A model for the synchronization of road clearing activities with multi-modal relief supply distribution after a natural disaster

Abstract

Natural disasters such as earthquakes can disrupt road networks, with some roads being covered with debris. Similar to the 2010 Haiti Earthquake and the 2010 Hurricane Igor, several communities may be isolated because of a lack of operational roads connecting them to the rest of the network. This isolation makes delivering relief supplies, such as food, medicine, and water, to the communities difficult. Hence, road clearing activities and the multi-modal distribution of relief supplies need to be coordinated. After road clearing teams reconnect the network by removing the debris, trucks can traverse the cleared roads to deliver the goods. In addition, if it is not possible to reach a community using trucks within a pre-determined time horizon, alternative transportations modes, such as helicopters and ships, may be used. Despite its importance for disaster preparedness and mitigation, the synchronization of road clearing activities with the multi-modal distribution of supplies has not received much attention in the literature. Therefore, this thesis aims to formulate, model, and efficiently solve the problem of coordinating road clearing with the multi-modal distribution of supplies. Special consideration is given to supplying coastal communities in an island context, so that the problem especially addresses road and maritime transport operations. First, this thesis proposes a new road clearing problem called the “Collaborative Multi-vehicle Prize Collecting Arc Routing for Connectivity Problem” (MPC-ARCP). In the MPC-ARCP, the road clearing teams start from multiple depots and reconnect isolated communities to critical infrastructures, such as ports and airports. The less resilient communities are prioritized to be reconnected first. A hybrid metaheuristic is introduced to solve the MPC-ARCP on large-scale networks. Then, this thesis presents the “Synchronized Road Clearing and Multi-modal Distribution of Relief Supplies Problem” (CIRCULATIONS), which maximizes the amount of relief supplies delivered to affected communities. A bi-level metaheuristic is introduced to solve this problem, which is applied to a real-world plausible scenario of a megathrust earthquake on the Cascadia Subduction Zone on Canada’s West Coast. The CIRCULATIONS’ results of the plausible medium impact Cascadia scenario indicate that barges will be crucial in transporting the supplies to Vancouver Island, whereas ferries are expected to deliver only to Texada Island as they are mainly non-operational in the initial 72 hours of the disaster response. In addition, 66% of the total demand is estimated to be delivered. Of this, barges and trucks will transport 65%, while helicopters will only deliver 1%. These results show the importance of combining two modes of transportation (i.e. marine and ground) to distribute as many supplies as possible to the affected communities. This thesis also discusses the study's limitations and presents several inquiry lines for future research.