How drone data is shaping the future of asset management
The use of drones has rapidly expanded throughout the asset management industry. However, while this evolving technology may have changed the way we inspect physical assets, has is it changed the way they we manage them?
The biggest issue with inspection methods of the past was surrounding the quality and quantity of the data collected. Inspections were often conducted manually, in challenging environments, and with restrictions around access. Assets often had to be shut down during these inspections, which becomes costly and time consuming. The data recorded was often limited to drawings done by hand or single images taken by cameras with notes attached.
At first, early adopters took to using drones to inspect their assets. It was quick and efficient, allowing photo’s or video to be captured of problematic areas. It negated the need for manned inspections at heights, and reduced the need to use EWP’s, scaffold or rope access.
However, while this has been a great new capability, has it been truly innovative? Has it fundamentally changed the way we make decisions and manage assets? In this capacity, the drone merely changed the method of conducting the inspection. These ‘spot check’ images were still simply incorporated into our existing reports and workflows.
These inspections and workflow methods have limited accuracy and are dependent on the person undertaking the inspection. They tend to be focused on identifiable defects and not a complete overview. If a defect wasn’t identified at the time, what implication might that have down the track? Also, consider when a new defect is identified. Could you go back and assess that specific area from earlier inspection reports? Well, no, because it was never previously recorded as the defect wasn’t present during the last inspection.
While these methods may have sufficed until now, how can we be expected to further optimise decisions, maintenance programmes and strategies without the full picture?
Now, imagine knowing your asset so well, that when needed, you could recall every inch of it down to the smallest detail. You know exactly what the condition is of any area, without a doubt. Imagine being able to access that level of knowledge from anywhere at any time.
This ability is our new reality. Through the creation of life like 3D digital “twins” an asset can be captured in its entirety, providing unprecedented levels of knowledge and insight across its lifecycle. An asset can be repeatedly captured time and time again to provide an effective way to monitor trends and make accurate predictions.
We are entering an era of complete asset transparency through the delivery of powerful drone data, and this is the breakthrough the industry has been waiting for.
We now have the ability to use a tool such as a drone, to capture vast amounts of information. However, where the value is created is the ability to then process that information into highly detailed and usable data with a few important features.
Firstly, the data is reliable and repeatable. The most powerful benefit of having consistent non-subjective data is the ability to “rewind” time and asses the entire assets condition from previous inspections, not just “spot” areas. This not only allows the manager to track and monitor how a defect has deteriorated over time, but also allows them to assess the root cause of a defect, or even compare it against previous rectification works.
Secondly the data must be accurate and detailed. While data can be captured in a relatively quick time by all types of drones; to be of any great value it must be “life like” in detail and accurate. Using specific drones designed for inspections, along with correct capture techniques, dimensionally accurate 3D models can be delivered with sub millimetre levels of detail. After all the idea is to be a digital equivalent or “twin” of the actual asset.
Thirdly, the data must be easy to use. This is where the use of dedicated 3D inspection platforms will deliver true insight. The ability to visualise high-resolution imagery is only just the start. The right tools are required to allow for true collaboration and cooperation. Managers are now able to digitally annotate of areas of concern and compare datasets over time. You can accurately measure distance, gradients and volume, and most importantly you can share the data with colleague’s, management and contractors.
These key features are enabling management teams to make fully informed decisions about their assets and create better strategies through unprecedented levels of knowledge.
The next big evolution to come out of all this data will be the application of artificial intelligence to automatically identify defects. “AI” in asset management is still in its infancy and technically known as ‘supervised learning’. Presently human intervention is still required to verify the correct identification of defects, and to correctly dismiss false detections.
To learn, the machines must be fed vast amounts of data through human input. This will take time as the more data the machine is fed, the smarter it becomes and the more accurately it will identify defects. The challenge however will be the amount of data needed to train a machine. The machine needs to learn every type of defect on every type of surface to correctly identify defects on a complete asset. For instance, to teach it to detect rust on steel could take anywhere up to one million images to gain acceptable levels of accuracy. And, this is only to identify a defect, not its’ severity.
We live in a digital age where we expect and demand information instantly. We depend on more and more knowledge to better base our decisions and strategies. The ability to virtually inspect an asset from anywhere at any time is an invaluable tool required for a world with rapidly evolving workflows, working habits and client expectations.
So when should you get involved with drone data? The simple answer is, yesterday. The sooner you start collecting valuable data, the sooner you will have a deeper understanding of your asset. Having this critical information at your fingertips will empower you to manage your asset portfolio effectively and with confidence.
Derek Feebrey is CEO of Trendspek.
A drone data provider specialising in solutions for the asset management industry.