Quantifying hazards resilience by modeling infrastructure recovery as a resource constrained project scheduling problem

Abstract. Reliance on infrastructure by individuals, businesses, and institutions creates additional vulnerabilities to the disruptions posed by natural hazards. In order to assess the impacts of natural hazards on the performance of infrastructure, a framework for quantifying resilience is presented. This framework expands upon prior work in the literature to improve the comparability of the resilience metric by proposing a standardized assessment period. With recovery a central component of assessing resilience, especially in cases of extreme hazards, we develop a recovery model based upon an application of the resource constrained project scheduling problem (RCPSP). This recovery model offers the opportunity to assess flood resilience across different events and also, theoretically, between different study areas. The resilience framework and recovery model have been applied in a case study to assess the resilience of buildings infrastructure to flooding hazards in Alajo, a neighborhood in Accra, Ghana. The results show that for the three flood events investigated (5, 50 and 500-year return periods), the 300-day resilience of the buildings infrastructure in Alajo was quantified as 0.94, 0.82 and 0.69, respectively. In practical terms, each value reflects the ability or inability of the system to maintain its function during the reference period for the given flood event, with zero corresponding to a complete loss of function and one when unaffected. This information is valuable for identifying the vulnerabilities of buildings infrastructure, assessing the impacts resulting in reduced performance, coordinating responses to flooding events, and preparing for the subsequent recovery.