Article Post

Utility industry's data wealth challenges can become lucrative assets with proper care

Asset Health Framework

U.S. utilities provide reliable electricity that has driven the economy for more than a century and provided a quality of life so good and seamless that most people never give electricity a thought. In fact, the National Academy of Engineering ranks electrification as No. 1 among its 20 Greatest Engineering Achievements of the 20th Century.

Today, utilities are at a crossroads of sweeping change brought about by the advent of renewables, changing economics and demographics, increasing customer expectations, and a dizzying array of regulatory changes. While many of these changes are creating new challenges for utilities, they also bring new opportunities for improving the industry’s ability to continue providing affordable, safe, reliable, sustainable electricity.

While many of these changes in the renewables and customer engagement areas often get top billing, there also are some interesting shifts happening on the grid operations side of the business, including how a utility manages its T & D assets and a data-rich operating environment. The Utility Analytics Institute leads a group on asset health, and the group has explored some of the challenges and opportunities faced by a large investor-owned utility (IOU) in the western U.S. This article will examine some of the challenges the IOU has faced.

Data, data everywhere

The list of data sources available to utility leaders today is long and growing. Data comes from EMS/DMS/SCADA/Distribution Automation, Outage Management Systems, geospatial and GPS applications, and Enterprise Asset Management (EAM) Systems. Add legacy data from old systems, manual data (like hand-written notes on map books in trucks), and spreadsheets, and data quality becomes a challenge. Data quality must be a key initiative in an organization where data is considered a strategic asset.

The IOU’s principal maintenance and reliability strategies manager projects a vision for integrating the use of operational asset data such as loading, voltage, harmonics, exposure to faults, magnitude of faults, exposure to transients, temperature, and operational signature. If gathered and analyzed effectively, these factors can help bridge the gap between managing an aggregate inventory of assets in a globally optimized way, and managing individual assets in a way that is both locally and globally optimized.

As the entire utility industry faces the challenge of fully leveraging data, efforts like the one profiled here start to affect data- and analytics-driven improvements in asset management practices.

Another example includes the efforts of the Asset Health Focus Community of the CIM Users Group (CIMug). The group’s initial efforts are documented in the Electric Power Research Institute (EPRI) Technical Update, “Standard Based Integration Specification: Common Information Model Framework for Asset Health Data Exchange.”

As a starting point, the EPRI report includes the Asset Health Integration Framework highlighted in the image above.

A copy of the EPRI report can be downloaded here.

A framework for the notorious “Four Vs” framework

The “Four Vs” of big data – volume, velocity, veracity and variety – enter into many discussions about analytics. While this addresses data challenges, it’s also about building the technology infrastructure that makes analytics (asset analytics in this case) possible. The data sources are growing, and so are the technologies that manage it. Did anybody know what a “data lake” was a few years ago? How will cloud data management applications facilitate better, faster, more reliable data-driven solutions? What technologies or systems will ultimately provide that path to data nirvana where data is open, available, clean, distributed and secure?

The IOU director responsible for this area sums up their infrastructure requirements by saying they need more direct access to data that’s protected for security reasons. They need a platform to dump large quantities of data without fear of corrupting the source, while still making the data useful for asset management applications both locally and across the enterprise.

A leap (okay, maybe a few steps) forward

If a utility’s asset management team has the world’s best data and the newest technology in place, what could go wrong? A lot actually.

Business processes and maintenance practices are important. A lesson learned in early attempts to implement EAM (formerly known as “CMMS”) revealed that it’s not enough to automate a manual process. The process needs to be reinvented to reflect the improvements available in the EAM. This holds true as utilities move deeper into-data-driven decision-making. Many asset management practices are decades old and do not account for the availability of data or affordable computer processing power.

The bulk of movement in the asset management world is toward predictive operations: change out that asset before it fails because doing so increases reliability, averts safety issues, and can save money over time.  T & D utilities have not made the leap from reactive (time-based) maintenance, to predictive (condition-based) maintenance, but the industry is on its way. The IOU leaders sum it by saying that as data improves, so does their ability to accurately predict equipment failures. It changes how they identify T & D infrastructure replacement scope and manual processes that will be replaced by analytical processes that better determine the optimal scope based on the annual objectives. IOU leaders see a trend toward increased analytics and data scientists in an era where operational processes will be influenced by analytic processes.

It’s okay to ask why?

Data and technology often leads to a focus on the “what” and “how” of what we are doing. In reality, we should focus on the “why” of our activities. Aligning analytics initiatives with key business drivers is critical.

The leaders at the IOU are moving toward predictive analytics and developing talent and capabilities in that area. For example, they’re creating a risk score model that predicts the likelihood and consequence of transformer failure. The CEO has sponsored the initiative, and multiple use cases have been successfully completed. The success has generated senior-level interest in developing a dedicated advanced analytics function. Predictive maintenance has the potential to completely transform how the IOU identifies, funds and executes infrastructure replacement. Today when utility revenues are flat or shrinking, gaining efficiencies is key.


Effectively manage large-scale data migration challenges (over 70 legacy systems/data types readily supported)  Twitter: @dsiintl

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.