Multi-region ARIO analysis

Stanford Urban Resilience Initiative, 2023
Original program by Stephane Hallegatte, 2014
MATLAB codebase for quantifying post-disaster economic recovery for multi-regional analysis. Based on the original ARIO model by Hallegatte, 2014 and single-region ARIO model by Stanford Urban Resilience Initiative, 2022.

License: GNU General Public License v3.0 (see LICENSE)

Updates

08/11/2023:

• A new model parameter input.sub_ratio is introduced:
  • When input.sub_ratio = 1, the inventory from same sector but different regions are allowed to substitute each other for production;
  • Otherwise, the inventory from same sector but different regions are not allowed to substitute each other as per in the previous version.

08/15/2023:

• A hypothetical example is posted with required inputs and output from the analysis for reference.
• The preprocessing scipt is posted to process and format the input data from the example for multi-regional analysis.

07/23/2023:

• This program requires a set of pre-formatted inputs to run.

Program overview

This repository contains the following:

• mr_wrapper.m: wrapper script governing analysis settings
• run_mr_ario.m: wrapper function for multi-region ARIO program
• mr_ario_preprocessing.m: script for preprocessing (1) economic inputs, (2) damage/loss inputs, and (3) I-O inputs for multi-region ARIO program
• functions: functions for simulating economic recovery, and auxilaliary post-processing tools
• inputs: raw (1) I-O table, (2) region names, and (3) sector names pre-processed (1) economic inputs, (2) damage/loss inputs, (3) I-O inputs, and (4) behavioral parameter distributions
• output: analysis output

Example overview

This repository contains the data for a test-run example:

• The example assesses the impacts to a two-region economy (Region1 and Region2) from the initial damages.
• The example extracts the economic information of the analysis regions (Region1 and Region2) from a three-economy I-O table (for Region1, Region2, and Region3) using the preprocessing script.
• The output results of the example are provided in the outputs folder.

fn_run_mr_ario overview

The function fn_run_mr_ario is the main function governing the ARIO analysis, and is called by the program wrapper, sr_wrapper after analysis settings are defined.

• fn_load_default_ario_settings: loads hardcoded analysis parameters
• fn_initialize_mr_variables: initializes containers for all timesteps in the analysis. Sets initial, pre-disaster values where applicable.
• Main loop: for timestep $k = 1:n$:
  • fn_compute_demand: computes percentage of sector-level recovery, updates reconstruction demand rate and total final demand for the next time step
  • fn_compute_prod_lim_by_cap: computes production constrained by production capacity.
  • fn_compute_prod_lim_by_cap_sup: computes production constrained by production capacity and supplies available.
  • fn_get_output_econ_metrics: computes the actual satisfied demand and actual supply for the next time step.
  • fn_update_input_econ_metrics: update input economic metrics for the next time step.
  • fn_get_output_econ_performance: update output economic metrics for the next time step.

Input files

• mr_ario_building_damage.csv: contains individual damage observations/simulations for each building (or building cluster) in the simulated inventory. Used directly by the program to construct sector-specific recovery curves, control the rate of reconstruction for each sector.
• mr_ario_econ_data.mat: economic data for each sector in the economy, including:
  • exports
  • final demand
  • imports
  • value added
  • local demand
  • fixed assets
  • total output
• mr_ario_economic_sectors.csv: list of sectors modeled in the economy (should include housing, if available). Each sector's ID, name, and non_stockable good status must be included.
• mr_ario_IO_data.mat: raw IO matrix (non-normalized) to be used directly in the analysis.
• mr_ario_loss_data.mat: variables that contain sector-level aggregate losses and reconstruction demand assignment.

Output files

• mr_ario_results.mat: .mat file containing mr_output, a MATLAB struct containing the results of the analysis.

Running an analysis

The steps to run a multi-region ARIO analysis are summarized below.

1. Place (i) mr_ario_building_damage.csv, (ii) mr_ario_econ_data.mat, (iii) mr_ario_economic_sectors.csv, (iv) mr_ario_IO_data.mat, (v) mr_ario_loss_data.mat, and (vi) behavioral parameter distributions csv's in the inputs directory.
2. Run the mr_ario_wrapper.m script, which calls on the run_mr_ario.m function to initiate the ARIO analysis with user-defined settings.
3. Analysis will be saved in the outputs directory.