osmsatlab 1.0.1

A library for calculating accessibility and equity metrics using OpenStreetMap data.

OSMSatLab

OSMSatLab is a Python library for comprehensive urban accessibility and equity analysis. It seamlessly integrates OpenStreetMap (OSM) service locations with WorldPop raster population data to calculate high-resolution coverage metrics.

Designed for ease of use, it automates data retrieval, processing, and visualization through a single unified interface.

Key Features

• Automated Data Retrieval: Automatically identifies countries, downloads WorldPop geotiffs, and fetches cached OSM data for any given area.
• Multi-Modal Analysis: Calculates distances using Euclidean, Driving (approximate), and Walking network metrics.
• Coverage & Equity: Computes:
  • Coverage Ratio: Percent of population within a specific service threshold (e.g., 1km).
  • Service Density: Services per 1,000 inhabitants.
  • Service Burden: Population per service.
• Integrated Visualization: Built-in plotting tools for population choropleths, service density maps, and distance histograms.
• Efficient Architecture: Uses spatial indexing (KD-trees) and raster-vector hybridization for high-performance analysis on large regions.

Installation

You can install osmsatlab using pip or Poetry.

Using Poetry (Recommended)

poetry add osmsatlab

Using pip

pip install osmsatlab

---

Quick Start

The core of the library is the OSMSatLab facade class. It handles initialization and data loading automatically.

1. Basic Initialization

Initialize with a bounding box (minx, miny, maxx, maxy). This automatically downloads population data and OSM services.

from osmsatlab.core import OSMSatLab

# Bounding box for Amsterdam
bbox = (4.85, 52.33, 4.95, 52.40) 

# Initialize lab (downloads data automatically)
lab = OSMSatLab(bbox=bbox, crs="EPSG:3857")

2. Calculate Metrics

Compute accessibility metrics for a specific service category (e.g., 'healthcare', 'education_school').

# Calculate Accessibility (Euclidean, 1km threshold)
access_metrics = lab.calculate_accessibility_metrics('healthcare', threshold=1000, metric_type='euclidean')
print(f"Coverage Ratio: {access_metrics['coverage_stats']['coverage_ratio']:.2%}")

# Calculate Per-Capita Equity
equity_metrics = lab.calculate_per_capita_metrics('healthcare')
print(f"Services per 1k people: {equity_metrics['services_per_1000']:.4f}")

3. Visualization

Use the built-in visualization tools to generate maps instantly.

from osmsatlab.viz import render_maps
import matplotlib.pyplot as plt

# Generate maps for healthcare access
render_maps(lab, "Amsterdam", service_category='healthcare')
plt.show()

---

Project Structure

The library is organized into modular components:

• osmsatlab.core: Contains the OSMSatLab facade class, the main entry point.
• osmsatlab.io: Data fetching modules for OSM (osm.py) and WorldPop (population.py).
• osmsatlab.metrics: Calculation logic for accessibility and per_capita metrics.
• osmsatlab.spatial: Spatial operations and index (KDTree) management.
• osmsatlab.viz: Visualization workflows, including render_maps and choropleth utilities.

Methodology

OSMSatLab uses a raster-based population approach combined with vector-based service locations.

1. Population: WorldPop 100m/1km resolution rasters are masked to the Area of Interest (AOI).
2. Services: OSM Points of Interest (POIs) are fetched and spatially indexed.
3. Analysis: For every population pixel centroid, the nearest service distance is calculated. The population count of pixels falling within the distance threshold is aggregated to determine total coverage.

License

This project is licensed under the MIT License - see the LICENSE file for details.