Resilience and safety of civil engineering systems and communities: A bibliometric analysis for mapping the state-of-the-art.

• A bibliometric analysis regarding resilience in civil engineering field was carried out. • 539 documents were found in Web of Science database for 1996–2020 period. • Research trends and main gaps were analyzed in eight sub-categories. • Recovery time and Downtime , Critical infrastructures , Probabilistic approach , and Community resilience are the fields where resilience gained the most attention. The interest in the concept of resilience has been growing consistently over the past few years to study the functionality and behavior of systems against natural and man-made hazards. Yet a comprehensive, updated review of methods and frameworks to assess and improve the resilience and safety of civil engineering systems and communities is lacking. In this paper, a bibliometric and visualization method is implemented to explore the status of resilience research in civil engineering applications by analyzing journal papers published from 1996 to 2020. The concept of resilience and safety is investigated through eight subject categories identified by the authors in the literature: recovery time strategies and downtime, critical infrastructures, probabilistic approaches, fuzzy logic approaches, structural health monitoring, health care facilities, emergency management and decision-making, community and urban resilience. Results show that resilience research has increased rapidly since its introduction, most notably in the past seven years. The analysis identifies two main research approaches: frameworks and conceptual models, and case study based. The latter is the most adopted methodology by the analyzed works. In terms of geographical distribution, most of them have been carried out in the USA, the United Kingdom, China, and Italy. The authors' keywords analysis reveals that recovery strategies, critical infrastructures, vulnerability, and community resilience and safety have attracted prominent attention in the past decade. Finally, we conclude that further multidisciplinary research is needed to model multi-hazard scenarios and cascading effects, to collect data, and to define new performance metrics. [ABSTRACT FROM AUTHOR]