Making Sense of Meter Data—Plunging Into Business Analytics

By Jill Feblowitz, IDC Energy Insights, with H. Christine Richards, freelance writer

Although a few systemwide implementation plans have been stalled, already a significant number of smart meters have been installed in the United States with plans for more.

As of spring, IDC Energy Insights estimates more than 21 million meters are installed at the top 80 U.S. utilities. Thirty-one of these utilities have plans for full deployment, which will bring the number of smart meters to more than 50 million during the next few years. IDC Energy Insights forecasts that by 2015, North American smart meter deployments will reach 88 million—a penetration rate of 51.4 percent of the addressable market, which includes residential homes and commercial and industrial businesses.

Smart meters are a rich source of data for utilities. Time series interval data—probably the most important data collected through AMI—can be collected over as little as 1-minute intervals and delivered multiple times per day. Other data associated with the smart meter is of great use to utilities, as well, such as time-stamped meter status and identification data. For smart meters used as a gateway for home-area networks, such as in Texas, data associated with appliances and thermostats can be made available, albeit with privacy protections.

At this stage, meter data is being collected by utilities for presentation to customers and creation of utility bills. The meter data is in this sense supporting transactions. Much more can be gained from analyzing that data, however. Enter business analytics.

End users throughout utilities could benefit from basic and more advanced analytics applications. Business value and needs drive the type of applications companies will deploy. Three key business value segments include customer value, operations and maintenance and capital investment.

Near-term Customer Value: Better Customer Service

Utilities can leverage more frequent consumption information from smart meters to help customers sort through billing questions and provide more services to customers. For example, analyzing energy usage data, utilities can help customers pinpoint the causes of high energy bills. Then utilities can recommend action, such as servicing an air conditioner or replacing an inefficient refrigerator. In some cases, analyzing meter data alongside account and billing history can help utilities identify potential fraud.

Through better analyzing customer payment patterns, utilities also can provide appropriate assistance for customers based on their payment histories. Utilities can use analytics to determine if a late-paying customer is chronically late and might require additional utility support. Layer on analytics and home energy management systems, and utilities can consult customers who have bill troubles about lowering costs through energy conservation.

Using analytics to better understand customer preferences, attitudes and behaviors, utilities can identify and better satisfy customers’ needs. Utilities offer many programs—energy efficiency, solar incentives, time-based pricing and demand response—and more programs are coming onboard every day. Every customer is not the same, so segmenting the market and implementing appropriate marketing plans for each segment will help stretch marketing dollars and garner greater participation.

Long-term Customer Value: Customer Empowerment

More advanced analytics will enable customers to plug in and manage energy devices rolling into their homes and businesses, such as electric vehicles, renewable and distributed energy resources, smart appliances and home energy management systems. Some of these devices will enable energy consumers to become power producers, as well. More advanced business analytics can offer opportunities for consumers to understand the power-trading marketplace and efficiently sell their excess power.

To support customer decisions about energy production and consumption, utilities can leverage more advanced analytics and better access real-time data to offer time-based pricing plans. This means customers can select pricing plans that better serve their needs based on their energy lifestyles. Customer smart meter data, customer segment information, weather information and pricing data can enable utilities to refine pricing in real time based on customer responses to price signals.

Near-term Operations, Maintenance: Pinpoint Outages, Refine Asset Management

Utilities are investigating the value of having smart meters on the grid from an operational perspective. The secret of outage management is that it largely has depended on incoming outage calls to locate potential outage fault location. With interval data and the ability to aggregate data based on location, there is more information to pinpoint faults so workers can go more quickly and directly to the source of problems.

One utility recently proved the value of having smart meter data available to prevent outages. The company is looking at consumption data from smart meters associated with individual transformers during peak consumption. Taking this data together with weather and expected increases in nonweather-related load for these meters, the utility can predict which transformers might be closer to the end of their lives and at risk of blowouts. The utility then can decide to deploy personnel to monitor the equipment more closely or deploy sensing devices to monitor transformer health.

Long-term Operations, Maintenance: Supporting Demand Response

Utilities can understand how to operate demand response systems better. Interval data from smart meters, demand response program information systems and geospatial repositories can provide utilities with the connectivity needed to support the dispatch of demand response as a variable resource, much like how utility-owned—and soon more customer-owned—wind and solar are incorporated onto the grid.

Near-term Capital Investment: Assess Impacts, Optimize Portfolios

It’s important to understand how changes in energy demand impact the grid. Demographic information, geocoding, revenue meter data, production meter data and geospatial data can help utilities better forecast demand growth and capital investment requirements for the lowest levels of the distribution grid. Utilities can determine through analytics whether new or upgraded infrastructure is needed in a particular area to meet the area’s changes in energy demand. The utility in the example also uses transformer-level analysis of revenue smart meters to plan replacement of transformers before they fail.

