New approaches in Statistical Process Control (SPC)

statistics
programming
Published

October 15, 2021

Statistical Process Control has been out there at least for a century but real life applications continue to be difficult to implement. A good mixture of pragmatism and rigor is required to come up with a tool that can be easy to develop and maintain, is able to consume data in a continuous manner and provides realistic advice to effectively control production.

Many rules have been developed since the creation of the technique but most come from an era where the control was visual (e.g. 10 points above the center line). Nowadays implementation of SPC in software packages makes it possible to utilize more advanced and flexible algorithms and to see the outputs in real time.

One of these more recent approaches is Change Point Analysis which is claimed to be “capable of detecting subtle changes missed by control charts”. To go further we’re sharing here an interesting paper on the topic: Change Point Analysis: A Powerful New Tool For Detecting Changes.