Measuring system resilience through a comparison of information- and flow-based network analyses

Abstract

AbstractQuantifying the properties of complex, self-organizing systems is increasingly important for understanding the development and state of modern systems. Case studies have recommended sustainability frameworks predominately in literature, but little emphasis has been placed on methodological evaluation. Data availability is often an obstacle that constrains conventional flow-based network analysis, but a novel information-based technique (QtAC) developed by zu Castell and Schrenk overcomes these constraints by modelling interactions between agents as information transfers. This study compares the QtAC method to conventional flow analysis by applying both to the same 90-year dataset containing socio-economic data from the island of Samothraki, Greece. Resilience indicators, based on Ulanowicz’s ascendency analysis, are derived on both the information- and flow-based networks. We observe that the resulting dynamics of the information-based networks align closer with complex system dynamics as theorized by the adaptive cycle model. Additionally, we discuss how QtAC offers different interpretations of network indicators when compared to usual interpretations of flow analysis. Ultimately, QtAC is shown to provide an alternative for complex systems analysis if the data situation does not allow for conventional flow-analysis. Furthermore, we show that the combination of both approaches can yield valuable new insights.