A revolution in data analytics is hitting electric cooperatives, and the experts following it say the co-op world will be transformed.
"We're really just starting to scratch the surface," says Jim Spiers, NRECA senior vice president for Business and Technology Strategies. "The balance of our careers, data is going to be key, and the tools to understand data—artificial intelligence, machine learning, data mining—all of these are going to be important aspects of how we get value."
Increased operational efficiency, better planning, new services, more member engagement, and better rate design are among the benefits observers see ahead.
Cobb EMC provides an early glimpse at some of the ways the data revolution is likely to change co-ops.
The Marietta, Georgia-based co-op analyzed member data from its own databases, along with more data it acquired by working with a marketing and survey firm, to develop a sophisticated profile of its members that goes far beyond standard demographics.
Cobb, one of the nation's largest distribution co-ops, serves more than 180,000 members in a territory spread across five counties that includes suburban Atlanta. The data-driven research allowed segmentation into five primary categories to determine the most effective ways to reach those groups.
[panel title="Sidebar:%20'The%20Cobb%205'" sub-title="Using%20data%20to%20reach%20members%20most%20effectively%20" link-text="Read%20more" link="/remagazine/articles/Pages/Technology-Trends-The-Cobb-5-Data-Analytics.aspx"]Advanced data analytics helped Georgia's Cobb EMC segment its membership into five groups, which they use to identify members' interests and define the best method for outreach.[/panel]
"In the co-op world, this type of information is going to become very, very valuable as we move into the future," says Kevan Espy, Cobb EMC senior vice president for marketing and corporate communications. "When you look at how well you can target market and how you can save money—the efficiency—it's just huge. Why send out a bill insert for 180,000 members when you know half aren't going to look at it?"
Spiers says the growth of co-op data analytics is part of a larger transformation in business and industry driven by the massive amount of raw information available through digital communications. Personalized, highly targeted marketing based on consumer data is already common in retail. Companies also have come to rely on sophisticated data analytics for new levels of efficiency in supply chain and distribution management.
Cooperatives are beginning to more fully embrace the use of big data, overcoming concerns about the perceived cost and complexity, Spiers says, along with privacy questions. Some co-ops are even hiring data analysts, recognizing that the benefits can outweigh the costs.
Consumers who are worried about privacy should consider that co-ops, created to put their members' needs first, are particularly sensitive to privacy and security concerns, he adds.
'Shifting the Power'
[image-caption-right description="%20." image="/remagazine/articles/PublishingImages/woman-desk-dec-2018.jpg" width="150" /]
Cobb EMC's use of different forms of communication to reach unique member subsets is one example of how analytics allows co-ops to engage on a new level. Vern Dosch says it's only the beginning.
Dosch is president and CEO of
National Information Solutions Cooperative (NISC), which is in the business of data and counts Cobb EMC as a member. Better use of data could allow cooperatives to tailor services to meet specific needs of individual members, he says, providing things like time-of-use meter readings to help manage their power use or a choice of what time of month they'd like to be billed.
"This is shifting the power—no pun intended—from the utility to the consumer," Dosch says. "Today, the data is allowing us to educate the member and provide them with options."
It can also tell a cooperative where it needs to engage members with new information. For example, Cobb EMC's analysis indicated an interest in electric vehicles among significant segments of its membership, so the co-op's EV program was accelerated, Espy says.
"We started doing some drive events," he says. "We developed some education plans. We incorporated [information about electric vehicles] into our annual meeting."
The bottom line, Dosch says, is that applying analytics to solid consumer data allows a co-op to get ahead of the curve when it comes to members' priorities and interests.
"Proactively being an advocate for the member is what the data and data analytics allows us to do," he says.
'Actively Controlling the Grid'
[image-caption-right title="." image="/remagazine/articles/PublishingImages/woman-data-dec-2018.jpg" width="150" /]
Data analytics is already shifting co-op operations and expectations. Electric utilities once considered 5 percent line losses acceptable, says R.B. Sloan, president and CEO of SEDC, the utility software solutions co-op.
"We now have co-ops that are talking about 3 to 2 ½ percent, and I think through analytics that's very obtainable," Sloan says. "To me, that's an indicator of things coming down the road."
Cooperatives are cutting line loss by analyzing data coming in from several sources.
"We look at AMI [advanced metering infrastructure] data. We look at SCADA data. We look at billing data. We look at transformers on the line, lights on the line, and more," Sloan says.
With that information, co-ops can get a detailed profile of kilowatt-hours supplied and actually used on a particular segment of line, which allows them to track down losses and, Sloan says, "optimize the use of their systems."
An even higher level of analytics lies ahead as more and more devices on the grid are equipped with sensors and remote-control capability, says Steven Collier, director of smart grid strategies at Milsoft Utility Solutions.
"One of the big future applications of data analytics is being able to know the output of all those devices and being able to control them—rooftop solar, for example, or a solar farm," he says.
The gains in efficiency will reduce costs, benefiting rate-payers while increasing system reliability. They will also significantly reduce demand for new generation, Collier says, playing an important role in a future where new power plants could be extremely difficult to build.
The change in the volume of data available to utilities will be geometric in scale.
"People thought we had big data when they were taking meter readings from AMI every hour," Collier says. "We're talking about a whole new phase. The data now, there's so much more of it; it's much more diverse, and I'm going to take action based on it. I am now actively controlling the grid."
But that control will be largely automated because of the speed at which the system will move.
"This is no longer going to be done by a dispatcher in a dispatch center," Collier says. "This is going be done automatically. … It's real-time monitoring and control of the grid to optimize supply to load, to optimize voltage."
On the local level, better analytics will also enable automated distribution management systems (DMS) that can proactively reduce peak demand, optimize operations, and detect faults.
