Withings Takes Efficiency to the Next Level with Instrumental

Withings teams up with Instrumental to speed through development and delight customers with their new sleep tracker.

Withings understands the importance of efficiency. The company planned and executed a one-day DVT, the fastest we’ve ever seen, and runs an exceptionally organized factory. Even so, they reached out to Instrumental to bring their efficiency to the next level. During DVT and the mass production run of Sleep, the Withings team used Instrumental’s intelligent defect discovery and monitoring tools to protect the new product’s quality. With visibility into their assembly line and the ability to catch issues they previously didn’t know existed, Withings was able to streamline defect discovery and failure analysis. They then moved into mass production confidently and ramped smoothly, with fewer defects.

Sleep, an IOT sleep tracker, gives users insights into their sleep cycles, including heart rate tracking and snore detection. It is designed to be simple and easy to use, and automatically syncs to Withings’ health app via WiFi.

The Problem

Sleep would not only be Withings’ first sleep tracker, but also the first product that the Withings team built after being acquired by Nokia. The pressure was on to make sure every unit met the high standard of quality that their customers had come to expect. The only visibility they had into the assembly line came from sending engineers to the factory and hoping they would be in the right place at the right time to catch issues, which wasn’t an effective way to prevent defective units from shipping to customers. Today’s failure analysis process typically involves disassembling physical units in or shipped from the factory, which is often slow and inefficient, so even when Withings did discover defects, it was difficult to figure out the root cause.

Withings was about to enter DVT when they engaged Instrumental. The industry standard DVT takes between five and twenty business days, but Withings planned to do their entire DVT build in one day. Powering through DVT speeds up time to production, but leaves the team with very little time to iterate if anything goes wrong. They were worried about unforeseen problems arising during mass production.

The image on the left shows reversed polarity (black and red wires are in the wrong positions) that Instrumental Dev helped detect; on the right is a normal unit.

As they moved into mass production, the Withings team leveraged Instrumental’s machine learning capabilities to help them find issues without needing to provide “golden units.” For example, Instrumental Dev identified that on several units, the two wires connecting the pump to the PCB were inverted. This recurring issue could not be caught by functional tests, and if these units had shipped to customers, it could have resulted in latent field failures, damaging product and brand perception. Another issue Instrumental Dev’s tools helped Withings catch was with the sleep tracker’s microphone. By using Instrumental Dev’s technology to run an anomaly search on the microphone region, Withings engineers discovered that a late tooling change caused a part to interfere with and crush the microphone. They then used Instrumental Dev’s Monitor feature to keep track of that region, and other problematic regions, to ensure consistent quality throughout mass production.

“In past builds we let some defects through NPI and fixed it in MP for speed, but with Instrumental we were able to detect and fix problems a lot faster in NPI, so we had a much smoother MP build.”


The Results

The Withings team implemented Instrumental Dev at a key point in their new product’s development and used it to inspect more than 10,000 units, with tests set up to monitor 12 regions. By automating defect discovery and streamlining failure analysis, Withings was able to speed through DVT and work through mass production headache-free. Withings’ team discovered issues they otherwise wouldn’t have, and the data and analysis tools housed in Instrumental’s software empowered them to correct those issues before they turned into bigger problems.

Thanks to Instrumental Dev, Withings’ team was able to uphold the high quality standard Withings’ and Nokia’s customers were accustomed to, even with a brand new product.

Interested in learning more?

Subscribe to our newsletter to stay informed about new features, case studies, and AI trends in manufacturing.