All the latest news from the industry weekly compiled by the editorial team for you free of charge.
This eMail is already registered.
An unexpected error occured.
Please accept our Terms of Use.
Registration successful.

Aniruddha Banerjee

Co-founder

SwitchOn

1 Rating

AUTOMATION OF QUALITY INSPECTION PROCESS Applying Artificial Intelligence is now technically and commercially feasible

May 13, 2022

…says Aniruddha Banerjee, Co-founder, SwitchOn, in this interview with Vidisha Sonone. He talks about how the start-up is driven to automate quality inspection processes for the manufacturing industry and how new age technologies can maximise productivity and reduce costs. Excerpts...

How did SwitchOn’s journey begin?

Most of the manufacturing processes are completely automated with machines and robotic arms. However, the quality inspection processes in most plants are predominantly manual, with lines of manual workforce picking up products and identifying defects. Critical quality functions cannot be slow, error-prone, and subjective. This drives us towards automation.

What made you think of AI for your product DeepInspect?

We realised there were some fundamental flaws with the conventional inspection systems. While the conventional systems were good for presence-absence or colour-detection applications, complex surface inspection on reflective, metallic, plastic surfaces was still not possible. There was a need for a huge amount of data before the AI systems could reach the 95%+ accuracy mark. DeepInspect, the AI-based solution works only with <200 good pictures to get almost 99.9%+ accuracy within weeks of deployment.

Which areas of the manufacturing industry can benefit from new age technologies like AI, AR, VR, and Machine Learning?

Technologies, such as AI and Machine Learning, have finally matured and can be both technically and commercially applied. We fundamentally believe that the plant operations and supply chain processes of manufacturing facilities have been the slowest adopters of technology & hence are ripe for large-scale benefits.

Do you think the manufacturing industry is too quaint and rigid to accept digitalisation?

Most of the manufacturing applications are quite safety-critical, relying on high-quality standards. Previously, digital technologies were not mature enough to handle complexity, accuracy, and stability; hence, the adoption was reasonably slow. I feel that this is a great opportunity for state-of-art technology to help take legacy manufacturing to the next level.

What is next for SwitchOn?

We have just started scaling our solution across large enterprises and looking to double down on our traction in both automotive and consumer goods. We are also building a product that is scaling 2X quarter-over-quarter, and we are looking to expand our team by 3X in the next few months.

What problem are you aiming to solve?

We help precision manufacturing industries drastically reduce their manufacturing defects through the use of a one-of-a-kind field-trainable AI system. Automotive and CPG companies globally employ 25%+ of their workforce on inspecting product quality and yet face 500+ customer complaints per year per site leading to $120M+ in losses. We are aiming to eliminate defects in manufacturing lines.

Companies related to this article
Related articles