Managing the utility’s enterprise information
TransAlta Utilities, and others, tackle data
Published In: Intelligent Utility Magazine May / June 2012
ENTERPRISE INFORMATION MANAGEMENT AND smart grid integration are the heart and lungs of the new utility world. That is, if we accept that smart grid integration is the heart, then enterprise information management is surely the lungs of the intelligent utility.
Enterprise information management, at its core, means the structuring and governance of information assets across the utility, overarching the historical boundaries that have defined IT information and data, operational information and data, customer service information and data, and so on. In so doing, the utility has the ability to take enterprisewide ownership and responsibility for the data, including its accuracy and quality, in order to improve both the data itself, and the company's operations.
While enterprise information management (EIM) is not a new term, its importance has come to the fore more recently, as intelligent utilities are amassing more data through devices throughout the grid, and more information about electricity usage patterns, load shapes and more.
At the recent Utility Analytics Summit, a number of utility directors discussed their road maps to enterprise information management and integration, and shared their experiences to date.
Strong business team critical in efforts
Hydro One, one of the largest electricity delivery systems in North America, controlling 100 percent of the transmission in the province of Ontario as well as a 123,528-kilometer distribution system, set up its information management program in 2008 and set both short-term and long-term goals for it at the time, said Brad Bowness, Hydro One's director of business architecture and IT, with the strategy being set in the first year.
"It took us one-and-a-half to two years to implement the recording stack," he said. In 2009, the system's foundation was established, and it went live. With that experience then under its belt, Hydro One spent 2010 and 2011 stabilizing and enhancing the program, and started 2012 with what Bowness described as "continuous improvement."
"Taking two years to stabilize the system is important," he noted.
Hydro One's asset analytics initiative brought information from the operations technology side of the house and correlated it with other data. "A strong business team is critical (in the information management program effort)," Bowness said. "During the initial implementation, we had a key business team, who have lived and breathed it through the entire life cycle."
A valuable work in progress
Dean Balog, a self-described "data dude" and TransAlta's director of enterprise applications, discussed the three areas of focus of TransAlta's EIM: data governance, business intelligence and data architecture and modeling.
It's a work in progress: the data governance program launched the week before the February summit, Balog said. "This year, the focus is on process, roles and responsibilities, and a focus on data quality," he explained. "We need to have people understand what data governance is, and their roles in it. That's the role of the pilots."
The utility is also focusing, in its early implementation, on data quality. Here is an area in which the old adage, "garbage in, garbage out," holds true, as decisions are made based on the data available, and bad data can result in less-than-optimal decision-making.
TransAlta already has some early data governance successes. Data is now a priority, directly from the CEO on down. As well, Balog said, "business is on board and motivated." With that, he also shared lessons learned: "Persistence is important-three to five false starts are normal," he said. "Look for small wins, and communicate success widely. Finally, external domain experts can help to explain the concepts."
Balog says the next steps for TransAlta's EIM program include a transition from pilot projects to a sustaining EIM practice, with information life cycle management rolled into the mix. The utility also plans to link data governance to business intelligence governance. New tools will be investigated to assist the utility in its EIM focus. And, of course, a policy review is on the "to-do" list, as well.
A single version of the truth
First Choice Power, based in deregulated Texas, formed a centralized analytics group a few years ago. The prime challenge, according to Lloyd Tokerud, the utility company's manager of analytics, was the complexity of the data. First Choice's centralized analytics model promotes what Tokerud describes as "truth and consistence." (This has also been described by others as a "single version of the truth" for data over time.)
Further, the centralized solution adopted by First Choice means education and training are reduced, and live data sessions with executives accelerate decisions.
"Our philosophy is, first, start big, and start in the middle with one thing that matters most," he said. "And," he added, to knowing chuckles from the audience, "speed, accuracy, scope: pick two, because you can't have all three."
First Choice uses visualization to accelerate business intelligence, or "prove it, then build it."
"We gather the information manually, then use a visualization tool to rearrange the data in real time for consensus, so that we can build only what's relevant," he said. "This kind of approach has worked well for us."
Making decisions more easily
The underpinning of Southern California Edison's analytics model, SCE's Ron Grabyan, manager of business intelligence services, said, is "enabling the business to get to the data to make their decisions. We've been doing continuous improvements all along, small changes. We want to make big changes."
Part of those changes include migrating SCE's entire data warehouse into an environment that will allow analysts to get their data faster.
The current environment, he said, has many challenges, including report response time, data loads, and build and maintenance costs. "In the 2012 environment, however, we are not developing as many objects."
The benefits of this new approach are many, but these two are significant. In the area of performance improvement, SCE is getting reports approximately two to three times faster, and data load is proving to be three times faster.
"We can provide an environment for analysts to get their data faster," Grabyan said.
Consumers Energy has approached enterprise information management in a somewhat different fashion, according to Stephen Hirsch, director of data quality for the utility.
Consumers Energy, an integrated gas and electric utility in Michigan, launched its information quality program in the mid-2000s as it began replacing a range of legacy systems with an enterprise-wide system.
"The high-level goal of our project," he said, "is to enable customer experience value, to use customer data for competitive advantage, to facilitate safe and reliable actions and to help ensure consistent financial performance."
In terms of program value and outcomes, Consumers Energy is looking for quality, accurate data that enables: detailed segmentation (enabled by customer data); a better categorization of customers to facilitate providing the appropriate programs to the appropriate customers; an improved understanding of parent/business partner relationships; and lower invoice processing and anti re-work costs.