In this chapter, the author talks about how to detect the signal of assignable cause from a process behaviour chart. He first presents the characteristics of a predictable process, and then gives three major detection rules for an unpredictable process. However, my concern is how can we communicate them to the audience since it is hard to explain (for now since I don’t understand the rationale behind it yet) the logic behind the XmR chart (Is it common in the industry?).
In a predictable process, approximately more than 85% of the values would fall within 50% of the limits (upper and lower). There will be a relatively fewer number of values, say 15%, near the limits. Provided that, we have the following three detection rules for an assignable cause:
- Points Outside the Limits: A single point outside the computed limits on either the X Chart or the mR chart, should be interpreted as an indication of the presence of an assignable cause that has a dominant effect.
- Runs About the Central Line: More than 8 successive values, all on the same side of the central line may be interpreted as an indication of the presence of an assignable cause that has a weak but sustained effect.
- Runs Near the Limits: 3 out of 4 successive values all within the upper 25% of the region between the limits, or all within the lower 25% of the region between the limits, may be interpreted as an indication of the presence of an assignable cause which has a moderate but sustained effect.
The goal of monitoring the process behaviour chart is to turn the unpredictable process into a predictable process by fixing the assignable cause in the system.
Process behaviour chart only handles one types of continual improvements, that is, taking action on the process. There are other continual improvements, including setting up plans, goals and target (taking action on the outcomes), and actions to align goals and process together.