QGIS Plugin — ProxiSanté

QGIS plugin developed in Python that automatically finds the fastest route from a starting point (Lambert-93 coordinates) to the nearest hospitals. Developed in pair as part of a university project at the University of Montpellier.

Python PyQGIS Qt Designer Graphical Modeler Routing Network Analysis

📄 Full Report (PDF)

01 — Features

The plugin automates road network analysis to calculate optimal routes to hospital facilities from anywhere in the territory.

📍 Coordinate Input

Enter starting coordinates in Lambert-93 (X,Y). Automatic format validation and error handling.

🛣️ Network Analysis

Calculate fastest routes to all hospitals via road network, with travel time in minutes.

🏆 Automatic Ranking

Sort hospitals from fastest to slowest. Extract and highlight the optimal route in red.

🗺️ QGIS Visualization

Automatic loading of results in QGIS with differentiated styles: blue for all paths, red for the fastest.

02 — User Interface — Qt Designer

The interface was designed with Qt Designer and integrated into QGIS via PyQGIS. Signals and slots enable communication between widgets and business logic.

Final interface — coordinate input, scale and layer selection

Qt Designer — signals and slots editor (mScaleWidget, mMapLayerComboBox, lineEdit)

Code excerpt — retrieving parameters from the interface and validating coordinates:

# Retrieve selected layers from ComboBoxes
textMapLayer1 = self.dlg.mMapLayerComboBox.currentText()
textMapLayer2 = self.dlg.mMapLayerComboBox_2.currentText()

# Retrieve and validate starting coordinates
Coordonnees_depart_texte = self.dlg.lineEdit.displayText().strip()
point_depart = self.Obtenir_Coordonnees_depart(Coordonnees_depart_texte)

def Obtenir_Coordonnees_depart(self, Coordonnees_depart_texte):
    try:
        Coordonnees = Coordonnees_depart_texte.strip().split(',')
        if len(Coordonnees) != 2:
            raise ValueError("Invalid coordinate format")
        lat, lon = map(float, Coordonnees)
        return QgsPointXY(lat, lon)  # Create a QGIS geographic point
    except (ValueError, IndexError) as e:
        QMessageBox.warning(None, "Error",
            f"Invalid format. Use: X,Y (e.g., 711980,6242711). Error: {str(e)}")
        return None

03 — QGIS Graphical Modeler

Algorithm processing is encapsulated in a QGIS graphical model that orchestrates the different steps: spatial indexing, route calculation, travel time calculation in minutes, sorting and optimal route extraction.

QGIS Graphical Modeler — ProxiSanté algorithm

Fig. 1 — QGIS Graphical Model: complete processing pipeline from inputs (hospitals, road network, start) to outputs (sorted paths and optimal route)

① Spatial Indexing

Spatial index on hospitals and road network to optimize geographic queries.

② Route Calculation

Fastest route from starting point to each hospital in the network.

③ Time in Minutes

Conversion and calculation of travel time in minutes for each calculated route.

④ Sorting & Extraction

Sort in ascending order, selection and extraction of the fastest route to the optimal hospital.

03b — Python Model Export

The graphical model was exported to Python as a QgsProcessingAlgorithm class. This script is then imported and executed directly from the main plugin file, encapsulating the entire processing pipeline in a reusable QGIS algorithm.

Architecture — calling the model from the main plugin:

# Import the exported model from the graphical modeler
from .Temps_de_trajet_vers_hopitaux import Temps_de_trajet_vers_hopitaux

# Initialization and execution of the model with interface parameters
self.modele = Temps_de_trajet_vers_hopitaux()
self.modele.initAlgorithm()
results = self.modele.processAlgorithm(parameters, context, feedback)

Key excerpt — travel time calculation in minutes and hospital ranking:

# Travel time in minutes from path length (speed 80 km/h)
'FORMULA': 'round($length / (80 * 1000), 4) * 60'

# Ascending sort to find the fastest hospital
'EXPRESSION': 'to_real("temps_minutes")'
'ASCENDING': True

# Extraction of rank #1 path (fastest)
'FIELD': 'Classement', 'VALUE': 1

The complete code is available on the Codeberg repository.

04 — Results and Visualization

Results are automatically loaded into QGIS with differentiated symbology. Paths to all hospitals appear in blue, the optimal route to the nearest hospital is highlighted in red.

ProxiSanté result in QGIS — blue paths and optimal route in red

Fig. 2 — Result in QGIS: paths to all hospitals (blue) and fastest route (red). Copy coordinates in Lambert-93 or WGS84 via right-click.

Code excerpt — automatic loading and styling of results in QGIS:

def charger_resultats(self, pathResult1, pathResult2):

    # All paths to hospitals — thin blue line
    vlayer1 = QgsVectorLayer(pathResult1, "Paths to hospitals", "ogr")
    if vlayer1.isValid():
        symbol1 = QgsLineSymbol.createSimple({
            'line_color': '0,0,255,255',   # Blue
            'line_width': '1'
        })
        vlayer1.renderer().setSymbol(symbol1)
        QgsProject.instance().addMapLayer(vlayer1)

    # Optimal route — thick red line
    vlayer2 = QgsVectorLayer(pathResult2, "Nearest hospital", "ogr")
    if vlayer2.isValid():
        symbol2 = QgsLineSymbol.createSimple({
            'line_color': '255,0,0,255',   # Red
            'line_width': '2'
        })
        vlayer2.renderer().setSymbol(symbol2)
        QgsProject.instance().addMapLayer(vlayer2)

05 — Installation — QGIS Plugin

The plugin can be installed directly from the QGIS plugin manager and appears in the Extensions menu and toolbar.

ProxiSanté Plugin in QGIS plugin manager

Fig. 3 — ProxiSanté Plugin installed in QGIS — plugin manager with description, category and links

View repository View Python script