Predictive analytics can empower utility companies to deliver better customer service and become more profitable while enabling them to respond in real time to any number of issues including outages, thefts, and spikes in energy use.
Utilities can analyze the massive amounts of data they capture via new smart meters to operate more efficiently, enhance customer experience, make better buying decisions, predict and detect outages, and protect their revenue by detecting thefts, notes Scott Zoldi, vice president of analytic science at FICO, in an article on Information Management.
However, rather than just store and analyze this data on a static, historical basis, utilities can use batch analysis to process the data in real time, Zoldi says.
For example, utilities can cluster customers based on their energy usage every day. At a macro level, utilities can detect common patterns for better load balancing and determining which customers are more likely to benefit from programs to help reduce their energy use, according to the article.
“This is the basis of predictive analytics, where you can forecast with a high degree of probability future behavior,” Zoldi says.
Additionally, to really benefit from Big Data and predictive analytics, electric utilities can build “streaming analytics infrastructure[s] that use real-time data to help them make the right decisions at the right time,” he notes.
Streaming analytics can also help utilities fight fraud. Utilities lose as much as $6 billion every year from stolen power, including when contractors or homeowners install bypass lines, according to Zoldi. By measuring energy use, utilities can predict where the fraud is likely to occur on the network.
Additionally, with streaming analytics, a utility can differentiate between two customers who use the same amount of energy each month, but exhibit very different usage patterns, he says. This information will allow utilities to determine which customers they should target with energy-reduction programs.
Utility companies can also use analytics to detect outages and predict failures, Zoldi notes. Typically there are fluctuations in electric usage, voltage, and other equipment parameters that occur before a failure.
Studying this short-term behavior and typical long-term behavior, utilities can predict the probability of possible failure in the near future—enabling them to more proactively manage inventory and repair critical network infrastructure, Zoldi says.
Using predictive analytics to drive insights from Big Data, utilities can “generate significant efficiencies in an industry with annual revenues of more than $300 billion,” according to Zoldi.