In this lesson, we go over I-MR charts and what are the advantages in using them over other methods in Six Sigma. If you have any questions, feel free to leave a comment below and we’ll get back to you!
Individuals and moving range, or I-MR charts, are among the most useful in Six Sigma in the science of quality management. The chart consists of individual data in the order they were generated in the upper chart and differences between successive data, or moving ranges, in the lower chart. The fundamental job of the chart is to find evidence of process instability. If the chart shows no rule violations, as is the case here, there’s a good chance that the process is stable and predictable or in control. In other words, there’s a good chance that the process will work again.
Tomorrow, if the process does show rule violations, as this other example does, then the process is not stable and predictable. It’s out of control and may or may not work the next time you try it. Here’s an example of applying I-MR charts to the time it takes to collect accounts receivable. It applies equally well to the dimensions of a part, or to electrical resistance, or of many other things. This chart shows no rule violation. So it’s fair to assume that unless something disturbs the process, we can predict where new data will fall.
In addition to detecting process instability, I-MR charts can be used to compare groups of data. Much as you do in a t-test or ANOVA. They can also be used to evaluate the accuracy and precision of measurement systems. There are some errors that are commonly taught. One is that the data need to be normally distributed for the charts to work. This is incorrect. Walter Shewhart, who invented the charts, made no assumption of normality, relying instead on the Chebyshev inequality, which works for all distributions. The charts do not require any particular distribution of data. Another is the users should use X bar and R charts instead to ensure normality.
Now X bar and R charts are wonderful charts, arguably stronger than I-MR. But they require more skill and are frequently done incorrectly with disastrous results. There is no requirement for normality, so there’s no point in trying to get it by using X bar and R charts. Third, it is often thought that the moving range chart must be in control before the individuals chart can be interpreted. This is not so and out of limit condition and the moving range chart, simply means that there’s been a shift in the data that is too large to easily explain by ordinary variation. It means no more and no less. Technically, we say that the average moving range is robust to outliers.
Used properly, by I-MR charts provide an abundance of information with a minimum of computation. They are one of the most insightful tools you can learn.