← Cases
ChinaManufacturing2026 – ongoing

Forvia

AI use cases at the Yancheng seat-track plant, a WEF Lighthouse Factory

Forvia's Yancheng plant reported a 46% die change gain and a 7.5-month payback on its AI use cases

WorkingCompany's own claimOfficial ↗

Problem

The Yancheng plant makes about 30 million sets of automotive seat tracks a year. It needed to keep quality high, reduce destructive testing and shorten die changes.

Approach

The plant ran more than 40 AI use cases using machine learning, deep learning and generative AI. They included a multimodal quality control system, a big data model that replaced or optimized destructive testing, and generative AI for die change optimization. The World Economic Forum named the plant a Lighthouse Factory in January 2026.

Results

46%improvement in die change optimization with generative AI, per ForviaCompany's own claim[1]
over 90%accuracy of the big data model that replaced or optimized destructive testing, per ForviaCompany's own claim[1]
7.5 monthstime to return on investment for the AI solutions, per ForviaCompany's own claim[1]

Takeaway

The figures come from Forvia's own announcement and the company did not say how it measured them.

Categories

Business function
Production and quality
Innovation type
Process innovation
Evidence
Company's own claim

Sources

Source summaries

[1]Forvia19 Jan 2026

Forvia's press release reports the plant's recognition as a Lighthouse Factory and lists its AI work, including more than 40 use cases. It gives a 46% gain in die change optimization, over 90% accuracy for the testing model and a 7.5-month payback.

Full details are in the original sources.

Added 1 Nov 2026 · Updated 1 Nov 2026
Open case page ↗