Sponsored content.
Utilities are increasingly being asked to do more with aging infrastructure, incomplete asset records and rising expectations for reliability and resilience. A recent Factor This and DTECH webinar explored how visual inspections powered by artificial intelligence are helping utilities close long-standing data gaps and turn infrastructure itself into a systemwide dataset.
The live webinar, Infrastructure as a Dataset: Unlocking Utility-Wide Insights from Visual AI Inspections, first aired Dec. 9 and is now available on demand. The discussion examined how utilities are moving beyond manual, cycle-based inspection workflows by combining aerial imagery, computer vision and machine learning to build what speakers described as a “living infrastructure dataset.”
Lyra Park, vice president of utility sales at Zeitview, said many utilities underestimate how incomplete or inaccurate their existing asset data has become, particularly when legacy infrastructure predates modern GIS systems.
“More often than not, we see a data profile that has either no data, and even if there is data filled in, it’s often inaccurate,” Park said, adding that this makes it challenging to address all the initiatives and work streams utilities have been tasked with.
The webinar focuses on how aerial inspections using drones can dramatically improve both asset location accuracy and condition visibility across transmission and distribution systems. Speakers described how utilities are using high-resolution imagery and AI-assisted analysis to detect a greater number of equipment issues than traditional ground inspections, while also providing field crews and engineers with better information before trucks roll.
Brenda Green, manager of engineering programs and technical services at Powder River Energy Corp., shared real-world experience from her cooperative’s multi-year inspection program covering roughly 200,000 poles across a vast rural service territory. Green said the volume of data collected was initially daunting but ultimately transformative.
“You don’t know what’s wrong on your system until you know what’s wrong on your system,” she said. “Addressing all of those defects and anomalies and everything that you find is a big deal.”
