Batch RR2 Model Pipeline

This repository contains the batch_rr2 model pipeline to estimate Risk Rating 2.0 (RR2) flood insurance premiums using CPRA Coastal Master Plan 2023. This repository contains the batch_rr2 model pipeline to estimate FEMA’s Risk Rating 2.0 (RR2.0) flood insurance premiums at the structure level under multiple Coastal Master Plan scenarios.

📦 Installation

From the source (recommended)

bash git clone https://github.com/adilurrahim/batch_rr2.git cd batch_rr2 conda create -n batch_rr2 python=3.10 conda activate batch_rr2 pip install -r requirements.txt

Getting Started

Step 1: Project Structure

data/ ├── structure_csv/ │ └── mp23_pdd_clara_structure_info_costs_2024_06_18.csv ├── OccupancyMapping/ │ └── OccupancytoTypeofUseMapping.csv ├── InsurancePolicyDistribution/ │ ├── ZipCodeData/ │ │ └── tl_2021_us_zcta520.shp │ └── FEMAPolicyCounts/ │ └── fema_risk-rating-zip-breakdown_2021.csv ├── ProcessedData/ │ └── CommonData/ │ ├── County/ │ ├── CRS/ │ ├── FlowLine/ │ ├── RiverPolygon/ │ └── HUC12/ │ └── ScenarioSpecificData/ │ └── 2/ │ └── FWOA/ │ └── Lower/ │ ├── CoastLine/ │ ├── Elevation/ │ ├── FloodDepth/ │ └── Levee/

Step 2: Run the Pipeline

bash python scripts/main.py \ --years 12 22 32 42 52 \ --plans FWOA FWMP \ --scenarios Higher Lower \ --structure_csv data/structure_csv/mp23_pdd_clara_structure_info_costs_2024_06_18.csv \ --data_path data/ProcessedData \ --tables_dir scripts/rr2_tables \ --existing_geo \ --output_dir output \ --column_setup full \ --insurance 32 \ --occupancy_map data/OccupancyMapping/OccupancytoTypeofUseMapping.csv \ --parallel

Use the --insurance flag to restrict calculations to insured structures only. If a number is provided (e.g., --insurance 32), it controls the number of iterations for stochastic assignment. If used without a value (i.e., --insurance), the default is 10 iterations.

Add --parallel for faster geographic attribute extraction.

Add --existing_geo to use precomputed geospatial attributes (if any).

Example inference scripts to batch-run premium calculations across multiple simulations using predefined bash commands are stored in infer_scripts folder.

Customization via CLI

| Flag | Description | |------|-------------| | --years | List of years (2 12 22 32 42 52) | | --plans | Scenario plans ( FWOA FWMP) | | --scenarios | Scenarios (Higher Lower) | | --structure_csv | Path to structure CSV | | --data_path | Root directory for all processed GIS and raster data | | --tables_dir | Root directory for all FEMA rating factor tables | | --existing_geo | Path to existing geographicdf CSV file to skip extraction | | `--outputdir| Path to save the results | |--columnsetup| Choose betweenpremium(default) orfullcolumns | |--insurance| Flag to calculate premiums for insured buildings only, number of iterations for insurance flag assignment | |--occupancymap| Optional: Path to occupancy-to-TypeOfUse mapping CSV | |--parallel` | Optional: enable parallel processing of geographic attributes extraction |

Output Files

• MP2023_<scenario>_<plan>_C000_U00_V00_SLA_<year>_RR2_geo.csv: Geographic attributes for each structure
• MP2023_<scenario>_<plan>_C000_U00_V00_SLA_<year>_RR2_prem.csv: RR2 premium results

Output Columns

--column_setup premium

• structure_id
• Building Premium
• Contents Premium
• Increased Cost of Compliance Premium
• Mitigation Discount
• Community Rating Systems Discount
• Full-Risk Premium

--column_setup full (adds geographic info)

In addition to the above: - County, HUC12, CRS, LeveeSystemId, Elevation - StructRelElev, DTR, ElevRiver, ERR, DrainageArea, RiverClass, DTC

📁 Project Structure

├── data/ # Input structure CSVs, occupancy maps, processed data ├── output/ # Output folder for premium calculations ├── rr2_tables/ # FEMA RR2 tables (CSV format) ├── scripts/ # Main logic and functions │ ├── helper_functions.py │ ├── geographic_attributes_chunk.py #for parallel processing │ ├── geographic_attributes.py │ ├── rr2_premium_functions.py │ ├── insurance_utils.py │ ├── table_loader.py │ └── main.py ├── infer_scripts/ # Bash scripts to run example jobs ├── requirements.txt ├── README.md

Python Requirements

``` Python 3.10+

Install dependencies:

pip install -r requirements.txt ```

License