FEATURE Machine Vision y
Improving quality in the
automotive supply chain Each year, approximately a million vehicles are recalled in the UK for safety repairs. These figures do not include cars that just malfunction and need repair. Whether the problem is engines, HVAC systems, gear components, tyres, software, airbags or steering wheels, faults in vehicles pose a serious safety risk. Harel Boren, CEO of Inspekto, discusses what can be done to improve quality and reduce costs in the automotive supply chain.
T
he automotive supply chain is incredibly complex — structured in tiers from raw materials to larger structures, to the original equipment
manufacturer (OEM) building the final product. The average vehicle has around 30,000 individual parts, sourced from a range of suppliers, which will be transported between tiers — sometimes halfway around the globe. All this means that assuring and monitoring quality across this vast supply chain has historically been a tricky process. In addition, Automotive OEMs continuously manage the quality of products their suppliers, and their supplier’s suppliers, provide. During production and just before shipping, manufacturers perform visual quality assurance (QA) so that any defective products are not sent on to the next tier. As the dire statistics show, these huge efforts generate sub-optimal results for OEMs across the globe.
The factory has eyes
However, the traditional visual quality assurance ecosystem is not fit for this purpose. Random checks deliver little more than nothing. The next best thing is human inspection – which fails to recognise defects in 25 per cent of products inspected, not to mention the tedious work and the cost of labour that is taken away from actual production. The last resort would be machine vision
inspection. However, because machine vision inspection solutions are so expensive, a minimum of £70,000 and sometimes in excess of £150,000 for each project deployed at one point on a single line, automotive manufacturers are installing them only at major junctions. Even so, these solutions are prone to fail with any change to the production line. Whether it is lighting, handling or the product itself, any change will require costly adaptation of the machine vision solution, along with a long and nerve wrecking wait. In addition, because machine vision
inspection traditionally relied on a systems 30 June 2020 | Automation
integrator to hard-engineer a solution onto the line, there are significant wait times in proof of concepts, commissioning and delivery of components. Once it is installed, a lengthy training process is required to teach the system all possible defects that could occur, followed by iterative optimisation, validation and commissioning. This too, deters manufacturers from using machine vision.
Autonomous is the answer
As of November 2018, this no longer applies. The launch of Autonomous Machine Vision (AMV) as a new category in machine vision, introduced affordable inspection systems that can be readily installed by the company’s own personnel. This means that automotive manufacturers can install visual QA at any point they want to address on the production line and do so as soon as the need comes up.
Installing visual QA at every required point on the line means that any quality issues that occur in between major junctions are identified and defective products can be
removed from the line. This reduces wasted energy and labour spent on a product, which will inevitably be scrapped. On top of that — if your buyer, the next tier manufacturer, claims that your business sent them a faulty product, you will be able to prove that it left your premises undamaged. In addition, AMV enables plant optimisation. Using TRACKS, the archiving and traceability application that can be installed on every INSPEKTO S70, manufacturers can trace exactly where a defect was introduced and what machine caused it, so that they can optimise their process to reduce the risk of defects being introduced in the future. It feels as though every day, another vehicle is being recalled. However, with the introduction of Autonomous Machine Vision, recalls will decline.
CONTACT:
Inspekto Web:
www.inspekto.com
automationmagazine.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44
