Basic Network Visualization and Routing (QGIS3)¶
Creating, visualizing, and managing networks is an important part of GIS. Many types of physical infrastructure such as roads, railways, utilities can be modeled as networks with lines and nodes - with properties attached to them. In this tutorial, we will learn how road networks are commonly modeled and apply some styling techniques to visualize the routing properties. We will also use QGIS3’s built-in tools for network analysis that to find shortest path between 2 points along the network.
Overview of the task¶
We will take a layer of road centerlines for Washington DC, visualize the connectivity and build a network to find shortest path between any 2 points in the city.
Other skills you will learn¶
How to use data defined overrides to align an arrow symbol based on line direction.
Get the data¶
District of Columbia government freely shares hundreds of datasets on the Open Data Catalog.
Download the Street Centerlines shared by DCGISopendata data as a shapefile.
For convenience, you may directly download a copy of the datasets from the links below:
Data Source: [DCOPENDATA]
Locate the downloaded
Street_Centerlines.zipfile in the Browser panel. Expand it and drag the
Street_Centerlines.shpfile to the canvas.
You will see a new line layer called
Street_Centerlinesadded to the Layers panel. This layer represents each road in Washington DC. Select the Identify tool in the Attributes Toolbar. Click on any road segment to see what attributes are attached to it. There are standard attributes like road name, type etc. there is an attribute called
DIRECTIONA. This is an import attribute for routing as it specifies whether the segment is two-way or one-way. It contains 4 different values.
Two Wayfor two-way streets.
One Way (Digitizing direction)for one-way streets where the traffic is allowed in the direction of the line (start-point to end-point) and
One way (Against digitizing direction)for one-way streets where the traffic flows in the opposite direction of the line. There is also
Unknownvalue where we will assume two-way traffic. We will now use the information in that attribute to display an arrow on streets that are one-way.
Click the Open the layer Styling Panel button in the Layers panel. Select the
Rule-basedrenderer from the drop-down menu.
We will create a new style with a filter for only the one-way roads. Click the Add rule + button.
In the Edit rule dialog, click the Expression button.
In the Expression string builder dialog, expand the Fields and Values section in the middle-panel. Select
DIRECTIONAattribute and click All Unique in the right-hand panel. The 4 values that we discussed earlier will appear. Having these values here as a reference helps when building the expression. Also, you can double-click on any value to add them to the expression.
The goal is to create an expression that selects all one-way streets. Enter the following expression and click OK.
"DIRECTIONA" = 'One way (Against digitizing direction)' OR "DIRECTIONA" ='One Way (Digitizing direction)'
Next, change the Symbol layer type to
on center pointunder Marker placement.
Click on the
Simple markersymbol. Scroll down and pick the
filled_arrowheadmarker. You will see that the arrow-like symbol now appears on the one-way streets. But all of them are pointing in a single direction, whereas we know that our filter contains roads in multiple directions. We can further refine the symbols with a data-defined override for the Rotation value.
Click the Data defined override button next to Rotation.
We can put a conditional expression that returns different rotation value depending on the one-way direction. A simple 0 or 180 degree rotation works to account for the direction, but it works only for horizontal lines. To align the arrow-head perpendicular to all the lines - we need to account for the angle of the line in the expression as well. The
angle_at_vertexfunction helps us find the angle and use it in the expression. Enter the following expression and click OK.
CASE WHEN "DIRECTIONA" = 'One Way (Digitizing direction)' THEN angle_at_vertex($geometry, 1) - 90 WHEN "DIRECTIONA" = 'One way (Against digitizing direction)' THEN angle_at_vertex($geometry, 1) - 90 + 180 END
Now you will see the arrow-heads aligned to the correct road direction and angle. To keep the style uncluttered, we are choosing to display arrows only on one-way streets. Unlabeled streets are assumed to be two-way. Now that we have the network styled correctly, we can do some analysis. Go to.
Search for and locate thealgorithm. Double-click to launch it.
In the Shortest Path (Point to Point) dialog, select
Street_Centerlinesas the Vector layer representing network. Keep the Path type to calculate as
Shortest. Next we need to pick a start and end point. You can click the … button and click on any point on the network in the canvas. If you want to replicate the results in this tutorial, you can enter
-76.99730092166396,38.887624846748984as the Start point and
-76.99154831062152,38.89151000569929as the End point. Expand the Advanced parameter section. Choose
DIRECTIONAas the Direction field. You must be familiar with the one-way direction values for forward and backward traffic flow. Enter
One Way (Digitizing direction)as the Value for forward direction and
One way (Against digitizing direction)as the Value for backward direction. Keep other options to their default values and click Run.
The algorithm will use the geometry of the layer and provided parameters to build a network graph. This graph is then used to find the shortest path between the start and end points. Once the algorithm finishes, you will see a new layer
Shortest pathadded to the Layers panel that shows the shortest path between start and end points.
You will see that there are many possible paths between start and end points. But given the constraints of the network - such as one-ways, the result is the shortest possible path. It is always a good idea to validate your analysis and assumptions. One easy way to validate it is to use a third-party mapping service to see if their results match with the ones we derived. Here is the shortest path suggested by Google Maps between the same start and end points. As you can see the recommended shortest route matches exactly with our results - validating our analysis.