35 years of spotting relationships in shop-floor data, built into an app.
I've spent 35 years in quality, Six Sigma, and Lean within automotive manufacturing, and one question comes up again and again: is this actually related to that? A tool change and a scrap rate, a temperature and a dimension, one process step and another downstream — spotting a real relationship early, and knowing how strong it actually is, is often the difference between fixing a problem and chasing a coincidence.
Checking that properly usually meant a spreadsheet, or software that lived on a desktop somewhere else — never quite at hand standing on the shop floor with a fresh set of readings. And with more than two variables in play, it's easy to eyeball the wrong pair and miss the relationship that actually mattered.
So I built the tool I wished I'd had, as a companion to CpK. Enter up to four columns of data — by typing, speaking, or scanning a photo — and see a correlation matrix across every pair at a glance, so the relationship worth a closer look finds you rather than the other way round.
From there, the app fits several curve types and picks the best one by adjusted R² rather than raw R², so a more complex curve only wins if it genuinely explains the data better, not just for having more parameters to bend with. Every result comes with its equation, R², adjusted R², and p-value, and a "Speak Data" option reads any column back to you for cross-checking against a physical sheet.
Everything runs entirely on your own device — nothing you enter ever leaves your phone, which matters in a manufacturing environment as much as it does anywhere else.
Questions, feedback, or just want to talk shop about correlation and curve fitting? I'd like to hear from you.
Email Me