Climate change is a pressing global challenge, resulting in rising temperatures, shifting precipitation patterns, and more frequent extreme weather events. Particularly in coastal regions, these climate change effects are leading to more frequent and intense natural hazards, such as tropical cyclones and flooding, which pose significant threats to civil infrastructure systems. Additionally, the changing environment also accelerates the aging and degradation of these systems, threatening their long-term performance and increasing potential risk across the built environment. In risk assessment, the expected long-term loss of systems caused by hazards has been commonly used as a standard criterion for decision-making in previous studies. However, under the profound influence of climate change, a single indicator (i.e., the expectation) is no longer sufficient to capture the increased uncertainties. Other statistical metrics such as median, standard deviation, skewness, and kurtosis should be evaluated. Furthermore, the probabilistic distribution of loss is needed to quantify these metrics for a more comprehensive decision-making process.
This study addresses this limitation by proposing a maximum entropy approach to assess the probabilistic long-term loss of infrastructure systems subjected to hazards under climate change. This analytical approach can efficiently evaluate these risk indicators, saving significant time by avoiding the heavy computations required by the conventional Monte Carlo simulation in risk assessment. The framework is applied to assess the long-term loss of a coastal bridge subjected to tropical cyclone-induced waves and storm surge. The long-term loss is evaluated by conducting vulnerability analysis of the bridge and risk assessment using a stochastic process model. The non-stationary characteristics of hazards due to climate change (i.e., changing frequency and intensity) are incorporated in the risk assessment based on the stochastic model. This framework is especially flexible in considering the impact of climate change on hazard frequencies and intensities, making it beneficial for decision-making to enhance the system resilience.