Ways to edit network datasets

Once a network dataset is built, it can be edited. The following list provides some examples of ways you can edit the network dataset:

• Sources can be added or removed from a network dataset, and connectivity can be modified.
• If the network supports turns, new turn sources can be added or existing ones removed.
• Attributes can be added, removed, or redefined.
• Hierarchy can be modified.
• Direction information, set up when the network was built, can also be changed.

Changes to the schema of the network datset are made in the Network Dataset Properties dialog box.

In addition to the changes in the schema of the network, the basic network can be edited by editing its sources. If a new road has been added, creating a new feature in the streets source and rebuilding the network will ensure that the network dataset is updated. If the speed limit field value on certain roads has changed and that affects the travel_time network attribute, rebuilding the network will ensure that the correct values for travel_time are stored for the roads in question.

Note:

It isn't possible to rename the source feature classes of a network dataset. If renaming them is required, you need to delete the network dataset, rename the feature classes, and re-create the network dataset.

Editing source features

Since network elements are created from features of network sources, all edits must be made to the sources.

• Features in edge and junction sources are edited in the same manner as features in any other feature class in ArcMap. Once the sources are edited, you will need to ensure that all new features have shared vertices with intersecting features. This can be achieved by using the Integrate geoprocessing tool or by using topology.
• Turn sources can be edited by editing turn features or turn source fields.

Learn more about editing features in ArcMapLearn more about the Integrate toolLearn more about topology in ArcGISLearn more about editing turn featuresLearn more about editing turn feature class fields

Once the source features are edited, the network should be built to incorporate the changes. ArcGIS uses dirty areas to track the parts of the network dataset that need to be built due to edits. Dirty areas serve various purposes, including the following:

• Dirty areas minimize the time it takes to build a network. Rather than always building the entire network, the build operation only builds portions of the network that are covered by dirty areas.
• Dirty areas serve as visual cues to let you know if and where a network needs to be built.

You can view any dirty areas using ArcMap or ArcCatalog.

• In ArcMap, open the Layer Properties window of the network dataset layer, click the Symbology tab, and check the Dirty Areas check box.
• In ArcCatalog, click the network dataset in the Catalog Tree window and click the Preview tab.

Dirty areas appear as rectangles and, by default, have purple outlines with purple hatching inside.

Editing source features in an ArcSDE geodatabase

Data must be registered as versioned if you plan to edit the source features of a network dataset that is stored in an ArcSDE geodatabase. As with any source-feature edits you make, they will eventually need to be built into the network dataset. In a multiuser environment, dirty areas function as part of the mechanism for resolving conflicting edits and make it possible to version network dataset data.

If you are familiar with dirty areas in versioned topologies, the concept of dirty areas in versioned network datasets is similar.

Learn more about dirty areas and versioning of network datasets

Legacy:

Prior to ArcGIS 10, network datasets didn't support dirty areas or versioning. For this reason, you could edit source features with or without registering the data as versioned. With the release of ArcGIS 10, however, this changed, and now source features must be registered as versioned before they can be edited.

Copying network datasets

In ArcCatalog or the Catalog window of ArcMap, you can copy and paste a geodatabase-based network dataset from a feature dataset into another feature dataset with the same spatial reference. All feature classes that participate in the network dataset are also copied over. The new network dataset must be built before it can be used.