Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GREAT-EER: Graph Edge Attention Network for Emergency Evacuation Responses

Attila Lischka, Balázs Kulcsár

TL;DR

This work identifies and proposes the Bus Evacuation Orienteering Problem (BEOP), an NP-hard combinatorial optimization problem with the goal of evacuating as many people from an affected area by bus in a short, predefined amount of time and achieves near-optimal solution quality.

Abstract

Emergency situations that require the evacuation of urban areas can arise from man-made causes (e.g., terrorist attacks or industrial accidents) or natural disasters, the latter becoming more frequent due to climate change. As a result, effective and fast methods to develop evacuation plans are of great importance. In this work, we identify and propose the Bus Evacuation Orienteering Problem (BEOP), an NP-hard combinatorial optimization problem with the goal of evacuating as many people from an affected area by bus in a short, predefined amount of time. The purpose of bus-based evacuation is to reduce congestion and disorder that arises in purely car-focused evacuation scenarios. To solve the BEOP, we propose a deep reinforcement learning-based method utilizing graph learning, which, once trained, achieves fast inference speed and is able to create evacuation routes in fractions of seconds. We can bound the gap of our evacuation plans using an MILP formulation. To validate our method, we create evacuation scenarios for San Francisco using real-world road networks and travel times. We show that we achieve near-optimal solution quality and are further able to investigate how many evacuation vehicles are necessary to achieve certain bus-based evacuation quotas given a predefined evacuation time while keeping run time adequate.

GREAT-EER: Graph Edge Attention Network for Emergency Evacuation Responses

TL;DR

This work identifies and proposes the Bus Evacuation Orienteering Problem (BEOP), an NP-hard combinatorial optimization problem with the goal of evacuating as many people from an affected area by bus in a short, predefined amount of time and achieves near-optimal solution quality.

Abstract

Emergency situations that require the evacuation of urban areas can arise from man-made causes (e.g., terrorist attacks or industrial accidents) or natural disasters, the latter becoming more frequent due to climate change. As a result, effective and fast methods to develop evacuation plans are of great importance. In this work, we identify and propose the Bus Evacuation Orienteering Problem (BEOP), an NP-hard combinatorial optimization problem with the goal of evacuating as many people from an affected area by bus in a short, predefined amount of time. The purpose of bus-based evacuation is to reduce congestion and disorder that arises in purely car-focused evacuation scenarios. To solve the BEOP, we propose a deep reinforcement learning-based method utilizing graph learning, which, once trained, achieves fast inference speed and is able to create evacuation routes in fractions of seconds. We can bound the gap of our evacuation plans using an MILP formulation. To validate our method, we create evacuation scenarios for San Francisco using real-world road networks and travel times. We show that we achieve near-optimal solution quality and are further able to investigate how many evacuation vehicles are necessary to achieve certain bus-based evacuation quotas given a predefined evacuation time while keeping run time adequate.
Paper Structure (25 sections, 1 theorem, 3 equations, 9 figures, 9 tables, 2 algorithms)

This paper contains 25 sections, 1 theorem, 3 equations, 9 figures, 9 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Machine Learning-Based Methods to Solve Orienteering Problems
  4. Capacitated Orienteering Problem
  5. Bus Evacuation Problem
  6. OP for Emergency Evacuation
  7. Contributions
  8. Problem Formulation: Bus Evacuation Orienteering Problem
  9. Optimization Program for the BEOP
  10. Markov Decision Process for the BEOP
  11. Solution Method
  12. GREAT-EER Framework
  13. Reinforcement Learning
  14. Experiments
  15. Environment
  16. ...and 10 more sections

Key Result

Theorem 1

The Bus Evacuation Orienteering Problem with satisfied triangle-inequality for the travel times is NP-hard.

Figures (9)

  • Figure 1: GREAT-EER Framework for the BEOP: a BEOP instance is transformed into a graph that encodes the pairwise shortest paths between all relevant nodes. Then, the GREAT encoder is used to process this edge-based graph. Finally, a pointer network iteratively selects nodes to visit and their order. The resulting tour is returned as the output.
  • Figure 2: Simplified visualization of the MDP for BEOP showing the different action cases.
  • Figure 3: Validation performance (evacuation quota) per epoch for GREAT-EER models with 100 evacuation points
  • Figure 4: Validation performance (evacuation quota) per epoch for GREAT-EER model with 20 evacuation points
  • Figure 5: Example evacuations generated by different solution methods in San Francisco given 1.5h evacuation time and 3 buses for evacuation with 50 seats each.
  • ...and 4 more figures

Theorems & Definitions (1)

  • Theorem 1