How To Automate Industrial Inspection — Part 1
In this blog, the first instalment of our “How to Automate Industrial Inspection” blog series, we will be covering different technologies and processes that will help you to transform your business operations.
Asset operators are seeing a dramatic increase in inspection data from drones and mobile IoT devices that are demanding unsustainable levels of human effort to analyse and report. Many see Artificial Intelligence (AI) as the engine that will drive the automation of these processes, but they are not clear how to harness it. There are many AI algorithms available that attempt to identify some defects in industrial assets, but these are typically standalone solutions. Asset operators with thousands of assets and millions of data points require an integrated application that uses AI as part of an automated workflow, connecting defects to assets and transforming data into actionable information.
Inspection2 software uses AI to automatically analyse inspection images saving up to 70% of engineering time compared to traditional methods. A powerful AI engine that really drives OPEX savings in inspection analysis demands an application that is specific to industrial assets and solves the problem in a more focused way than off the shelf AI tools.
To meet that demand we developed Asset Mechanics.
What is Asset Mechanics?
Inspection2 Asset Mechanics is a framework within which we combine AI tools with other automation technologies to more quickly and accurately analyse digital inspection images of industrial assets. It breaks down an asset into its component parts to identify defects specific to those components.
Let's take a look at the example below of an electrical distribution tower
- The asset ID and type is registered in the software — each asset type has a specific Asset Mechanics configuration including common components, defects and analysis rules
- The software breaks the asset down into the main components that it identifies in the imagery — this enables the defect identification
- Inspection2 AI tools analyse the images for defects associated with each specific component — e.g. in a glass insulator we look for cracks but not corrosion.
- In the example, contamination is identified on the insulator
Inspection2 Asset Mechanics delivers a significant increase in the accuracy and speed of analysis saving thousands of hours of engineering time. It is also a critical factor in delivering full automation of inspection analysis and reporting which we will explore in future blogs of the “How to Automate Industrial Inspection” blog series.
About Inspection2
Our mission is to automate industrial inspection delivered through our integrated application. We manage image data from IoT sensors, drones and other devices; analyses and reports on anomalies using 3D digital twin, AI and process robotics technologies. Our software transforms huge volumes of data into actionable information which is easily shared with enterprise applications, unleashing powerful management information and the potential of automation.
