CAPABILITY ANALYSIS


DETERMINING PROCESS CAPABILITY

A process is said to be operating in statistical control when the only source of variation is due to common causes. This state, however, does not imply that the parts or products produced by that process are acceptable, that is, that the parts meet the required specification. In fact, there are four basic possibilities for the relationship of statistical control and what is generally referred to as capability, which is the fact that a significant percentage of parts produced (99.73 “99.97%) will be within the designated specification. These four possibilities were shown in Figure 12.1.

One of the most important concepts to note as related to these relationships is that a process must be in statistical control before capability may be established. It is statistical control that allows us to determine that the output of a process is predictable; without such confidence, any assessment of the potential of that process to produce parts within specification is unfounded.

CONTROL CHARTS AND SPECIFICATIONS

As we already have stated several times, specifications for a process cannot be compared with the control limits for that process. The reason for this is that specifications deal with single parts in relation to a specification, whereas the control limit is a calculated number based on averages of the samples. In fact, earlier we talked about apples and oranges when one attempts such a comparison. Therefore, the two following points must be restated:

  1. To compare specifications, which are written for individual units, against limits from control charts, which are a result of groups of units being measured, we must convert the control limit values.

  2. To have a process under control is not equivalent to having a process that produces parts within specification. Figures 12.2 through 12.5, for example, reflect in-control processes; however, only two are also capable. In fact, only one of the distributions shown is desirable!




Six Sigma and Beyond. Statistical Process Control (Vol. 4)
Six Sigma and Beyond: Statistical Process Control, Volume IV
ISBN: 1574443135
EAN: 2147483647
Year: 2003
Pages: 181
Authors: D.H. Stamatis

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net