This repository contains all of the code that is required to replicate the work that I undertook during my thesis on modelling mass evacuation of cities using agent based modelling. This is a breakdown of how the folders are organised.

• v1: Early experiments
• v2: Videos for an earlier version of MassEvacDemo
• v3: PRE2016, TGF2015 paper
• v4: Thesis, Flood paper, Topology paper

For all intents and purposes, the instructions provided for setup below primarily only apply to v4 as it contains the most up to date analyses presented in the thesis. The earlier versions v1, v2 and v3 have been kept primarily to preserve the record of the state of the analyses during various stages of paper submissions and they have not been maintained in quite some time. One day, if I have more time, I will look into those too.

There are various Python 2.7 dependencies which will need to be installed as listed in requirements.yml by running the following command:

conda env create -f requirements.yml
source activate massevac

A local install of OpenStreetMap geospatial data and GPWv4 population data on a PostgreSQL server is required. The instructions on how to do this are available in SetupDB.sh which has currently only been tested on Ubuntu 16.04 LTS.

You will need to add a file at v4/.dbconfig with your database credentials as follows:

{
 "host": "localhost",
 "user": "username",
 "password": "password",
 "dbname": "gis"
}

It is also necessary to install ffmpeg to produce visualisation of evacuation as follows:

conda install -c menpo ffmpeg
chmod +x $(which ffmpeg)

After all the dependencies are met, test run an agent based evacuation simulation by running the following in the terminal:

cd v4/
python 0A_abm_test.py