By Krishna Iyengar
The auto component industry is one of the key sectors where Industry 4.0 has a lot of scope for implementation. Artificial Intelligence, Machine Learning, Deep Learning technologies can change the auto component industry as we know it currently. Also, implanting these Industry 4.0 technologies will significantly reduce the level of errors in the manufacturing process.
Expensive post-sales product recalls, warranty claims, high volumes of scrap generation, reduction in yields, project deadlines getting missed – all lead to both financial as well as brand-reputation losses. These challenges are, many a time, the result of insufficient inspection during the production processes; particularly crucial for the automotive industry that is constantly changing with customer demands.
The Indian auto components industry is seeing healthy growth with the country emerging as a global hub for auto component sourcing and the industry already exports over 25% of its production annually. To keep the momentum of high growth, manufacturers should leverage automated Quality Control which will allow them greater efficiency as well as an increase in productivity. Measuring the quality and detecting any non-conformance to standards are integral parts of the Quality Control process. The only way to achieve high levels of quality is by ensuring 100% inspection, 100% of the time.
Without automated quality control, organisations resort to manual quality inspection and that comes with a whole host of challenges.
Challenges with manual quality inspections impact productivity and brand reputation
Inspection in the production processes done manually can be subjective, biased and solely dependent on the perception and experience of the Quality Inspectors, leading to inconsistency in making decisions on the product or component quality. It can become a very expensive option, especially where the production volumes are very high, and manual inspection causes production bottlenecks, bringing down efficiency and productivity.
With QC personnel having to adapt to unique as well as changing quality requirements, fatigue sets in easily and with defective components getting missed and being passed through as good cannot be avoided. Also, working on large volumes of products day after day can lead to health issues and ergonomic constraints too, among QC personnel. Detecting defects in different environments or lighting situations, 24X7 can also be challenging, eating too much into the time of domain experts, who have to create complicated models factoring in various features.
Some organisations have adopted computer vision to help with inspection, but because that still relies on human intervention for decision making, it is not cost-efficient. With the actual quantum of defects that can be detected by vanilla machine vision, high false positives and lower precision of defect detection are commonplace. To add to this, continuous re-programming of the inspection equipment makes it an inflexible proposition, to adapt to changes in the products or components.
New-age technologies drive efficiency, accuracy and speed in quality control
Today, AI/ML technologies are taking the automotive industry to the next level of efficiency. Deep learning is an AI function that mimics the workings of the human brain in processing data, but with the speed and accuracy of computerised systems. Deep-Learning-driven technologies enable 100% inspection of the production components, even while discovering hidden patterns in the data.
Automakers can now control quality on the shop floor by leveraging AI to create competitive inspection standards and to make the QC function very effective at a much faster rate. An AI-powered automated defect detection system, because of its speed and because it can be done 24×7, leads to increased production and higher accuracy.
One of the other most significant plusses of this technology is that it frees up human labour from routine and lower cognitive tasks, and Quality Inspectors can get trained for performing jobs that require higher-order skillsets; like using their expertise on the subject to help the organisation refine the product and the QC processes.
Innovation with AI/ML technologies provides several benefits. It reduces defective pieces to slip through, detects numerous featured areas on the product within a few seconds thereby increasing speed of inspection and is capable of adapting to the changes in the product. Efficiencies across the processes are improved significantly leading to several business benefits. AI-based inspection can also help handle complicated QC processes, incorporating methods like X-ray or penetrative dye inspection – taking precision of inspection to higher levels.
Despite several advantages of using AI/ML and DL technologies for Quality Inspection, the manufacturing industry lags behind in their adoption. Companies are hesitant to invest in upgrading to AI-based infrastructure. To adopt AI-based transformation, the change should be led by the top management in any organisation to put the necessary infrastructure, policies and governance in place. It makes business sense for manufacturing companies to rely on an AI-based Solutions and Implementation partner to support the parts inspection processes that will drive better business outcomes.
(Krishna Iyengar is Co-founder and CTO of Jidoka Technologies.)
(Disclaimer: The views expressed in the article above are those of the author’s and do not necessarily represent or reflect the views of Autofintechs.com. Unless otherwise noted, the author is writing in his/her personal capacity. They are not intended and should not be thought to represent official ideas, attitudes, or policies of any agency or institution.)