The mission of this group is to bring together utility professionals in the power industry who are in the thick of the digital utility transformation. 

Post

Data Intelligence Is Creating Real Progress in the Utility Sector

image credit: © Wrightstudio | Dreamstime.com

This item is part of the Special Issue - 2020-12 - Data Analytics & Intelligence, click here for more

The utility industry has felt pressure to undergo a digital transformation for a long time now, and with the recent COVID-19 pandemic, it only became more apparent. Why though, is it so important? In the end, there are multiple goals enterprise management wants to reach: to enhance the customer experience for their users, to create a more comfortable workplace for staff, and to achieve the ideal revenue flow. All of them can be achieved with help of technology. 

Recent events have woken everyone up to the need for resilience in the time of emergency. Automation, remote access technology, collaboration platforms help businesses become more flexible and better adjusted for whatever disruptions may come. That is why a lot of companies started to focus on adopting the appropriate software and building their infrastructure accordingly. However, many miss one of the most important elements — to get real value out of digital transformation, one needs to know the exact points of its application.

Your access to Member Features is limited.

The topic of big data is trending all over the technology news websites right now. Many companies are collecting and storing massive volumes of data, in hopes to enhance their business performance. Is this data useful, though? The answer depends on what happens after it is gathered. Without the tools to process the massive amount of information, it becomes just a heavy ballast laying around, taking up space. It is impossible to get the real value from such data without machine learning models and trained artificial intelligence, as there is too much of it. However, if wielded correctly, it can unlock a powerful potential for business opportunities, shows the recent research by Deloitte.

The means to know what customers need, what employees would benefit from, and which pain points in the system should be covered first, is data intelligence. Data intelligence is the practice of using artificial intelligence and machine learning to analyse and transform massive datasets into intelligent insights. In this article we will look at some examples of using data gathered by utilities, artificial intelligence (AI) and machine learning (ML) creatively.

Asset Management

For a while now, utilities are installing smart meters, upgrading distribution lines and new hardware, implementing other innovative solutions that generated terabytes of data, that should be categorized and managed correctly. With help of IoT analytics, utility companies can decipher which signals are more relevant, and build the reaction protocols to be quicker and more effective. 

Given the proper reporting from the devices and their adequate analysis, the usual maintenance schedule can be replaced with the intelligent one, based on the current situation rather than on the average estimation. For example, this way underperforming assets can be found and optimized fast, contributing to revenue growth.

Edge algorithms, combined with real-time reporting from the assets, help integrate distributed energy resources (more often renewable ones) into the grid, preserving its stability.

The result can be an extended life period of the devices, the possibility to run facilities at peak performance without the risk of outages, renewable sources integration, faster power recovery after natural disasters, better user experience and fewer disruptions.

Fraud Prevention and Better Security

Knowing the patterns of consumption creates an opportunity for better detection of unusual spikes and faster identification of unsanctioned activities. Additionally, targeted investigations are much easier conducted and yield more results with sufficient usage, billing and payment data. When done well, analytics can consistently reduce fraud losses by 3 to 5 percent in mature environments and by over 30 percent in evolving contexts.

Many companies use a combination of the following methodologies to achieve better results:

  • Out of pattern analysis. The process of comparing user activity with peer group behaviour or historical data of own past behaviour, which helps identify unusual events.
  • Linkage analysis. The process of using networking activity analysis, which helps identify entities associated with known types of fraud, as well as practices used by those similar to them.
  • Rule development. The process of creating and applying rules for basic business activities to spot unusual trends, as well as specialized rules for specific transactions.
  • Model development. Helps create fraud-scoring tools and detailed statistical analytics to provide quantitative insight into possible fraud activity.

The rule-based model executes the risk profiling of the customers, simplifying the creation of the preventive strategy. More thorough monitoring of the particular premises may not only reduce the unbilled resource usage, but help identify the potential problems like leakages, equipment dysfunction, irresponsible facility handling, etc.

Better Customer Service and Education

According to Acquia, 41% of brands noticed that more customers return after being targeted by personalization initiatives. This is a very expected result, as in the age of being bombarded by many offers, people crave connection and are more likely to support a company that cares about their unique needs. Utilizing data to anticipate the needs of end-users can become crucial for utility providers to prevent complaints, solve issues quicker and foster trust between the company and consumers. Using ML and data available from household devices, companies can also predict and take measures to prevent customer churn.

Outage management can also benefit from the data analysis, with faster identification of the weak points in the grid and potentially risky assets. Preventive measures can then be taken, prolonging the life of the devices, strengthening the grid, avoiding costly disruptions and enhancing customer satisfaction.

Consumption pattern identification and analysis will give utilities a powerful tool for educating their customer base about power conservation and smart usage. This can prove equally beneficial for the company and the users, saving money for the latter and supporting the sustainability of the former. 

Providers can use data for both personalized offer packages that help households be more frugal and for delivering more customized invoices, which supply more information for the customers who want to manage their bills. Livework has found, that 42% of energy customers are not satisfied with their billing experience. Such a situation is damaging for the provider-user relationship, undermining trust and lowering the chance of retention. Intelligent data analysis can provide the means to enhance the billing experience for the customers and therefore help both parties understand each other better.

Moving the Industry Forward

Utility providers with a customer base of any size can gather and analyze data to achieve the above advantages. It is also important for companies to build partnerships and share the results of their efforts with each other in order to perfect the grid and make the industry better for everyone involved. A smarter grid and more satisfied customers will create a healthy eco-system, with opportunities for faster and easier energy trading, implementations of sustainable practices and increased loyalty.

Everyone participating in the energy business has a responsibility to support the industry’s progress by taking practical steps, and one of the vital ones is careful and smart data handling. 

Rost Bitterlikh's picture
Thank Rost for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member

Discussions

Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Daniel Burillo's picture
Daniel Burillo on Dec 16, 2020

What can utilities do to ensure quality?

Rost Bitterlikh's picture
Rost Bitterlikh on Dec 17, 2020

Data quality assurance actually might be a great topic for a separate article, thank you. 

Very roughly, each company should set data quality standards it is going to use thereon to sift through the massive of data collected. All incoming data should be profiled against the main aspects (e.g.: data consistency, completeness, and patterns) and also checked for duplicates to avoid storing and processing identical data. Undoubtfully, his process should be automated. The data pipeline needs to be clearly defined and carefully designed, with the obligatory integration of data lineage traceability. Additionally, there shouldn't be any changes introduced to the existing datasets without automated regression testing.

This, of course, is a bare minimum, which can be then built upon.

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »