Sustainable Reconstruction Planning from Natural Disasters (Earthquakes): A Systematic Mapping Study of Machine Learning and Technological Approaches

Natural disasters have various adverse effects on human lives, making it challenging for authorities to manage post-disaster situations with limited resources. Due to the extreme extent of the damage, the huge amount of resources needed to restore life to normality makes such a situation challenging. For this purpose, different methodologies have been proposed to effectively handle these types of situations. All these methodologies consider different aspects of the post-earthquake context, taking into account core parameters such as the time and cost required for reconstruction, as well as the people directly affected by the earthquake. In this paper, we conduct a Systematic Literature Review (SLR) of various state-of-the-art techniques proposed for different phases of post-earthquake situations, specifically for reconstruction planning with sustainability considerations. All these proposed solutions are differentiated on the basis of input data, parameters, and type of solutions (data sciences, civil engineering, socio-economics, and modelling). The time range chosen to filter out relevant studies is between 2000 and 2025. Eventually, we reviewed 55 related articles out of 47,539 analysed from seven different digital libraries. The findings of this SLR reveal that optimization and simulation-based approaches dominate the current research landscape, with a growing trend toward data-driven and AI-assisted reconstruction planning. However, only a few studies focus on integrating socio-economic, environmental, and physical infrastructure aspects, which represents a major research gap. These findings provide insights that can guide future researchers in designing more comprehensive frameworks to improve post-earthquake reconstruction in a sustainable manner by prioritising economic, social, and environmental infrastructures, as well as facilities for affected individuals, thereby utilising available resources more effectively.