Given maps of an evacuee population, shelter destinations and a transportation network, the goal of intelligent shelter allotment (ISA) is to assign routes, exits and shelters to evacuees for quick and safe evacuation. ISA is societally important due to emergency planning and response applications in context of hazards such as floods, terrorism, fire, etc. ISA is challenging due to conflicts between movements of evacueegroups heading to different shelters and transportation-network choke-points. State of the practice based on Nearest Exit or Shelter (NES) paradigm addresses the former challenge but not the latter one leading to load-imbalance and slow evacuation. Recent computational development, e.g., capacity-constrained route planning (CCRP), address the latter challenges to speedup evacuation, but do not separate evacuee groups going to different shelter destinations. To address these limitations, we propose a novel approach, namely, Crowd-separated Allocation of Routes, Exits and Shelters (CARES) based on the core idea of spatial anomaly avoidance. Experiments and Hajj case study (Makkah) show that CARES meets both challenges by providing much faster evacuation than NES and much lower evacuee-group movement-conflicts than CCRP.

No abstract

No abstract

Given a graph and a set of service center nodes, a Capacity Constrained Network-Voronoi Diagram (CCNVD) partitions the graph into a set of contiguous service areas that meet service center capacities and minimize the sum of the shortest distances from graph-nodes to allotted service centers. The CCNVD problem is important for critical societal applications such as assigning evacuees to shelters and assigning patients to hospitals. This problem is NP-hard; it is computationally challenging because of the large size of the transportation network and the constraint that service areas must be contiguous in the graph to simplify communication of allotments. Previous work has focused on honoring either service area contiguity (e.g., Network Voronoi Diagrams) or service center capacity constraints (e.g., min-cost flow), but not both. Our preliminary work introduced a novel Pressure Equalizer (PE) approach for CCNVD to meet the capacity constraints of service centers while maintaining the contiguity of service areas. However, we find that the main bottleneck of the PE algorithm is testing whether service areas are contiguous. In this paper, we extend our previous work and propose novel algorithms that reduce the computational cost. Experiments using road maps from five different regions demonstrate that the proposed approaches significantly reduce computational cost for the PE approach.