Translating Data Analytics Opportunities into Utility Abilities

Data Analytics

by Bradley Williams, Oracle Utilities

Technology being utilized by utilities is drawing in a wealth of new, rich data, at a volume unparalleled in the history of the industry. While at first overwhelming in its sheer size, this new data offers utilities distinct opportunities to drive increased operational efficiency and reliability, improved customer service and greatly enhanced customer relationships.

Utilities can now collect outage, interval, voltage, tamper event and diagnostic flag data via smart meters. Using this data, they can identify and correct outages more quickly, provide customers with power usage information, identify power theft, and aggregate usage to assess asset risk and asset replacement issues.

Data from other sources is flooding in as well, including customer feedback from various communications channels –including social media channels–alternative distributed generation data, weather and forecasting data and data generated by sophisticated sensors, controls and grid-healing elements all along the utility grid.

Analytics unlocks New Value

Big data and predictive analytics are being described as transformative in nature; twin pillars of the new utility. Terabyte after terabyte of new data streaming in from smart grid devices has forced utilities to re-examine how they look at information strategy, operational structure and customer engagement, as well as their ability as an enterprise to cope with change and take advantage of the new data flow enterprise-wide, rather than in distinct utility organizational silos.

As operational technology within the grid has increased and matured, there has been an evolution in the amount of data being collected and in the expectations of analytics capabilities. Once utilities could analyze data only on an ad hoc, hands-on basis, through white boards and complex spreadsheets. The industry is moving to a clearer analysis, however, incorporating more available data and analyzing it more quickly. Knowing what to look for, utilities can yield insights from real-time information streams. With historical and real-time data at hand, utilities are beginning to look forward with predictive analytics, creating value to proactively mitigate potential operational problems before they arise.

No Single Path to Analytics Nirvana

Beyond the need to use the new data to improve capabilities, each utility’s specific data analytics needs and approaches are as individual as the utility. There is no one single answer, no data analytics silver bullet for all utilities. Big data and business intelligence dashboards are not single answers. Utilities still need operational reporting; they need to know their key performance indicators (KPI), such as their system average interruption duration index (SAIDI), their system average interruption frequency index (SAIFI), and more.

To manage those needs, operational data analytics can provide the rear-view mirror, as well as the real-time view, of the utility’s operations. This can help a utility in its reporting requirements, but also in its ability to deal with outages more quickly and effectively. With predictive analytics in the mix, a utility can begin to compare historical data to identify trends, mash-up weather forecasts and forecast demand to more accurately predict energy usage and grid impact of renewable generation. In addition, they can turn reactive outage management into proactive outage management by monitoring transformer and other asset health rather than use historic run-to-failure asset management.

Customer data analytics, encompassing both structured meter data and more unstructured customer data, such as social media and customer relationship data, can provide the utility with a means to better provide customers with information about their usage patterns, target them for new programs, establish pricing programs, implement more effective demand response programs and alert customers to usage spikes that may indicate an issue with an appliance.

Translating Opportunities into Utility Abilities

To take advantage of what analytics solutions offer it’s important to look at analytics across the organization, beyond the utility organizational silos. There is great value to be had in sharing and leveraging diverse sources of data across the utility enterprise. To do that, however, a utility must take a step back and look at its current business processes to use the new data flowing in. Significant value can be achieved, but it will take looking far more broadly at the business cases each utility is setting for itself.

To fully tap into the wealth of actionable information available, it’s imperative that utilities ask themselves: “What else can we do? How else can we achieve value?” This will mean looking more deeply into the pool of predictive analytics, sharing data and collaborating across the enterprise silos, and experimenting with mash-ups of the disparate types of data the utility has collected. For example, how is customer segmentation impacted by load profile? What data can we aggregate and then analyze across all silos to give us a deeper, more holistic understanding of our business? The possible questions are unlimited and, once identified, can provide increasingly stronger business cases for the utility to pursue.

Seeding the Cloud

There has been much discussion in recent years about the possibilities for cloud computing and cloud analytics within the utility sector. Recent Oracle research in how North American electric utilities are using the increasing volumes of smart grid data, shows substantial interest in cloud-based solutions for data management and analysis, and utility implementations for specific business cases are growing in number.

Cloud-based solutions offer an answer to another issue prevalent within the industry: The data analytics skills gap within utility staffing is a prominent issue for companies trying to move ahead with clearly defined data analytics business cases. One of the biggest impediments utilities encounter when leveraging their data is a lack of people skilled in big data analytics and electric utility business knowledge. The skilled electric utility personnel are too busy doing their day-to-day business to learn new analytics tools, and the data scientists and analysts don’t know the utility operations well enough to know what to look for. Pre-packaged, cloud-hosted analytics applications offer utilities solutions to these issues without the capital investment in yet more data servers, with assigned utility skilled data scientists as part of the service. In addition, cloud-based solutions can improve the speed, security and scalability of data management systems, for those utilities already in the thick of things.

While cloud is still in the early stages of adoption within the utilities industry, cloud-based analytics will allow utilities to collaborate outside their walls, with other utilities to address more complex business issues and identify new KPIs. Analytics collaboration has been focused on collaboration within the utility, but cloud-based analytics offer a prime collaboration sandbox in which utilities can experiment together. They can use the cloud as a testing site to share analytical information and analytics best practices to benefit all involved.

This evolution will allow the utility analytics practice to stay fluid, rather than static. As new and more complex analytics issues are resolved collaboratively, they will migrate into utilities’ standard operating procedures, allowing those in the collaborative sandbox to address as-yet-undiscovered issues.

The Holistic View

As a number of utilities begin to implement new analytics within their operations, now is a good time for utilities to assess and reassess their analytics strategies. Most importantly, does the current or planned strategy allow for growth? If not, why not? Where do you want it to go? What do you want it to do? Have you made allowances in your strategy for what you are unaware of?

One of the greatest challenges the modern utility faces lies in its ability to translate the mountain of new data into true, actionable intelligence, and to leverage the information gained from analysis to make decisions that will improve business performance, service reliability and customer relationships. The value is there for those willing to start the journey.

Bradley Williams is vice president of Oracle Utilities Global Business Unit’s Product Management responsible for Outage Management, Distribution Management, Mobile Workforce Management, Work and Asset Management, and Load Analysis utility applications and Smart Grid Strategy.

More PowerGrid International Issue Articles
PowerGrid International Articles Archives
View Power Generation Articles on PennEnergy.com
Previous articleForming the Groundwork: Communications Infrastructure Reliability Near and Far
Next articleCooper Power Systems Smart Grid-Ready Control

No posts to display