quantities are determined and in a model of the first wave of COVID-19 in the Netherlands, we use these principles to evaluate governmental interventions to tackle the pandemic. Altogether, the thesis aims to contribute to our understanding of these systems, as well as to propose new methods to analyse them.