Scaling Up? Don’t Leave Statistical Process Control (SPC) Behind

Overview

For over 30 years, the staff at PQ Systems has been talking to people practicing statistical process control (SPC). Despite the endless variety in SPC applications, from sandwich spreads and steel springs to train wheels, there are recurring issues that appear time and again. One of the most common challenges is how to grow an SPC deployment to large scale.

Today, computers and software are widely used for SPC, but the basic workflow for SPC workers is little changed since the days of paper and pencil: 1) measure something, 2) visualize it on a chart, 3) learn from and possibly react to signals seen on the chart. This works well for a limited number of metrics.

In modern work processes, from complicated production lines to multi-day hospital stays, the number of metrics that might be monitored using SPC is staggering. Even when being careful to select the vital few metrics and ignore the trivial many metrics, you might be faced with 300, 500, or even 1000 control charts! In your operation, is it practical to think about deploying this many control charts? For most people, unfortunately the answer will be no. It is just not practical. Who will look at all these charts? How many workers can be trained and then made available for this SPC work?

The problem here is not with the technique of SPC. Instead the problem is one of deploying the technique on a large scale. This presentation will describe the challenges of growing SPC to a large scale, and suggest approaches to meet these challenges.