Digital twin-based decision support systems for natural disaster management: A systematic review of current trends and approaches

Abstract

Increased urbanization and extreme climate events in recent years have led to a rise in the frequency and severity of widespread natural disasters. The disaster management cycle is a common framework that helps emergency management organizations develop procedures that can minimize the devastating impacts of natural disasters. Recently, researchers have begun to explore how big data and artificial intelligence can be used to develop virtual digital twin (DT) models of disaster situations that provide real-time decision support for each phase of the disaster management cycle. This study presents a systematic literature review of the burgeoning research area of DT-based decision support systems for disaster management. We identify the various locations and disaster types featured in DT case studies, the data collection techniques and computational methods utilized to create virtual representations of disaster-affected areas, and the decision-making tasks that DTs have been applied to. Based on our findings, we highlight several research gaps in the current literature and provide a set of recommendations that can serve as guidelines for future studies. Overall, this review provides researchers and practitioners with insights into the current trends and future research directions for DT-based decision support systems for natural disaster management.