🗺️ QGIS Network Analysis: Shortest Path Finder for Philippine Roads
This project demonstrates how to perform network analysis using QGIS and Python. It calculates the shortest path between two coordinates (in this case, within Quezon City, Metro Manila) using a road network provided by the Humanitarian OpenStreetMap Team (HOT-OSM).
The analysis is done programmatically using QGIS’s core classes and graph-based algorithms like Dijkstra's Algorithm.
📌 Features
Load vector road data from a GeoPackage (.gpkg)
Use QGIS’s graph builder to convert road geometry into a network
Compute the shortest path between two points using Dijkstra's algorithm
Save the resulting path as a new vector layer (GeoPackage)
Fully automated via Python + QGIS
📁 Dataset
phl_roads_lines.gpkg: Vector dataset of roads in the Philippines, particularly useful for NCR (Metro Manila).
Source: Humanitarian OpenStreetMap Team
🧠 Requirements
QGIS (>= 3.x) installed on your system
Python (3.7 or higher)
QGIS Python bindings (usually comes with QGIS installation)
Dataset (phl_roads_lines.gpkg) in the project directory
⚙️ Setup and Execution
Install QGIS
sudo apt install qgis python3-qgis
Ensure the qgis.core, qgis.analysis, and PyQt5 modules are available.
Run the Script
python3 shortest_path.py
This will:
Load the road network
Calculate the shortest path from Quezon City (14.6760, 121.0365) to a destination point (14.5550, 121.0000)
Save the path in shortest_path.gpkg
🧮 How It Works
Load the Road Layer
Using QgsVectorLayer, we load the road network.
Define Points
Define start_point and end_point using QgsPointXY.
Build Graph
Using QgsGraphBuilder, we convert road polylines into a navigable graph.
Shortest Path Calculation
Apply QgsGraphAnalyzer.dijkstra() to compute the least-cost route.
Export Path
Write the result as a LineString into a new .gpkg file with proper attribute fields.
🧪 Output
✅ shortest_path.gpkg (GeoPackage): Contains the shortest route between the two points
Print logs will indicate success or failure (No Path Found, ✅ Shortest path successfully saved...)
