Leveraging Information Architecture for Data Mining and Analytics for Greater Grid Investment Value

Written by Brad Williams

Smart grid deployments are creating exponentially more data for utilities and giving them access to information they've never had before. We need to know how to warehouse data offline from the operational database, organize data for efficient data analysis and reporting, provide access to operational data on a broad scale at minimal cost and isolate other business users from critical operational systems.

Utilities are beginning to find greater value from smart grid investments through data mining and analytics tools. But amassing huge amounts of smart grid data, transforming it into knowledge and communicating it to stakeholders in a timely manner can be a daunting task. Raw data must be captured and transformed into meaningful information that can be understood and used in utility business operations. The business users need to then process the information in the context of their existing knowledge, to build a broader base from which to make decisions about tactical corrections and strategic modifications.

Somewhere in this data deluge lies the path to a more efficient tomorrow, but how will access to this new data change the way utilities drive their businesses, and will predictive analytics spur operational change?

Oracle's "Big Data, Bigger Opportunities" study, released July 10, 2012, surveyed executives at utilities with smart meter programs in place and gauged perceptions regarding the business impact of "big data," preparedness to handle data growth and plans to extract optimal business value from this data to better target, engage and serve customers.

The good news from the Oracle study:

  • The average utility with at least one smart meter program in place has increased the frequency of its metering data collection by 180 times, collecting data once every four hours as opposed to just once a month.
  • Utilities with smart meter programs in place say they are somewhat prepared to manage the data deluge, rating themselves 6.7 on a scale of 1 to 10.
  • Around three quarters of utilities are collecting critical information such as outage and voltage data, and many are using it to support business decisions, improve service reliability and enhance customer satisfaction.

Despite improvements, 45 percent of utilities still struggle to report information to business managers as fast as they need it, and 50 percent miss opportunities to deliver useful information to customers. Utilities see a need to improve their ability to translate information into actionable intelligence and leverage data for strategic decision-making. 64 percent say it is one of their top three priorities. In addition, utilities plan to leverage smart grid data to improve operations and customer service in many specific ways, the study found:

The Next Five Years: What Utilities Plan

Provide customers with information about their usage patterns


Implement and/or improve conservation and efficiency programs


Implement demand response programs


Compare historical data to identify trends and forecast demand


Improve compatibility with regulatory requirements


Alert customers of usage spikes


Use predictive analytics to minimize outages


Target customers for new programs


As utilities undertake such ambitious and comprehensive changes, we at Oracle believe they will do well keep a few key take-away principles in mind.

Get ready to use data across the enterprise. With data coming in from every corner of the business, utilities must not only make data collection a priority, but also invest in the systems and people needed to analyze a growing number of new data sources collected from smart meters and other smart grid components. Utilities need an effective way to organize, display and analyze this data, so that the network can quickly be viewed to understand trends, identify issues and resolve outages in an efficient, timely manner. This is the process part of the problem.

Tame the data. Business intelligence applications help users access, organize and use the data needed to monitor activity, make decisions, and meet goals. Business intelligence tools are available for specific utility applications to pull relevant data from utility core applications using pre-built extractors. The extractor transforms and moves the data into a data warehouse, pre-built data models or schemas that structure the data for fast, accurate analysis. Business intelligence tools then transform the data into information and move it to graphical and spatial dashboards. Specifically, such extractors and schema pull data from specific utilities production applications and format it for business intelligence purposes; spatial dashboards display information on maps; and graphical dashboards analyze and summarize information in charts, traffic lights, top 10 lists, alerts and similar tools. Drill-downs and filters help users focus on essential issues.

Tackle ownership issues. Data ownership resides in various organization departments. Smart meter/interval data should be considered enterprise-level data, so utilities must ensure they have an enterprise data strategy in place. Securing strong sponsorship from a business leader who understands how critical it is to have information drive performance is probably the most important aspect of deriving full value from technology initiatives.

Remember the customer. In addition to streamlining business operations, successful data management should greatly improve the customer experience, both through improved outage management/service reliability and stronger customer communication around smart grid changes and benefits.

Oracle's "Big Data, Bigger Opportunities" report indicates that the vast majority of utility executives are working to enhance their ability to glean real intelligence from smart grid data. The goal is ultimately to create new opportunities to improve service reliability and deliver useful information to customers. As Rodger Smith, senior vice president and general manager of Oracle Utilities, has put it, "utilities can benefit from establishing enterprise information strategies, and from investing in the systems and people needed to make better business decisions."



b williams

Bradley R. Williams is Vice President, Industry Strategy, at Oracle Utilities Global Business Unit, which is responsible for outage management, distribution management, mobile workforce management, work and asset management, and load analysis utility applications. He is in charge of Oracle's smart grid strategy. Previously, he was a research director at Gartner, director of technology development at PacifiCorp, and a supervising engineer at Southern California.