At the forefront of the latest and most exciting advancements in this technology is AI for industrial inspection (AI4II), which is set to transform the way manufacturers carry out the process of visual examination of high value components.
Artificial intelligence is beginning to fulfil its potential as the next breakthrough technology in driving productivity for high value manufacturers.
Its development has already delivered smarter and more efficient ways of working, as well as pushing down production costs.
Inspections prevent failures and faults
Manual inspection involves the physical assessment of machinery to identify and prevent system failures and faults, which can be time-consuming and costly to manufacturers.
AI4II has now paved the way for a digital solution which combats the challenges that high value manufactures face.
The challenges of industrial inspection
High value manufacturing is a core component of the UK economy, and the need to support high integrity engineering sectors has placed huge pressure on the rapid delivery of reliable and high quality evaluation methods to inspect components and ensure they are fit for function.
Manufacturers working in the aerospace, automotive and construction sectors, will no doubt be aware of the challenges faced when carrying out an industrial inspection and the need to speed up the process to cope with growing demand for high quality machinery that stands the test of time.
It takes years to build the necessary inspection skills
Visual inspections are routinely carried out to determine whether a structure, product, component or process meets the specification requirements.
Such inspections are usually completed by a trained individual who has knowledge and experience to visually identify faults as well as non-conformant quality and performance.
Time restricted inspections
Each engineer carrying out the process of a visual inspection will need years of experience to understand the intricacies of system failures and faults. Staff will be restricted to a number of hours to carry out this work and will also only be able to deliver a finite number of inspections in a given time.
What’s more, manual inspections are also open to a variety of challenges, from physically being able to access a particular system or part within a larger structure, to human error and recurring labour costs.
Adding innovation and technology to the process
AI provides an opportunity to introduce innovation and new technology to the visual industrial inspection process, offering an automated, highly reliable, digital solution to these sector-specific challenges.
AI4II has been designed to discuss and address some of these challenges and help organisations understand how they can overcome them.
Training AI through deep learning
The Centre for Modelling & Simulation (CFMS), a not-for-profit specialist in digital engineering capability, has produced three types of demonstrators which combine computer vision and artificial intelligence technologies to automate the manual inspection process and counteract some of the challenges that high value manufacturers face.
As part of the automated inspection process, the inspected component is photographed at regular intervals to produce a series of images. These images are manually annotated for features of interest (e.g. objects and defects) and then used to train the AI model to learn identify these features of interest automatically.
The AI learning curve
To enable this process to happen, AI4II uses deep learning technology, consisting of neural networks inspired by the biological neural networks used in the human brain.
Once the AI has learned how to identify defects, it will automatically process the subsequent photographs of the inspected component to speed up processes of identification for the next inspection e.g. flagging to an engineer to repair or replace a particular component.
Four benefits of AI for industrial inspection
AI technology can review feedback from millions of assets and process thousands of data sets and images in a significantly shorter period of time, with a much lower margin of error than using a human workforce. This works particularly well for manufacturers that need to carry out fast and routine inspections of assets
- Cost-savingBy reducing the need for an engineer to carry out a physical visit to undertake an inspection redirecting the focus of their highly-skilled workforces to repairing rather than troubleshooting defects and carrying out lengthy inspections
- Health and safetyAutonomous inspection provides a more efficient and streamlined solution for inspecting these potentially hazardous environments
- Data curationManufacturers can use the data harvested from AI4II to inform future technologies and materials, therefore informing future business decisions and in turn leading to the identification of further opportunities for the adoption of AI technologies in order to automate business processes
AIs value to manufacturers
It is without a doubt that AI will be a vital tool for high value manufacturers in the future, providing an opportunity to introduce innovation and new technology to the visual industrial inspection process.
By developing the AI4II demonstrators, CFMS is offering an automated, highly reliable, digital solution to sector specific challenges such as accessibility, human error and increasing labour costs.
Case study: Airbus A400M wing inspection
Carrying out a manual inspection of an Airbus A400 wing presents the manufacturer with a host of costly and time-consuming challenges. The inside of an aircraft wing, which needs to be routinely inspected either during manufacturing assembly stages or aircraft maintenance cycles, is a hard-to-access confined space which makes inspection physically challenging for an engineer to undertake.
With AI4II technology, a smart phone can be used to capture video footage of the inside of an A400M wing box. The footage from the smartphone camera can then be used to detect imperfections (such as foreign object damage, corrosion, cracks, dents and scratches) and used to train the AI to learn to detect the defects.
Once the defects have been identified, an engineer can be deployed to help fix the component. In the future, organisations could deploy a crawling robot, equipped with a camera and AI module to automate inspection in these cases.
You can watch the video of the wing inspection using AI technology.
You can follow Innovate UK on:
- Innovate UK Twitter @innovateuk
- You can find out more about the Industrial Strategy Challenge Fund here
- For more Innovate UK videos subscribe to our YouTube channel here
- Sign up for email notifications on funding, connections & support opportunities
- Follow Innovate UK on Facebook
Connect with Innovate UK on Linkedin