"It's a computer and a group of sensors that operate automatically, so that even the smallest co-op or a building or a city or a commercial operation could acquire and install it and have this automated system management," Collier says.
Overall, the experts say, data analytics combined with the artificial intelligence systems necessary to manage the vast inflow of data mean the smart grid will, at last, be truly "smart."
More Robust Forecasting and Planning
[image-caption-right title="." image="/remagazine/articles/PublishingImages/man-desk-dec-2018.jpg" width="150" /]
Data coming in from AMI, SCADA, and other systems is already enabling a shift in forecasting and planning. The volume of system information allows a much more granular look at grid performance. Just as important, it means that forecasting can be based on dynamic models that no longer rely on static snapshots of past performance to predict the future, but by tracking changes on a near-real-time basis.
Sloan says those capabilities should only accelerate as analytics becomes both more common and more sophisticated, and the benefits will reach into all facets of co-op operations and down to individual households.
"We recently started using artificial intelligence to identify trends," Sloan says. "We're working with one co-op that has been able to identify, on a daily basis, what is the expected use of a customer, and if there is a variation that is not explained by temperature or other things, then we will notify that customer that there is something going on."
Another example of the changes big data and analytics will make come in the area of finance and co-op audits.
"Normally when they perform the annual audit of the co-op, they take a sampling of the transactions," Sloan says. Now, auditors will be able to analyze every transaction as part of a review. "That's very practical if you have the right tools."
The experts note, however, that essential to better forecasting and planning will be ensuring that the benefits of analytics are actually employed. Simply using new capabilities for old approaches to planning will fail to take advantage of the full potential of data analytics.
Better Rate Design
[image-caption-right title="." image="/remagazine/articles/PublishingImages/man-woman-desk-dec-2018.jpg" width="150" /]
Data-driven improvements to operating should help utilities hold down rates, Dosch says. But there's a further benefit to analysis.
NISC is working with
National Rural Utilities Cooperative Finance Corporation (CFC) to develop rate structures based on verified use from meter data rather than traditional methods.
"What we're doing now is rate studies rather than using models," Dosch says. "We're saying, 'No, let's use actual interval data from the meter."
Analyzing consumer data can also help cooperatives tailor rate options to fit the lifestyles and desires of their members. Cobb EMC is rolling out several different rates that reflect, in part, insights gained from data they have gathered and analyzed.
One is a new "Night Flex" rate, which Espy describes as essentially a time-of-use rate.
"You get 400 kilowatt-hours for free between 12 and 6 a.m.," he says.
The rate reflects the high priority key member groups are placing on electric vehicles, which can be charged at night.
Another option allows members to pay their electric bill in equal monthly installments over the year. The idea is not new and is popular with many utilities, but Cobb EMC turned to its analytics to spread the word about the service it calls "Even Bill."
"We went through the data to find the people who use Even Bill and then went back into our system to see how many people profile like them," Espy says. "Then we can do a special targeted marketing to them to increase the adoption."
[image-caption-right title="." image="/remagazine/articles/PublishingImages/big-data-dec-2018.jpg" width="150" /]
The first challenge that comes with the new era of data analytics lies in the volume of information available. More data has been created in the past five years than in the previous history of the world, according to Bill Hoffman, a former data analytics expert for U.S. Bank and the CIA.
The volume of data creation is growing exponentially, doubling every two years, according to the experts, and by 2020 could amount to 44 trillion gigabytes. The data coming into cooperatives through AMI, SCADA, and other systems, along with outside sources, is only a small part of that total, but it still represents a massive shift in the amount of information a co-op has to process.
That total will only grow as smart devices and sensors proliferate along electric power lines and as more detailed consumer information becomes available through other sources.
Cobb EMC saw firsthand the challenge of volume.
"The first time I ran a query, my whole computer died immediately," says Nurdan Cornelius, Cobb EMC director of consumer marketing. "That's way too much data."
Milsoft's Collier agrees that the flood of data will be a challenge, and it will be critical for co-ops to acquire the capabilities or find the help necessary to stay on top of it.
"One of the key elements is going to be to take advantage of, and not get overwhelmed by, all the changes happening on the customer side of the meter," he says. "Be prepared to acquire the systems to deal with them."
A greater risk comes in believing that data analytics is something your co-op can ignore, Collier says.
"I think the biggest danger is trying to avoid this, thinking, we don't need to do this; we're a little co-op; our members aren't that sophisticated," he says. "If you do that, there are going to be sophisticated companies that are going to come in and do this for your consumers," by offering DMS and other services.
SEDC's Sloan says cooperatives, like all businesses, will need to avoid falling in love with data for data's sake and believing the answer to problems can always be found in acquiring more and more information.
"My training is as an engineer, so I love data, but at the same time, once you develop the capability of really delving deep into the data, it becomes almost a challenge: How much deeper can I get? I do think there's a point of diminishing returns. There's a point of paralysis by analysis."
The key will be to identify the places where greater analytics can truly benefit the cooperative and keeping those objectives in mind while investing in stronger collection and processing capabilities. Analysts say the focus should always be on how analytics can help the co-op succeed in its overall mission of serving its members.
NISC's Dosch notes that security and privacy concerns will always remain paramount when it comes to consumer information. Greater regulation to increase security and protections over consumer data could gain momentum at the state and federal level. But cooperatives have an organizational culture that gives them an advantage, because putting their members first has been part of their DNA from the beginning, he says.
In the end, Dosch and other analysts say the challenges that data and analytics present are outweighed by their potential to transform operations for the better.
"I know this is an overused phrase, but this is really a game-changer for the industry," Dosch says. "The efficiency, the savings, the empowerment of members—all that is enabled by data."