Implementing a geodatabase for local government, Part 3

Designing geodatabase behavior

Recall from the previous two modules that a small city in southern California is in the process of developing a pilot project to migrate their GIS data to the geodatabase data model. One of the main concerns that the city of Rome department managers have is that new development has been occurring throughout the city. They would like to see the pilot project incorporate at least one new subdivision.

All the subdivisions are recorded in detailed CAD drawings. The city engineers feel that additional database layers for capital improvements should be based on these CAD drawings. For example, new development means that utility networks, such as water and sewer, will need to be extended. The additional loads on these systems will also require improvements to the existing components. In fact, the Public Works Department is planning a major overhaul of the city's sewer and stormwater systems—in a major capital improvement project, they will be separating the systems for environmental reasons.

Sue, the GIS consultant from OMH Inc., is planning to add the CAD data to the Public Works personal geodatabase, thus providing the engineers with a familiar baseline dataset. More importantly, designing the sewer and stormwater systems in the geodatabase will also give them an opportunity to establish relationships between new and existing components and build intelligence into their GIS database.

 

Portion of CAD drawing showing city streets and parcels

 

Detailed CAD drawings provide the city engineers and planners with surveyed information about parcel and street data. The CAD data is going to be migrated to the pilot personal geodatabases and used as a baseline for further data development.

 

In order to prepare for this project, the Public Works Department would first like to create rules and behaviors for how the sewer system can be assembled in the database. They would also like to make data entry easier and more reliable when the database gets modified.

The GIS staff at the Public Works Department have been thinking about improving the database for a while. They have been closely following the development of the ArcGIS® Sewer/Stormwater Data Model, which provides many of the standards they desire for feature class definition, rules for editing, and symbology.

As part of the project, the city engineers want to build a completely accurate and realistic inventory of system components. For example, the inventory would include valves, manholes, meters, pumps, and other connection points.

Using the ArcGIS Sewer/Stormwater Data Model as a starting point, Sue and the city engineers are able to develop a UML diagram that suits their needs. They decide that certain features, such as pumps, can be effectively represented with subtypes based on the type of pump.

 

Diagram of Pump feature class and six subtypes

 

When a pump feature is added to the geodatabase, one of six different subtypes is selected. Each subtype shares the attributes of the Pump feature class.

 

For the pilot project, the engineers also want to specify domains for the subtypes to make data entry easier.

 

Diagram showing an attribute domain for pump diameter

 

A range domain named PumpDiameter is created and then assigned to both the InletDiameter and DischargeDiameter fields. The PumpDiameter domain is also associated with each of the subtypes. When a new pump feature is added to the geodatabase, the diameter values entered for the InletDiameter and DischargeDiameter fields can be immediately validated.

 

Every feature class and nonspatial table is considered in the same manner. For the pilot project, Sue works with a team of representatives from each department to design the geodatabase. Wherever possible, they use templates such as the ArcGIS Sewer/Stormwater Facilities Model to facilitate the design.

Once the UML diagrams are complete, Sue can export them to an XML Metadata Interchange (XMI) file and then use the ArcCatalog CASE Tools Schema Wizard to generate geodatabases containing feature classes with defined behavior.

After the CAD datasets are loaded into these empty geodatabase structures, the department staff can edit features and tables, digitize new feature classes, and use the validation tools to enforce database integrity.