An integrated model for locating backup facilities in flood disaster considering supportive strategies and conditional value at risk measure
Amirhossein Saffarinia,Farnaz Barzinpour,Ahmad Makui,Mohammad Reza Fathi
DOI: https://doi.org/10.3934/jimo.2024125
2024-09-20
Journal of Industrial and Management Optimization
Abstract:This research is conducted with the aim of proposing an integrated mathematical model for backup facilities in the area of accommodation and supply in case of flood disaster. We consider a combination of supportive strategies, such as construction primary and backup facilities, equipment and additional allocation to increase reliability of humanitarian logistics. A multi-objective, multi-period, and multi-scenario model is proposed to locate and allocate facilities in two preparation and response phases. In the developed model, the cost fluctuations are minimized via identifying Conditional Value at Risk (CVaR) as the objective function. Further, every decision-maker is capable of managing a percentage of long-term risks by adopting their confidence level. This model can also affect the recovery phase by using supportive strategies. In the proposed hybrid solution approach, the location-allocation criteria in the humanitarian logistics are classified based on their most related solution method. Thus, all the effective criteria are taken into account, and less important criteria are excluded. Given the safety criteria, a Geographic Information System (GIS) is also considered in this model. Finally, after conducting sensitivity analysis on supportive strategies like backup facilities strategy, management solutions are suggested to significantly decrease the severity and amount of disaster damages. The results reveal that backup facilities have an effective role in increasing the reliability of the chain, which can be greatly increased by combining with other supportive strategies. Also, adding the CVaR measure while considering scenarios with a low probability of occurrence and high impact, will reduce costs. The results show that the final model can reduce the costs by an average of 52%.
engineering, multidisciplinary,operations research & management science,mathematics, interdisciplinary applications