Analytics: The art of the possible

What it takes to get to "what's next" in analytics

At September's Utility Analytics Week we focused on what is happening now in utility analytics through many, many case studies. It is important to discuss today's successes with utility analytics, but it is also important to keep in mind what's on the horizon. That's why I particularly enjoyed the SAS workshop session at the event. This workshop certainly had a lot of case studies, but it also a looked at what is ahead with utility analytics and what it will take to get there. A visionary presentation by Paul Dick, principal industry consultant and utilities lead, Americas Energy Practice, SAS, focused on the art of the possible with analytics. In this article, we discuss some of Paul's key assertions, which involves being open to asking a lot of questions and redefining how utilities can best serve the needs of today's energy customers.  
What used to be solved, isn't actually solved
If you think you have something figured out, think again -- and again and again. Many times, when we feel that we have overcome a challenge, it is solved. However, many utility challenges that may seem like  they have been solved, are no longer solved. For example, Paul noted, "Think about energy forecasting, you think that energy forecasting is a problem that has been long solved. Well, it's no longer solved -- demand response programs, distributed generation -- all these different new technologies and processes that are driving utilities now make this an unsolved problem. That's a good example of what we need to go back and look at again." Advanced analytics can help utilities keep an eye on these changing conditions and adapt to an ever-evolving world. 
You need to redefine almost everything
At the heart of a utility is generation, transmission, distribution and customers, and those are still the core businesses of a utility. However, with rapid changes in the marketplace, utilities will have to think about them differently, and even redefine and add on to what has become familiar to the industry. "The next five years are going to be different than the last 100," said Paul. "We need to adapt faster to that and analytics is the way to do that." 
Redefining processes 
Many utility company processes need to change to meet tomorrow's needs and fully embrace analytics. "We're trying to apply analytics to processes that were designed a long time ago before we came across analytics and all of the possibilities available with them," noted Paul. It becomes a challenge when there is inflexibility in a company's approach. "Business units don't have as much flexibility as most of us probably would like, so we keep the same process and apply analytics to those, which can be a challenge," said Paul. Ultimately, companies need to take time to revisit those business processes, and understand how analytics can reshape processes for the better.
A kilowatt is no longer a kilowatt 
A kilowatt is no longer what it used to mean for utility companies, which creates more complexity and more opportunities for analytics to better understand those realities. "A kilowatt used to be produced and consumed in a centralized, one-way model," said Paul. "Now with battery power and distributed generation -- and other pieces of technology -- we now have the ability to defer when that consumption may actually happen. Other complicating factors include incentives to not generate power. It is no longer the same equation -- it is back to being an unsolved issue." 
Business models are changing
Even with utilities' core functions of generation, transmission, distribution and customers, what does it take to be a utility? And what is the role of a traditional utility? "There are now models out there that are going to allow companies to own every kilowatt within their environment and within their geopolitical area," said Paul. "And with distributed energy resources, we're going to have figure out how are we going to own assets when it comes to distributed generation. Things like this are going to present business model challenges as well as opportunities to actually own every kilowatt. Whether you're in a competitive market or not, distributed generation is your next competitor. Customers do have options now; they do have choices." 
Questioning what is possible
These challenges, while certainly challenging, mean that there are opportunities to grow and redefine what it means to be a successful utility in a rapidly changing world. So where are the opportunities? Analytics provide the opportunity to embrace these changes and challenges, better understand customers and create new opportunities for utilities. Some examples include:
  • Storm-related customer service | Certainly there are opportunities for better managing outage restorations during storms, but there are also many other customer service opportunities for utilities. There are advantages to advanced analytics models where you can actually communicate with customers in advance of storms, and perhaps let them know that "Hey, you know what, there's a 70% chance that this storm is coming and you may experience an outage." Paul noted that, "There's nothing wrong with that, it is just being a good provider of services, and telling the customer what the impact may be, and communicating the time of repairs in advance to your customers." Analytics also could unleash new opportunities to give more information  during and after a storm, such as Paul noted, "What does have power, where can I go? Tell your customers where there is service. You know is the movie theater on? Those are some things that need to happen during storms, in addition to restoration and lifeline services."
  • Electric vehicles | Another area that is actually starting to gain traction is the electric vehicle market. "They're starting to become much more popular and the manufacturers are starting to roll them off in higher quantities and there's actually an uptake on the sales of those," said Paul. "So how are we going to deal with that? Is the network sized and ready to go? A lot of people don't know, so it is something we need to answer before you're out there replacing transformer after transformer because three or four customers have PHEVs on them that are blowing up." Another factor to consider is the real-time grid management of these vehicles. Paul posed questions such as "What am I going to do with vehicle charging? There is a product and service opportunity there -- is it a flat rate? Is it free? Maybe I get to use the energy stored within that vehicle to put that back on the grid when I need it. Am I going to have differences in non-peak rates and peak rates? That's a big question, if you get a big enough impact, I want to know how did I affect my peak. It goes back energy forecasting and how all those dynamics change that `solved' problem." 
  • Grid parity | The grid parity model is another area of opportunity for utilities and analytics. According to Paul , the grid parity model asks "When is the cost going to be equal for me as a consumer? When am I going to be able to provide energy at a cost equal to what the central grid costs me?" At the end of the day, a utility must figure out where is grid parity going to occur, and what can the company do to change the grid parity. For example, what your business model might be is that distributed generation is part of your business, and you want to embrace it."  
With so many questions and new approaches on the horizon, analytics can help define -- and redefine -- what is possible for utility companies. If you're interested in these forward-looking perspectives, please consider attending Utility Analytics Summit 2014, which will expand on what's next for utility analytics with a focus on analytics strategy. Thanks for reading!
H. Christine Richards is the research director for the Utility Analytics Institute. You may reach her at