Modelling infrastructure interdependency at a local scale: value, methodologies and challenges

Natural hazards such as floods, bushfires, cyclones or hurricanes, can cause significant damage that disrupts our infrastructure systems. Climate change is making such extreme events more frequent and more severe. In addition, infrastructure systems have become more interconnected and interdependent. By interconnected, we mean that infrastructure systems use each other's output and operate together to provide joint services. The interdependence of two systems or components, on the other hand, refers to the effect of a decline of the performance of one system or component on another. Hence, a disruption to one infrastructure system may propagate into others and eventually affect various services that are critical for the well-being of communities. In recent years, cascading failures of infrastructure systems at a national scale have attracted significant attention. However, our understanding of these failures at a local scale remains limited. During major disasters, communities frequently face the challenge of disruptions that are locally triggered and propagated. Therefore, applying a systems-based analytical approach is likely to be beneficial for local authorities when making decisions on infrastructure planning and operations. To this end, the ability to model inter-connectedness of local infrastructure systems needs to be strengthened. The goal of this paper is to discuss the value of applying a systems approach to the analysis of infrastructure systems and their interdependencies and present and appraise various methodologies that can be used in implementing such as an approach. The paper is divided into four parts. First, fundamental elements of interdependency analysis are presented. The differences between interdependency modelling at national and local scales are discussed and illustrated. One important difference is that coarse spatial resolution of models at a national scale prevents them from yielding information that is useful for locally-triggered and locally-propagated disruptions. For example, in a national-scale model, the entire water or power distribution system for a local government area may be represented by a single node reflecting its role in, and effect on, the national network, without incorporating an understanding of its internal operation. Second, we describe the likely decision-making contexts within which infrastructure planning and management are conducted. Different local stakeholders with diverse concerns related to investment, planning, designing, and operations and management of infrastructures are usually involved in, and/or affected by, infrastructure decision-making. These stakeholders include local governments, infrastructure operators and utility companies, local businesses, households, individuals and local communities. Third, we review different methodologies available for modelling infrastructure interdependencies at a local scale. We discuss their suitability in light of the specific analysis goals and identify their respective strengths and weaknesses. We propose a preliminary framework for selecting appropriate methods for modelling infrastructure performance depending on the underlying decision-making context. Finally, we discuss important challenges present when tackling interdependency issues at a local scale. These challenges include lack of a comprehensive methodological framework linking different models, lack of a consistent benchmark and relevant datasets for judging the suitability of a method, and lack of a better understanding of the broader socio-economic impacts of infrastructure disruptions on local communities.