Prioritized Reconstruction of Healthcare Facilities after Earthquakes based on Recovery of Emergency Services

This repository contains the code developed for the paper:

Prioritized Reconstruction of Healthcare Facilities after Earthquakes based on Recovery of Emergency Services. In Review.

Please cite the paper if you use any data or analyses from this study.

Running the Code

• Code/greedy_hospital.py: Main Python file containing all functions to run the Greedy Algorithm
• Code/greedy_hospital_tmp.py: Main Python file containing all functions to run the Greedy Algorithm considering temporary facilities

Test One Scenario

To obtain the reconstruction order of a single scenario, open Code/lima_scenario.ipynb and run the notebook. This Jupyter Notebook runs a single test scenario using data in Data/lima_test.csv.

Section 4: Evaluation of Greedy Algorithm

To run the results from Section 4 of the paper (Evaluation of Greedy Algorithm), run Code/greedy_evaluation_mpi.py. Example command for 6 damaged buildings and 12 total buildings: python 3 greedy_evaluation_mpi.py 6 12 'output_folder/' 1 (1 represents simulation idx)

To run multiple simulations at the same time, run Code/greedy_eval_local.sh in the command line. Change parameters (number of damaged building, number of total building, and output folders) as needed.

To visualize the results of the evaluation, run Code/greedy_eval_visualize.ipynb. Make sure to change the folder name containing the output folders from greedy_eval_local.sh.

Section 6: Results and Discussion

To run the results found from Section 6 of the paper (Results and Discussion):

1. Run all 10,000 earthquake scenarios with Code/lima_simulation_mpi.py. Example command: python3 lima_simulation_mpi.py 0 9999 'lima_simulations.csv' 'lima_emergency_data.csv' 'travel_times.json' output_folder/'.
2. Run the Jupyter Notebook Code/lima_paper_results.ipynb