ArcGIS Tutorial: Understanding Connectivity

When you create a network dataset, you select which edges or junction elements will be created from the source features. Ensuring that the edges and junctions are formed correctly is important for obtaining accurate network analysis results.

Connectivity in a network dataset is based on the geometric overlap of line endpoints, line vertices, and points and follows the connectivity rules set to the properties of the network dataset.

　Connectivity Groups

Establish connectivity in the ArcGIS Network analyst extension to start with defining the connectivity group. Each edge source can be assigned to only one connectivity group, and each junction source can be assigned to one or more connectivity groups. A connectivity group can contain any number of sources. The way a network element is connected depends on the connectivity group where the element resides. For example, if you create two edges from two different source feature classes, you can connect if they are in the same connectivity group. If you are in a different connectivity group, these two edges are not connected unless you connect the two edges with a junction that participates in both connectivity groups at the same time.

Connectivity groups can be used to build multimodal transport system models. You can select the network sources that you want to connect to each other for each connectivity group. In the subway and street Multimodal network examples below, metro lines and metro entrances are all assigned to the same connectivity group. Note that the metro_entrance is also in the connectivity group where the streets are located. It constitutes a connection between two connectivity groups. All paths in both groups must pass at least one shared Metro entry. For example, the path solver might determine the best path for pedestrians between two locations in the city: walk from the street to the subway entrance, then take the subway, transfer to another subway at the transfer station, and finally get out of the other subway entrance. The connectivity group distinguishes between two networks and connects them through a shared junction (Metro entrance).

　　

　　Connecting edges inside a connectivity group

Edges within the same connectivity group can be connected in two different ways, depending on the connectivity policy used on the edge source.

• If you set up an endpoint connectivity rule, the line features become the edges that are connected only at the coincident end point.

　　

In this example, the line feature L1 becomes the edge feature E1, and the line feature L2 becomes the edge feature E2. Following this connectivity policy, one edge feature will always be created for a line feature. Building a network with endpoint connectivity is a way to build a cross-object model, such as a bridge. To build the model for this example, the bridge and the street are placed in the same connectivity group (1). A street source is assigned an "arbitrary vertex" connectivity rule so that street features can be connected at coincident vertices with other street features. The bridge source is designated as the endpoint connectivity. This means that the bridge can only be connected to other edge features at the end point. Therefore, any street that crosses under the bridge is not connected to the bridge. The bridge will connect to the other streets at the end point.

　　

If there is only one source in the network that you want to use to build the overpass (bridge) and underground channel (tunnel) model, you might consider using elevation fields on the flat data. For more information, see the "Elevation Fields" section later in this article.

• If the any vertex connectivity rule is set, the line features are divided into multiple edges at coincident vertices. If the purpose of building street data is to have streets intersect other streets at vertices, it is important to set this policy.

　　

In this example, the two polylines that intersect at the shared vertex position are divided into four edges, and a junction is formed at the shared vertex. Edge E1 and E3 are identified by the Source feature class and object ID of the line feature L1. Edge E2 and E4 are identified by the Source feature class and object ID of the line feature L2. Junction J3 will be the new system junction. Junctions J1, J2, J4, and J5 are either system junctions or junctions of coincident points of the source feature class.

　　Warning:

Not all cross-line features can generate connected edges. If they do not share any coincident endpoints or vertices, the junctions cannot be created at the intersection point through any connectivity policies. Therefore, the street data of the network dataset must be cleaned up beforehand to ensure that both the vertex and the endpoint appear at all required junctions.

　　

If you need to improve street data, you can use geoprocessing tools such as consolidation, split intersections, or build topologies on those feature classes and apply topology rules that force features to split at intersections when editing street features.

　　Connecting edges through junctions between connectivity groups

Edges in different connectivity groups can only be connected through junctions shared by two connectivity groups.

In a multimodal system example that combines a bus and street network, a bus station is added from a point source and is in two connectivity groups at the same time. Then, the point location of the bus station must be spatially coincident with the connected bus line and street line. When you add a point location for a bus station, whether the point can become a junction depends on the junction connectivity policy. For edges, junctions and edges are either connected at the end point or at vertices, depending on the connectivity policy of the target edge source. However, in some cases you may want to override this behavior.

　　

　　

For example, a bus station is connected to a bus line that uses an endpoint connectivity strategy, but you typically want to place a bus stop at an intermediate vertex. To achieve this, you will need to set a junction policy to override the default behavior of connecting junctions to a given edge.

To override the default behavior of junctions so that junctions are no longer formed at endpoints or vertices based on the edge source connectivity policy, set the connectivity of the junction source to overwrite. By default, the edge connectivity policy is respected.

　　

　　

　　Building an elevation model

The connectivity of network elements depends not only on whether they overlap in x and y space, but also on whether they share the same elevation. There are two alternative ways to build an elevation model: Use elevation fields and z-coordinate values that use geometry.

　　Elevation fields

In a network dataset, elevation fields are used to optimize connectivity at the end of a line. They contain elevation information obtained from the fields of the feature class participating in the network. This differs from establishing connectivity based on z-coordinate values, where the physical elevation information is stored in each vertex of the feature. Elevation fields apply to edge and junction sources. The Edge feature source using the Elevation field describes the elevation with two fields (one for each endpoint of the line feature).

In the following example, the EF1, EF2, EF3, and EF4 four line features belong to the same connectivity group and adhere to the endpoint connectivity rules. The elevation value of EF3 and EF4 is 0; The elevation value for EF1 and EF2 is 1. Therefore, at the intersection point, the EF3 only connects to EF4 (without connecting EF1 or EF2). Similarly, the EF1 connects only EF2, not EF3 or EF4. It is important to note that elevation fields are used to optimize connectivity, and they do not overwrite connectivity. Two edge elements, even if they have the same elevation fields, overlap each other, but if they are in two different connectivity groups, they are still not connected to one another.

　　

Many data vendors provide elevation field data that is used to build connectivity models. The ArcGIS Network DataSet Connectivity Model can use this elevation field data to enhance connectivity. Also, the interaction between elevation fields and connectivity models is also critical in building special scenarios such as bridges and tunnels.

　　Z-coordinate value of the geometry

If the z-values are stored in the geometry of the source features, you can create a three-dimensional network.

The model of the interior sidewalk is usually built with a 3D network. Consider that many corridors in a multi-storey building are indistinguishable in 2D (x, y) space, but in 3D space, they can be differentiated by their z-coordinate values. Similarly, elevators are moved vertically to connect the floors. Elevators are points in the X, Y space, but they can be modeled as lines in 3D space.

Z-coordinate values make it possible to construct a connectivity model of points and line features in three dimensions. In a 3D network dataset, to establish connectivity, the source features (specifically point, line endpoints, and line vertices) must share all three coordinate values: x, y, and Z values.

ArcGIS Tutorial: Understanding Connectivity