Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

An Optimized Evacuation Plan for an Active-Shooter Situation Constrained by Network Capacity

Joseph Lavalle-Rivera, Aniirudh Ramesh, Subhadeep Chakraborty

TL;DR

This work tackles evacuations during active-shooter events under capacity constraints by extending a Non-Homogeneous Semi-Markov Decision Process (NHSMDP) with a capacitated, multi-route routing framework called C-CASTERS. The method computes and sequentially reserves multiple safe routes while accounting for node/edge capacities and a dynamic shooter threat, and it is evaluated via discrete-event simulation on two building layouts (acyclic and cyclic). Across topology, evacuee distributions, and shooter behaviors, C-CASTERS consistently reduces casualties and time-in-shooting-range exposure compared with Naive ASTERS and Natural Response, and it mitigates crowding at bottlenecks. The results demonstrate the practical potential of capacity-aware reinforcement learning for real-time guidance in high-stakes evacuations, with implications for building design and emergency response planning.

Abstract

A total of more than 3400 public shootings have occurred in the United States between 2016 and 2022. Among these, 25.1% of them took place in an educational institution, 29.4% at the workplace including office buildings, 19.6% in retail store locations, and 13.4% in restaurants and bars. During these critical scenarios, making the right decisions while evacuating can make the difference between life and death. However, emergency evacuation is intensely stressful, which along with the lack of verifiable real-time information may lead to fatal incorrect decisions. To tackle this problem, we developed a multi-route routing optimization algorithm that determines multiple optimal safe routes for each evacuee while accounting for available capacity along the route, thus reducing the threat of crowding and bottlenecking. Overall, our algorithm reduces the total casualties by 34.16% and 53.3%, compared to our previous routing algorithm without capacity constraints and an expert-advised routing strategy respectively. Further, our approach to reduce crowding resulted in an approximate 50% reduction in occupancy in key bottlenecking nodes compared to both of the other evacuation algorithms.

An Optimized Evacuation Plan for an Active-Shooter Situation Constrained by Network Capacity

TL;DR

This work tackles evacuations during active-shooter events under capacity constraints by extending a Non-Homogeneous Semi-Markov Decision Process (NHSMDP) with a capacitated, multi-route routing framework called C-CASTERS. The method computes and sequentially reserves multiple safe routes while accounting for node/edge capacities and a dynamic shooter threat, and it is evaluated via discrete-event simulation on two building layouts (acyclic and cyclic). Across topology, evacuee distributions, and shooter behaviors, C-CASTERS consistently reduces casualties and time-in-shooting-range exposure compared with Naive ASTERS and Natural Response, and it mitigates crowding at bottlenecks. The results demonstrate the practical potential of capacity-aware reinforcement learning for real-time guidance in high-stakes evacuations, with implications for building design and emergency response planning.

Abstract

A total of more than 3400 public shootings have occurred in the United States between 2016 and 2022. Among these, 25.1% of them took place in an educational institution, 29.4% at the workplace including office buildings, 19.6% in retail store locations, and 13.4% in restaurants and bars. During these critical scenarios, making the right decisions while evacuating can make the difference between life and death. However, emergency evacuation is intensely stressful, which along with the lack of verifiable real-time information may lead to fatal incorrect decisions. To tackle this problem, we developed a multi-route routing optimization algorithm that determines multiple optimal safe routes for each evacuee while accounting for available capacity along the route, thus reducing the threat of crowding and bottlenecking. Overall, our algorithm reduces the total casualties by 34.16% and 53.3%, compared to our previous routing algorithm without capacity constraints and an expert-advised routing strategy respectively. Further, our approach to reduce crowding resulted in an approximate 50% reduction in occupancy in key bottlenecking nodes compared to both of the other evacuation algorithms.
Paper Structure (28 sections, 18 figures, 4 tables, 2 algorithms)

This paper contains 28 sections, 18 figures, 4 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Literature Survey
  3. Methodology
  4. NHSMDP States
  5. NHSMDP Actions and Transition Probabilities
  6. NHSMDP Reward Structure
  7. Evacuation Algorithm - Capacitated Value Iteration
  8. Experimental Design
  9. Simulation Environment
  10. Shooter Actions
  11. Evacuee Actions
  12. Graph Environments
  13. Acyclic School
  14. Cyclic School
  15. Parameters and Simulation Inputs
  16. ...and 13 more sections

Figures (18)

  • Figure 1: Possible Actions in $node_1$
  • Figure 2: Transition Probability for a Given Action $a_i$
  • Figure 3: Probability Transition Example
  • Figure 4: Two Different Graphical Layouts for Simulation Comparison
  • Figure 5: Escape Reward Value Affect Over Time - Cyclic School Floor: Rooms and Halls Distribution
  • ...and 13 more figures