## STATISTICS: YOU ASK. WE ANSWER.

2017-05-23 - **What do you want to know about statistics?**

*03 June 2015: Stephan Conrad*

Since the third edition of the AIAG Core Tool MSA 2002 manual has been published in 2002, the subject of ndc keeps haunting the world. It almost seemed like this statistic will continue to have a ghostlike, unremarkable presence and might even disappear slowly but surely. But then came the fourth edition of this manual in 2010 and the ndc gained in importance once again.

By now, people have had enough time to gather sufficient experience to apply this statistic; however, most of them have rather had a bad experience. This is the reason why we will have a closer look at whether the

”number of distinct categories (ndc)“ makes sense or not.

Mittlerweile finden sich in jedem Seminar zur MSA mehrere Teilnehmer, die nach Audits vor diesem Problem stehen, und immer öfters erhalten wir Beratungsanfragen zu diesem Thema. Da sich weder an den Eigenschaften des Messsystems noch an der zugrundeliegenden Merkmalstoleranz etwas geändert hatte, war oft unklar, warum dieses Messsystem nun nicht mehr geeignet sein sollte und was zur Optimierung beitragen könnte. Deshalb sollten wir dem ndc nun endlich etwas genauer auf den Zahn fühlen.

The AIAG MSA manual defines the ndc as the number of categories of measured values that can be reliably distinguished. To put it simple, you may count how many times the gauge repeatability and reproducibility GRR fits into the actual process variation.

You always truncate the ndc, unless it is less than 1. If it is less than 1, you have to round it up. The factor of 1.41 (=√2) does not refer to the 97% confidence interval as described in the MSA manual. It is calculated from the variation ratios given in the ISO plot...