Inspections are a tricky business. Companies within the industrial landscape spend considerable amounts of money and resources complying to the mandatory safety measures.
Most of the time, manufacturing plants have to be shut down and equipment must be allowed to cool for several days before a technician can safely inspect it. Because the machinery has to be idle, inspectors might miss signs of trouble that are only detectable during operations. Even with all the precautions, there could still be dangers resulting from malfunctions or the spilling of chemicals, and some parts of the machine that can’t even be inspected without disassembling the whole thing. Lastly, inspectors are still human, and even the most qualified human still makes mistakes.
After an inspection, the data collected still has to be processed and analysed, and that can take many weeks.
The conclusion? Inspections aren’t just dangerous, but also expensive and inefficient. General Energy knows this first-hand and is already working on a future-proof solution for the industry.
In June 2017, at its annual Minds + Machines conference, GE announced the birth of Avitas Systems, its new startup offering inspection solutions based on advanced robotics, artificial intelligence (AI) and deep-learning.
“Unplanned asset downtime is a top issue for the oil and gas industry, and can cost operators millions of dollars,(…) Avitas Systems will help enhance the efficiency of inspections, and can help our customers and others avoid significant costs by reducing downtime and increasing safety.”
Avitas Systems is taking emerging technologies in different fields, like NVIDIA’s deep-learning servers designed for gaming, and merging them with industry standards that have been making huge leaps forward recently, such as remote-controlled robotics.
Their autonomous inspection technology allows for human-controlled robots to either fly, crawl, or swim around the inspection sites at any time. Besides being able to reach places that we can’t, drones can also withstand harsher environments and enable field experts to remotely inspect locations from anywhere in the world.
“Remote experts can study issues at a distance because we bring the location to them,(…) A team in Houston can, from their office, investigate a problem with a pipeline in the Gulf of Mexico, for example. And because our system highlights areas of concern, the experts don’t have to review hundreds of hours of video. Instead, they can focus on the most important footage and help solve problems faster.”
The startup complements its hardware with state-of-the-art software: 3D modeling tools that create targeted paths for autonomous inspections; Avitas’ predictive analytics that can recommend intelligent inspection scheduling and planning to detect changes over time and predict anomalies; the data from both manual and autonomous inspection is collected almost instantly and fused with machinery performance analytics and external data sources (such as the weather) to create an extensive report; and a deep-learning database which is continuously evolving with each inspection, creating increasingly accurate results.
“We are often working with site plans measured in square miles, (…) When someone sees a pressure drop in a particular pipe, actually finding that pipe, or a particular flange or valve, can be quite difficult just because of the scale. The ability to use 3D models to look back and forward in time is a powerful way to focus on points of interest.”
In the end, Avitas Systems claims that clients have significant inspection time and cost reductions, resulting in annual savings of up to 25%. All of this while decreasing human risk, improving accuracy, and extending equipment life. Just one and a half years old, the Boston-based startup already employs 25 people and has over 50 clients around the world, with strategic alliances with top-notch companies such as Bureau Veritas, Kraken Robotics, Limelight Networks, NVIDIA, and Waukesha-Pearce Industries.
Avitas Systems is fully owned and developed by GE Ventures, a strategic investor for the digital industrial landscape. It works a bit like our partner 9.5 Magnitude Ventures: besides long-term capital commitment, the company also offers access to GE’s top leaders and domain experts, their network of customers and partners, and their marketing and PR coverage.