Determinants of Risk Disparity Due to Infrastructure Service Losses in Disasters: A Household Service Gap Model.

The objective of this article is to systematically assess and identify factors affecting risk disparity due to infrastructure service disruptions in extreme weather events. We propose a household service gap model that characterizes societal risks at the household level by examining service disruptions as threats, level of tolerance of households to disruptions as susceptibility, and experienced hardship as an indicator for the realized impacts of risk. The concept of "zone of tolerance" for the service disruptions was encapsulated to account for different capabilities of the households to endure the adverse impacts. The model was tested and validated in the context of power outages through survey data from the residents of Harris County in the aftermath of Hurricane Harvey in 2017. The results show that households' need for utility service, preparedness level, the existence of substitutes, possession of social capital, previous experience with disasters, and risk communication affect the zone of tolerance within which households cope with service outages. In addition, sociodemographic characteristics, such as race and residence type, are shown to influence the zone of tolerance, and hence the level of hardship experienced by the affected households. The results reveal that population subgroups show variations in the tolerance level of service disruptions. The findings highlight the importance of integrating social dimensions into the resilience planning of infrastructure systems. The proposed model and results enable human-centric hazards mitigation and resilience planning to effectively reduce the risk disparity of vulnerable populations to service disruptions in disasters.

Ensuring Access to Opioid Treatment Program Services Among Delawareans Vulnerable to Flooding.

Objective: To quantify potential flood-related access disruptions to medication-assisted treatment for opioid use disorder (OUD) among Delawareans. Methods: Spatial flood risk maps and infrastructure, services, and hazard risk, transportation networks, opioid treatment programs (OTPs) for the State of Delaware were integrated to visually display the relationship between these layers. A complex network theory-based simulation model was used to assess both direct (e.g., inundation with flood water) and indirect (e.g., isolation) impacts of floods. Results: Delaware is at increasing risk from flooding associated with storms and sea-level rise, which can lead to sunny day flooding during high tides. Of the 18 OTPs in Delaware, 4 are expected to be flooded in a 100-year flood and 7 are expected to be severely disrupted, increasing to 9 by 2035 and to 10 by 2050, with service reachability less than 15 square miles due to flood-induced isolation. Conclusions: Individuals utilizing OTPs for OUDs must be able to access treatment programs regardless of external disruptors like floods. Because these programs require consistent treatment adherence and in-person oversight by clinicians, timely restoration of services and continuity of operations for treatment facilities in post-disaster settings is critical for treatment compliance. Policy Implications: The State of Delaware has the third highest rate of drug overdose mortality in the U.S., with three-quarters of all drug-related deaths involving opioids. Impeded access to opioid treatment during a flood disaster can lead to relapse, overdose, and death. Hazard planning must develop policies and practices to address these risks.

Modeling of inter-organizational coordination dynamics in resilience planning of infrastructure systems: A multilayer network simulation framework.

This paper proposes and tests a multilayer framework for simulating the network dynamics of inter-organizational coordination among interdependent infrastructure systems (IISs) in resilience planning. Inter-organizational coordination among IISs (such as transportation, flood control, and emergency management) would greatly affect the effectiveness of resilience planning. Hence, it is important to examine and understand the dynamics of coordination in networks of organizations within and across various systems in resilience planning. To capture the dynamic nature of coordination frequency and the heterogeneity of organizations, this paper proposes a multilayer network simulation framework enabling the characterization of inter-organizational coordination dynamics within and across IISs. In the proposed framework, coordination probabilities are utilized to approximate the varying levels of collaboration among organizations. Based on these derived collaborations, the simulation process perturbs intra-layer or inter-layer links and unveils the level of inter-organizational coordination within and across IISs. To test the proposed framework, the study examined a multilayer collaboration network of 35 organizations from five infrastructure systems within Harris County, Texas, based on the data gathered from a survey in the aftermath of Hurricane Harvey. The results indicate that prior to Hurricane Harvey: (1) coordination among organizations across different infrastructure systems is less than the coordination within the individual systems; (2) organizations from the community development system had a low level of coordination for hazard mitigation with organizations in flood control and transportation systems; (3) achieving a greater level of coordination among organizations across infrastructure systems is more difficult and would require a greater frequency of interaction (compared to within-system coordination). The results show the capability of the proposed multilayer network simulation framework to examine inter-organizational coordination dynamics at the system level (e.g., within and across IISs). The assessment of inter-organizational coordination within and across IISs sheds light on important organizational interdependencies in IISs and leads to recommendations for improving the resilience planning process.

Robust component: a robustness measure that incorporates access to critical facilities under disruptions.

The objective of this paper is to integrate the post-disaster network access to critical facilities into the network robustness assessment, considering the geographical exposure of infrastructure to natural hazards. Conventional percolation modelling that uses generating function to measure network robustness fails to characterize spatial networks due to the degree correlation. In addition, the giant component alone is not sufficient to represent the performance of transportation networks in the post-disaster setting, especially in terms of the access to critical facilities (i.e. emergency services). Furthermore, the failure probability of various links in the face of different hazards needs to be encapsulated in simulation. To bridge this gap, this paper proposed the metric robust component and a probabilistic link-removal strategy to assess network robustness through a percolation-based simulation framework. A case study has been conducted on the Portland Metro road network during an M9.0 earthquake scenario. The results revealed how the number of critical facilities severely impacts network robustness. Besides, earthquake-induced failures led to a two-phase percolation transition in robustness performance. The proposed robust component metric and simulation scheme can be generalized into a wide range of scenarios, thus enabling engineers to pinpoint the impact of disastrous disruption on network robustness. This research can also be generalized to identify critical facilities and sites for future development.