Probabilistic analysis of interdependent infrastructures subjected to weather-related hazards

Simulation of the infrastructure performance using numerical models may significantly assist in developing strategies for improving its resilience to harmful effects of climate change. This paper presents a model for simulating the performance of interdependent infrastructure systems based on an extended network flow approach, i.e. infrastructure systems are considered as a network of nodes connected by directed edges. The model has been specifically developed to simulate the infrastructure performance at a community scale and has higher node resolution compared to typical models of infrastructure systems at the national level. The model is time-dependent so that the infrastructure performance can be assessed at discrete points over a period of time. Parameters describing the performance of infrastructure assets (e.g. production and flow capacities, demands) can be treated as random variables or probabilities can be assigned to failures of the assets. The application of the model is illustrated by the probabilistic assessment of the performance of two interdependent infrastructure systems – electrical power and water, damaged by flooding.