ArcUser Online

Water Main Shutdown Application Prevents Costly Mistakes
By Jennifer Coate, Project Manager for Weston Solutions, Inc.

Most water utilities rely heavily on visual review of paper maps or digital map displays to identify which valves to shut off in order to isolate a main break and determine which services will be affected by a shutdown. This approach requires senior personnel and can be both time consuming and error prone.

A customized GUI and toolbar were designed for the Main Break application.

The two major risks associated with using this method include shutting off water to critical services, such as hospitals or facilities serving kidney dialysis patients, and unintentionally shutting off service to a large area. The latter scenario can occur when a valve controls water flow to a large dead-end portion of a water distribution network. Usually, these situations are tough to identify. Water utilities generally rely on the best judgment of senior maintenance supervisors who know system details through many years of experience.

Using a GIS water main shutdown analysis application can reduce these risks by allowing field crews and management to accurately model and display water main shutdown scenarios prior to taking action. This type of application can also benefit utilities by

• Identifying the customers to be notified
• Identifying critical services that cannot be interrupted
• Determining the extent of a service interruption to avoid unintentional shutdowns
• Identifying broken valves and "next choice" options for shutting down mains
• Capturing leak history data that is useful in pipeline replacement modeling

Case Study—San Jose Water Company

San Jose Water Company (SJWC) is a privately owned water utility that services the City of San Jose, California, and portions of several surrounding communities. After reviewing an application built by Esri Business Partner Weston Solutions, Inc., for Seattle Public Utilities using Esri's Workstation ArcInfo, SJWC determined that a field-based version of the shutdown application would help SJWC maintenance crews analyze main breaks. Weston built the field application in ArcView 8.1 using Visual Basic and ArcObjects running against a personal geodatabase. The resulting application provides field crews the ability to

• Graphically select water mains, services, and hydrants and perform a trace through the water network to locate and highlight valves that need to be closed to shut off selected features.
• Simulate the effects of closing one or more valves to determine what portions of the water network would be shut down.
• Flag critical services and supply lines to determine which users should be notified in the event of a shutdown.
• Generate a notification report listing all customers affected by a water main shutdown.
• Capture leak history of the system for pipeline replacement modeling.

At first glance, no large shutdown risks are visible. Click on the image to zoom the display and see that a large portion of network was accidentally shut down.

The application was installed at SJWC's main office and on seven laptops used by field crew leaders. It has been in production since mid-year 2001. SJWC dubbed the application Main Break.

Building Main Break

Building the SJWC network was no small feat. It is composed of 18 distinct feature layers that contain more than 40,000 distribution mains, 258,000 service lines, 288,000 valves, and assorted other equipment such as hydrants, zone valves, and transmission mains. In addition to water data, parcel centroids were used for address matching and owner/address information. SJWC's GIS data is maintained in Intergraph MGE format and was migrated into a personal geodatabase for use in the application. Once loaded into a geodatabase, a geometric network was created.

Data preparation is essential when building this type of application. Tracing requires that all features in the network be connected. If parts of the network are disconnected, due to editing errors or incomplete data, a trace will return a large number of disconnected features and the analysis will be unusable. Building this application provided unforeseen benefits for SJWC because data errors were discovered and corrected during the preparation process.

Although beneficial, the data preparation process was time consuming. All features and the geometric network had to be created and re-created multiple times to determine proper snapping distances and identify complex edge features.

Continued on page 2