Survey shows overreliance on manual data collection and analysis for managing strategic assets
2020 has been an eye-opening year, causing leaders across myriad geographies, sectors, and industries to step back and rethink their preparedness strategies. This is especially true for utilities as record heat, storms, and natural disasters have wreaked havoc on critical infrastructure.
With one- to five-year planning cycles now in full swing, grid modernization conversations are underway. But how can utility leaders better prepare for and manage outages, while also working to combat them altogether through smart grid initiatives focused on mitigating climate change?
Better data collection and analysis methods can help utilities streamline operations while modernizing for the future.
Survey highlights urgent need
Strategic asset management should be the first step in any major initiative for utilities. As utilities continue to set forth or update existing integrated resource plans, they’ll require more analysis on which assets are performing optimally, which aren’t, and even where new assets should be placed.
But your analysis is only as good as your data. Today, utilities rely on four main methods for collecting geospatial data:
- Ground-based inspections
- Satellite imagery
- Manned aircraft
Many utility asset managers have been overly reliant on antiquated methods like ground-based inspections and notetaking to collect data, making it challenging to access real-time information when deeper insights are needed.
In fact, a recent survey focused on “Power Grid Inspection Insights to Improve Safety, Efficiency and Value” conducted by Utility Dive on behalf of PrecisionHawk found that:
- More than half of respondents across 100 utilities—ranging in size from small organizations managing between 1,000 and 10,000 miles of power lines to the largest utilities, each operating more than 100,000 miles—reported that they are still utilizing manual field inspections.
- Nearly two-thirds said their organizations are not using advanced inspection technologies very much or at all—including 10% who have not even considered them yet.
Not only are data capture methodologies largely outdated, but the data itself is dated. 40% of respondents responsible for utility company asset management and modernization revealed they’re only inspecting grid assets just once per year, with some saying they are conducting inspections even far less frequently, every two to five years.
With this, it’s no surprise that 38% report being “negatively surprised” by an asset’s condition during routine inspections.
When asked whether their organization had ever been surprised by the condition of a grid asset and what might have helped to mitigate any issues that resulted, survey respondents underscored the need for better asset management.
- “Sometimes [the damage] is significant, like a burnt-out pole top or cracked insulator. More frequent field inspections and better data analytics could help to prevent future issues.”
- “We found a gassing transformer. Not surprising, but certainly unpredicted. Automated online monitoring is the key to closing the unpredictability window.”
- “Some assets had failed. Improved data analysis and inspection standards would help.”
As these examples show, more timely asset condition insights can positively impact both utilities and their customers. It can mean the difference between uptime and downtime, customer cost savings and rate hikes, growing margins, and crippling expenses—even life and death, as one respondent said unnoticed asset problems have caused wildfires.
Clearly, opportunity exists for utilities to improve their data collection methods. In many cases, drones offer the best solution, providing a safer, cheaper and faster way to collect geospatial data while producing richer data for analysis. Despite this opportunity, only 22% of organizations surveyed use drones for asset inspections.
Combining AI with human judgment
Once you optimize your data collection methods, automating your data processing will further improve outcomes. Automation is the process of generating the desired outputs from a set of provided inputs with little to no human intervention.
Modern utilities combine artificial intelligence (AI) with human skills to create a streamlined end-to-end data workflow for managing strategic assets—one that increases accuracy and efficiency without incurring unreasonable costs. This is especially poignant given organizations surveyed spent an average of $8.5 million on ground-based inspections in 2019—money that could go a long way toward implementing better and safer methods.
With automation, humans are involved at the beginning and end of the workflow. At the beginning, data engineers research, design, and implement the machine learning models. At the end, data analysts analyze the outputs of the model for quality control. AI and machine learning handle the middle phases to create a manageable dataset that includes only the most relevant information.
By turning massive amounts of data into actionable insights, utilities can pinpoint which assets to repair or replace, and reallocate human resources to more valuable tasks.
Plan for the future
Utility companies may be leaving themselves in the dark when it comes to fully understanding what’s going on across grid assets. Their existing strategies aren’t capturing the telltale signs of impending failure or helping them properly plan for the future.
While even the smartest grid can’t outsmart Mother Nature, a modernized approach to grid modernization based in deep analytics will help utilities better plan for 2021 and beyond.
Kristen Ellerbe is presenting in a DISTRIBUTECH+ session titled, Quantifying and Evaluating Storm Resilience and Disaster Preparedness, which takes place on Thursday, November 12 at 12:30 Eastern Time. Click the link and register to see the full agenda! We hope you’ll attend.