Control Charting Issues Can Lead to Firefighting In {Business~Control Charting Performance Measures~X-bar and R Control Charting Issues That Can Lead to Firefighting~Firefighting In The Enterprise: Control Charting Issues May Be The Problem}

Tuesday, November 24th, 2009 at 1:50 pm

Organizations can have many problems because they are measuring the wrong things, which can lead to unbeneficial or detrimental behaviors. Almost every organization’s measurement and improvement systems need to be improved. Organizations need to avoid the basic red-yellow-green balanced scorecard measures, which, can lead to significant firefighting that does not benefit the enterprise as a whole.

To create a valid process capability statement data needs to be from a stable or “in control” process. Let’s have a sampling process so that if mulitple persons examine the same process their conclusions should be the same even if the process is considered “not in control”. With this statement, reference is made to how they sample from the process, not a chance occurrence. “Predictable” is often easier to understand than “in control”.

It is better to use ppm as a response for process capability and process performance indices, rather than Cp, Cpk, Pp, and Ppk, which can be very confusing and deceiving. A better practice is to use a probability plot to describe process capability/performance, since a probability plot offers more output flexibility and data understanding potential than process capability analyses that also provided Cp, Cpk, Pp, and Ppk outputs. When make a process capability/performance metric statement probability plots are also very useful, even though no specification exists. To get organizations out of the firefighting mode, probability plotting is not the most important issue relative to creating a no nonsense balanced scorecard measures system.

Corporate objectives, when implementing the Integrated Enterprise Excellence (IEE) measurement system, can be achieved. In the IEE methodology, there can become a measurement “pull” for the initiation of projects when a predictive metric does not produce a desirable improvement for the enterprise as a whole. To accomplish this goal, a measurement system is desired which is independent of how a sample collection system is developed.

30,000-foot-level balanced scorecard measures’ (control charting) primary purpose should be an overall view of customer experience. Assuming that there is consensus with this position, a couple questions need to be asked to determine if there is agreement as to what should be considered as a potential common cause and special cause input variable source in a 30,000-foot-level control chart. This is in contrast to the timely identification of an issue using a control chart to stop a manufacturing line for problem resolution because of an out of control signal, which is something that is typically conveyed in classes but does not often occur in the “real world”.

Let’s discuss a typical situation where process raw material is changed day by day. To be sure, consider that some raw material characteristic affects the process’ output. Should we consider raw material as a potential common cause variability source or a special cause variability source?

Raw material should be considered a source of common cause variability in most opinions. If we all agree and there is also agreement that control charting should provide information consistent with what we believe with respect to special and common cause identification, we will not use x-bar and R charts.

Why is this? With our current belief system, the fundamentals behind the creation of an X-bar and R chart can seem inconsistent. X-bar and R chart control limits are only a function of within subgroup variability. When considering x-bar and R control chart limits, variability between subgroups has no affect. The control limit calculations of an individuals control chart provides control limits that are a function of the variability between subgroups. The control chart upper and lower control chart limits might consider the variability between raw material lots; but this would not occur in an x-bar and R chart.

X-Bar and R Charts generated by statistical programs such as Minitab can show process capability, but we don’t think that is the way to go. Following X-bar and R charts can create a lot of fire fighting. X-bar and R charts are not used when making an IEE 30,000-foot-level assessment.

Believe it or not, some people find it hard to believe that the x-bar and R chart they learned about in their basic statistics class has issues. Several 30,000-foot-level articles can be found in the “On-line Resource Library” link at www.SmarterSolutions.com, which provides more details and shows an example, not only for a continuous response output but other outputs as well.

The volume, Integrated Enterprise Excellence Volume III – Improvement Project Execution: A Management and Black Belt Guide for Going beyond Lean Six Sigma and the Balanced Scorecard describes on how to create 30,000-foot-level charts for various situations and much more. You can also visit www.ieeblackbelt.com for more information.

Share and Enjoy: