Short-notice flood evacuation plan under dynamic demand in high populated areas. (May 2022)

• Record Type: Journal Article
• Authors: Insani, Nur; Akman, David; Taheri, Sona; Hearne, John
• Is Part Of: International journal of disaster risk reduction. Volume 74(2022)
• Publication Date: 2022-05
• Subjects: Flood evacuation -- Split delivery -- Multi-trips -- Closing time windows -- Early stage disaster; Emergency management -- Periodicals; Risk management -- Periodicals; Disaster relief -- Periodicals; Hazard mitigation -- Periodicals; 363.34
• DOI: 10.1016/j.ijdrr.2022.102844
• Languages: English
• ISSNs: 2212-4209

Abstract

Abstract: Driven primarily by climate change, natural disasters are expected to continue increasing throughout the years. A multitude of approaches for modelling disaster evacuation have been proposed for over decades and most of them focus on minimizing the travelling times or costs after a disaster strikes. However, evacuating all vulnerable people to safer places with sufficient vehicles before a disaster strikes is a better option. Owing to this operation's importance at such a crucial time, we have developed a short-notice flood evacuation model by utilising public transportation to evacuate an uncertain number of people, particularly in areas with a high population. The proposed model aims to find the minimum number of vehicles required and its sequence trips to save all evacuees before the water reaches the predicted areas. A closing time window at each evacuation point and evacuees-dependant service times are taken into account to tackle the flood propagation. Due to the limitation of available vehicles at the early stage of the disaster, the model considers splitting the demands and reusing the vehicles, i.e. multi-trips, to cover all the necessary routes within a time horizon. We apply the new model to some synthetic and real-world instances. In addition, we study the trade-off between the minimization of the number of vehicles and total travelling times. The results show that the proposed model has effectively saved more resources when split deliveries and multi-trips are imposed.