The incorporation of artificial intelligence (AI) and robust optimization methods for the planning and design of relief logistics networks under relief demand–supply uncertainty appears promising for intelligent disaster management (IDM). This research proposes a data-driven hybrid scenario-based robust (SBR) method for a mixed integer second-order cone programming (MISOCP) model that integrates machine learning with a hybrid robust optimization approach to address the above issue. A machine learning technique is utilized to cluster the casualties based on location coordinates and injury severity score. Moreover, the hybrid SBR optimization method and robust optimization based on the uncertainty sets technique are utilized to cope with uncertain parameters such as the probability of facility disruption, the number of wounded individuals, transportation time, and relief demand. Additionally, the epsilon-constraint technique is applied to seek the solution for the bi-objective model. Focusing on a real case (the Kermanshah disaster), our analytical results have demonstrated not only the validity but also the relative merits of the proposed methodology against typical stochastic and robust optimization approaches. Besides, the proposed method shows all casualties can be efficiently transported to receive medical services at a fair cost, which is crucial for disaster management.

Identifying and characterizing post-disaster isolated areas are critical to the success of large-scale disaster management. A post-disaster isolated area (PDIA) refers to an area that can hardly be reached because of the destruction of traffic networks amid a disaster. Lacking relief and medical resources also inflicts psychological impacts on vulnerable dwellers in a PDIA. We believe humanitarian relief can be planned prior to disaster devastation. If a connected area has installed a relief facility, such as a hospital, the road damage may not severely affect the residents in PDIAs. This study enables the exploration of PDIAs characterized by the possibility of disaster occurrence and social vulnerability; and moreover, identifying the size of connected areas, also called responsible areas, to facilitate emergency relief supply and distribution in PDIAs amid a disaster. This research represents a joint venture with a national-level government agent, targeting a highly vulnerable territory that permits, efficiently and effectively, identifying and characterizing PDIAs from the perspective of social vulnerability. We adopt multi-data sources that incorporate socioeconomic, geographic, and disaster impact data gained and inputted from a national earthquake impact information platform. By conceptualizing and incorporating a syncretic disaster-risk index into the clustering metric, managerial endeavor becomes possible. We find that the chosen sizes of responsible areas of PDIAs are decisive, and by managing to maintain at least one relief facility in each PDIA, the impact on the dwellers can be mitigated.

When a disaster strikes (e.g., COVID-19), we observe "forward" buying from the retailers and "reactionary" buying from the consumers. These two buying activities are intricately intertwined with the retailer in the middle. In response to the disaster, the retailers buy forward on the upstream side, and the consumers on the downstream engage their reactionary buying. Our study investigates how these two buying behaviors correlate with market resilience under the moderating effects of self-regulation and heteronomy. Based on the samples of 136 personal protective equipment (PPE) retail firms and 140 end-customers, our empirical results indicate that heteronomy has a significant influence on alleviating the effects of forward buying and reactionary buying on market resilience. Furthermore, the reverse effect of market resilience on forward buying is verified. We provide managerial implications of the correlation between the two disaster-induced buying behaviors and market resilience under the moderating effects of self-regulation and heteronomy.