Organizations can have many problems because they are measuring the wrong things, which can lead to unbeneficial or detrimental behaviors.  Most businesses can use improvement in their measurement and improvement systems.  Red-yellow-green balanced scorecard measures, should be avoided because they can lead to firefighting measures that do not benefit the overall enterprise.

Data should be from a stable process, or “in control” process, to create a valid process capability statement.  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 a better term 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.  Probability plots are also very useful to make a process capability/performance metric statement even though there is no specification.  To be sure, probability plotting isn’t the critical issue relative to implementing a practical balanced scorecard measures system to get an organization out of the firefighting mode.

Corporate objectives are achieved in the Integrated Enterprise Excellence (IEE) measurement system.  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, a measurement system is needed which is independent upon how someone decided to create a sample collection system.

The primary purpose of a 30,000-foot-level control charting 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.  Also, consider that some raw material characteristic does affect the process’ output.  Should raw material be considered as a potential common cause variability source or a special cause variability source?

Common cause variability should apply to raw material, as most will agree.  If there is agreement on this 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?  The underlying principals supporting the X-Bar and R chart may seem inconsistant with our preconceived ideas. X-bar and R chart control limits are only a function of within subgroup variability. Variability between subgroups has no affect on x-bar and R control chart limits.   The control limit calculations of an individuals control chart provides control limits that are a function of the variability between subgroups.  X-Bar and R charts do not consider the variability between raw material lots as the individuals control chart upper and lower control chart limits would consider.

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.  Blindly following X-bar and R charts can lead to a lot of fire fighting.   X-bar and R charts are not used when making an IEE 30,000-foot-level assessment.

It can seem counter-intuitive for some to accept that the X-Bar and R chart taught in a statistics class has problematic 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.

Popularity: unranked [?]