Evaluating Routing Strategies for Emergency Evacuation: A Spatially Explicit Agent-Based Modeling Approach

Code Base Documentation

By Rebecca Vandewalle, Furqan Baig, Santiago Nunez-Corrales, Jeon Young Kang, and Shaowen Wang

This repository contains code needed to run an agent-based emergency evacuation simulation in Python. Agents can use one of 3 routing strategies to determine a path from their initial location to out of the evacuation zone. Simultaneously, a wildfire object can spread throughout the simulation and close roads.

Start here

Run FireABM_Demo_Notebook.ipynb for an overview of how the simulation works.

Launch with CyberGISX

Purpose

This code directory contains code needed to replicate the experiments using a spatially-explicit agent-based model of wildfire evacuation in a forthcoming manuscript by Vandewalle, Kang, and Wang, as well as companion documentation and a code demonstration Jupyter notebook.

The main code base is flexible and can serve a variety of purposes; not all available parameters are used in the manuscript.

Due to the time taken, everything needed to run the same experiments is included in this repo, but the actual simulations runs for the experiment are intended to be run on HPC such as Keeling/Virtual Roger.

What does the code do

The code models the process of evacuation on a road network in which roads are progressively closed by wildfire spread. Individual households, represented by a vehicle, must navigate out of the danger zone and re-route if the road they are currently on becomes blocked by the wildfire. The forthcoming manuscript specifically looks at patterns in evacuation clearance and congestion that change based on how vehicle routing decisions are modeled. Specifically three driving strategies are compared, 2 based off of common modeling assumptions (quickest path and shortest path), and one that attempts to more closely model evacuee behavior (preference for major roads). These strategies are described in more detail in the manuscript and the demonstration notebook.

Code and documentation contents

This package contains all files needed to run experiments and generate output used in the forthcoming manuscript.

How to run the Jupyter Notebook

You can just copy this repository to your own Jupyter notebook file system, install any missing dependencies (see requirements.txt, and run the notebook.

An easier way is to open the notebook directly with CyberGISX (requires registration) by clicking this link: Launch with CyberGISX

To register for CyberGISX you will need a GitHub account and a working email address. Go to the registration page and fill out the form with your information. You may need to validate your email address. Once you have registered you will be able to directly open this notebook in CyberGISX using the above link.

This notebook and repository will also be available on the CyberGISX community notebook page to view and for users to run using CyberGISX (link TBD).

How to run a simulation on Keeling/Virtual Roger (HPC)

See 0_Replicating_the_manuscript_experimental_procedure.txt for steps taken to run simulations used for the manuscript

1. Make sure your directory structure is set up as expected (see 3_Expected_base_directory_structure.txt)
2. Make sure you have the required python libraries installed (see 4_Required_Python_packages.txt)
3. If on Keeling or another HPC environment, set up a virtual environment (see 5_Creating_a_virtual_environment_for_Osmnx.txt)
4. If you want to set initial agent positions by population, get households data from the US Census (see 6_Gathering_household_data.txt)
5. If you want to make your own wildfire simulation to use in the model, use FlamMap to generate a fire (see 7_Creating_a_simulated_wildfire_with_FlamMap.txt)
6. Determine which core script to use (see 8_Which_script_version_to_use_-_opt_or_opt_Keel.txt)
7. To run batch jobs determine which scripts you want to run (see 9_Running_batch_jobs.txt)
8. Determine which simulation parameters to use (see 10_Simulation_run_parameters.txt)

Simulation Output

Running the simulation will generate videos, such as the one at the beginning of this readme, as well as text results (see 11_Simulation_output_structure_and_explanation.txt). 12_Example_simulation_run_times.txt contains run times to help estimate how long a simulation might run. Finally, 13_Graph_Files_with_names_ending_in_orig.txt discusses which results files are provided.

Important files

See 1_Index_of_provided_files.txt for short description the provided files in the repo

• FireABM_Demo_Notebook.ipynb: the core notebook demonstrating how the simulation works
• combined_rslts_8_13.txt: Combined results file containing results from all the sumulation runs that were run on HPC
• Graph_Results_Final.ipynb: Use this notebook to create all of the figures used in the manuscript taking result data stored from combined_rslts_8_13.txt
• Rebuild_Road_Graphs.ipynb: This notebook demonstrates all preprocessing steps taken to prepare the road network, households data, and wildfire perimeters for use in the simulation