From 30% to 80% yield
across 12 production lines.
How a global Tier-1 automotive supplier turned 700+ process variables into four decisions — and a 2.7× first-year ROI.
The challenge
Yield stuck low, and no one could say why.
On a safety-critical process, first-pass yield sat far below target. Engineers had the data — hundreds of process signals per part — but classic studies could only test a few factors at a time, and the real driver turned out to be an interaction between parameters no one had paired. Every week the line kept running meant more scrap and more rework.
What Jemba did
700+ variables, ranked to the 10 that mattered.
- Identified the 10 most critical factors from 700+ variables.
- Replaced months of trial-and-error with a data-driven method.
- Concrete gains validated by the plant’s own process engineers.
Download the complete analysis.
Inside the PDF
- How ML identified the 10 most critical process factors from 700+ variables.
- The data-driven approach that replaced months of trial-and-error.
- Concrete production improvements validated by process engineers.
Download the full case study
Get the complete analysis including methodology, data insights and results.
🔒 We respect your privacy. No spam, ever.
“We analysed the data with our usual Six Sigma tools, and Jemba confirmed our analysis with a broader, faster view. Excellent work.”
Get the same answer for your line.
Book a 15-minute demo or start a free trial. 48-hour setup, no data scientist required.