New customer components coming online also can impact utility capital investment needs. These components could include electric vehicles (EVs), energy storage devices and small-scale power generation sources. Plugging EVs onto a single feeder, for example, can stress an undersized transformer and risk outages for a neighborhood or a larger portion of the grid. Install distributed generation, and it introduces bidirectional power flow on a grid built for one-way power flow. Utilities can use analytics to begin to understand which areas will be most likely to gain these new components and to better plan for capital investments in those areas.

Long-term Capital Investment: More Information for Forecasting

Once more new components of the grid deploy, utilities can use new data from these components to optimize capital-investment decisions. As the intelligent grid rolls out, for example, utilities no longer will have to assume about customers’ energy needs. They will have access to consumption patterns through smart meters and home energy management systems. With more customer data that can feed into models, utilities can make fewer assumptions to reach capital investment decisions.

In the supply and portfolio optimization arena, utilities will need to find ways to incorporate distributed generation better and more renewable generation resources. Many moving parts in the generation portfolio will require analytics to process the increasingly complex and dispersed data sources. As carbon pressure builds, utilities will seek ways to defer carbon-emitting generation investments through better peak–demand management, renewable generation integration and supporting the electrification of transportation.

General Guidance

Given the rapidly changing marketplace, utilities must ensure their first forays into business analytics turn into long but successful adventures. Factors for ensuring the long-term success of business analytics include:

  • Executive sponsorship. Executive support of companywide analytics efforts is critical to ensuring analytics receive the funding and support they need to be successful.
  • Stakeholder collaboration. The increasingly cross-cutting nature of analytics means many groups are involved in the success of the analytics, including business folks from all parts of the energy value chain and information technology personnel.
  • Repeatable analytics. Out-of-the-box solutions can help maintain consistency, but likely some level of customization for utility analytics will remain. When customizing analytics, utilities should ensure the applications and business processes associated with these analytics are repeatable. Flexibility in analytics applications will ensure current analytics programs can adapt to tomorrow’s needs.

Whether dipping their toes or plunging into business analytics, utilities can benefit from basic dashboards to advanced optimization analytics.

Jill Feblowitz is vice president, utilities and oil and gas for IDC Energy Insights. For more than 25 years she has worked as a consultant and in the field. At IDC Energy Insights, she manages analysts who provide research-based advisory and consulting services. She contributes to the utilities and oil & gas blogs in the IDC Energy Insights Community ( Her Twitter handle is jill@idc.

This article was written with assistance by freelance journalist H. Christine Richards.


Designing the Flow of Smart Grid Data

By Guerry Waters, Oracle Utilities

Meter data management (MDM) was the first application specifically designed for the smart grid era. Initially envisioned as the first stop in smart meter data’s journey through utility business processes, MDM consolidated data validation, editing and estimation (VEE) with bill determinant production—offloading these tasks from other meter-to-cash software.

This initial focus on smart meter data’s meter-to-cash role was appropriate; without revenue, utilities do not function. Aside from customer and employee safety, revenue processes are the highest utility priority to which data can contribute.

Meter-to-cash, however, is not the only utility priority benefitting from smart meter data. Power restoration, outage prevention, asset monitoring, conservation programs and load control are a few essential utility activities this data can enhance.

But if MDM is the data entry point, must nonrevenue priorities wait for the data until MDM completes its billing tasks?

Waiting cannot be an option. For real-time operations such as load control, delayed data is useless data. And neither utilities nor their customers see power restoration as less important than sending out bills.

There is an answer to competition among applications for the same data: a gateway application that gathers and disseminates data and commands.

Gateway-based data flow avoids the bottleneck potential of sending all meter data to the MDM. Instead, each data-using application needs just two communication channels between itself and the gateway:

  • An input channel to receive all applicable data from every device, and
  • An output channel to handle all outgoing messages.

Similarly, each smart grid device needs just one input and one output channel to communicate with all applications.

A gateway provides a straightforward data flow, making it easy to create efficient business and operational processes that span the enterprise. Managers can design and change processes without the time-consuming need to redesign the flow of data and commands between applications and hardware.

A data-gathering and command-delivery gateway offers tangible benefits:

  • It protects utility investments. A switch to a new MDM requires only one new integration between the application and gateway. There is no need to create new paths between the MDM and metering headends or between the MDM and other data-needing applications.
  • It facilitates new uses for smart grid data without negatively affecting other data uses. Simple application configuration permits a gateway to provide data to multiple applications simultaneously.
  • It accelerates application development, configuration and use by providing standard templates or hooks for attaching any application to any device. In addition, a common set of commands that can be used from any application eases staff training.

The proliferation of smart meters, sensors and other devices is increasing utility data volume to astounding proportions. Preventing data bottlenecks is an escalating priority. A gateway prevents those bottlenecks and clarifies and speeds data flow, which helps utilities extract maximum value from smart grid data.

Guerry Waters is vice president of industry strategy at Oracle Utilities.

More PowerGrid International Issue Articles
PowerGrid International Articles Archives
View Power Generation Articles on
Previous articleMeasuring Smart Distribution
Next articleDenmark Case Study: Improving Grid Efficiency

No posts to display