You are here

Gas Customer Analysis: A Business Solution

This month I wanted to continue looking at some unique customer solutions that have provided great value within a utility. Before I dig into this month’s specific topic, I wanted to briefly review why certain GIS-based solutions make such a large operational difference and provide the greatest return on investment.

If you’ve ever received a presentation from an Esri sales representative you are likely to be familiar with the concept of the five patterns of GIS, as detailed in this picture:

Gas Customer Analysis Five Patterns

This is a great talking point with management because it shows how GIS can impact the entire organization, far exceeding the GIS department. Staying in line with this methodology, we always strive to focus on finding innovative solutions to operational and workflow-related challenges instead of strictly focusing on the underlying technology. Various pieces of product software, customizations, and systems integration may be blended into the solution, but the end game is all about finding a solution that bridges the five patterns to maximize value for the customer.

This month’s topic is a great example of a business-oriented solution because it began with a business challenge. In this example, a utility had a process that was taking an exceptionally long time to determine a list of customers associated with a gas outage. They had their gas data within an Esri and Telvent-based GIS.

When an outage (planned or emergency) occurred, they would use the maps to visually determine the isolation area of the outage within the gas network. They would then correspond the gas outage area to the land base to gather the street names and address ranges. This information was then manually cross-referenced with their customer information system (CIS) to get a list of the customers at those locations. The list of customer and meter data was then formatted into spreadsheets to create a master outage list as well as gas relight forms to be used in the field to validate that the gas was turned off and then back on along with any corresponding pilot lights.

This process worked but it could take up to four or more hours to put all of this information together. The business challenge that was presented was whether a better solution could be put in place.

Around the same point in time, a separate initiative had been completed to geocode all of the gas customers from CIS into the GIS. This placed a point feature on the map based on a match against an address within either the utility-managed land base or more commonly against the Esri TeleAtlas web service. The points were non-networked and mostly represented a general location for the customer as opposed to a precise meter location. But it was a major leap forward. The solution centered on a tool we called Gas Customer Analysis because it analyzed gas customers by consolidating data from multiple different systems and rendering it quickly into a usable format.

The solution began with better utilizing the GIS data. Instead of using the map purely as a visual indicator of the extent of an outage, the first goal was to use the gas data as the basis for identifying the customers. The user can select a set of gas mains within the GIS using any available technique. 

At the simplest level, a manual selection using the Esri select features tool works well for small areas. But the power really comes in when you start using the Telvent gas valve isolation trace. This is a network-based trace that can originate from any linear point on the gas network. It will trace out and identify the full extent of the outage all the way to the gas valves that must be closed to isolate the area. You can then further exclude specific valves for any reason and find the next best isolation points. And as you may have guessed, this trace can return a selected set of gas mains.

This image shows an example of a set of selected gas mains, the blueGas customer Blue Point point features represent the geocoded customers: 

Because the customers are not connected into the geometric network, we needed to find some way to get to them via their proximity to the selected gas mains. Because the selection set could be in any irregular shape it was determined that a buffer would provide the most value. But we wanted to interactively provide the users with the ability to manipulate the buffer and thus began the creation of a customization within the GIS:


Gas customer analysis form

The top section of this custom form first informs the user about the pressure ranges of the selected gas mains. The next section then allows them to easily and interactively create a buffer around the selected gas mains:


Gas customers buffer slider

By simply dragging the buffer slider and clicking the "Draw Buffer" button, the user can create a buffer that provides the best coverage for geocoded customers on the map.

The initial setting of 100 ft may not provide enough overlap with the customers:


Gas customers buffer results 1

But the user can easily manipulate the buffer for maximum coverage:


Gas customers buffer results 2

By then clicking the "Get Customers From Buffer" button, the tool automatically reads in the basic customer information from all of the customer points contained within the buffer:


Gas customers get customers from buffer

This step alone shrunk the time to get the initial list of customers down into just a few minutes. The user can then click on any customer entry to see it flashed on the map. If certain customers should be excluded from the list, they can be selected and removed from the list. And just as easily, the user can click the "Add Customers" button to easily draw an extent on the map to get additional customers that might have been just outside of the buffer.

The next challenge was to determine if there were any customers contained within the CIS that might not be on the map. For safety reasons it is very important to ensure that all customers in the area are included in the outage list. This was accomplished by creating a systems integration point from the custom GIS tool into the CIS to check for any missing customers. 

But instead of just doing random searches, the tool assists the user by recommending address ranges to search within the CIS based on the current list of customers. Clicking the "Search CIS" button will prompt the user with a recommended search dialog:


Gas customer CIS serach

The tool takes the address range of the current customers and then analyzes them to look for odd or even street number trends. This approach quickly presents the user with a qualified list to search against the CIS database, which in this case was maintained within a mainframe DB2 system. Clicking "Search" searches against all customer records in the CIS using the specified criteria. Any matched customers are then compared against the current list of customers within the tool. Finally, any missing CIS records are added to the outage list within the custom tool, providing the user with a solid list of customers for the outage. 

The final challenge of the solution was to allow the user to export the customer and meter data into both a master outage list and a number of gas relight forms that could be split among a team of servicemen that will be working the outage in the field. The user accomplishes this with the click of the final two buttons:


Gas customers analysis export button

 The same process that previously took up to four or more hours has been reduced to minutes, and provides the user with a much more reliable and consistent pattern for gathering and reporting this data.

 We like to use this as a good example of a business-oriented solution because it started and ended with a business challenge. The solution used the core Esri and Telvent product software, extended the products with a custom tool to simplify and empower the user experience, and added a seamless systems integration point to the CIS to ensure the integrity of the customer data. These technical components provide a holistic business solution that can be tied back to each of the original five patterns of GIS discussed at the beginning of the article.

We hope this article will inspire you to think outside the box about your current business challenges. Avoid focusing solely on implementing a specific technology or piece of software, but approach the problem by looking for the best blend of technology to maximize the value to the organization.