Both Lean, and Six Sigma methods can be used to minimize waste and improve efficiency in business operations. In our blog post “Is your Shift Schedule Lean?” we addressed potential waste sources in shift schedules. This is a Lean approach to evaluating a shift schedule.
What about Six Sigma and Shift Schedules?
While Lean’s focus is on eliminating waste in the overall process under consideration, Six Sigma looks at minimizing variability in the process using a data-driven evaluation process. Root causes are identified and verified by actual data. Six Sigma methods can be mathematical, but they are very powerful too.
We can use Six Sigma methods to answer questions like: Is the shift schedule causing productivity to vary in our operation?
A common belief in shiftwork operations is that a particular shift (often believed to be night shift) has lower performance than the other shifts. Similarly, shift workers working on a particular day of the week (weekend days are a common target) may be believed to have lower performance. Or, perhaps the first shift worked after a long break is believed to be a low-performance day. Six Sigma methods can help us test those beliefs and separate the normal operating variability from real, shift-schedule, related problems.
Let’s look at an example of using Six Sigma methods to evaluate a shiftwork operation. Several years ago, Shiftwork Solutions worked with a consumer products company considering a schedule change from 24×5 (Monday-Friday, 8-hour shifts) to a 24×7 schedule. The average daily production volume, by shift, is shown in the table below:
It doesn’t take a Six Sigma Blackbelt to conclude that the third shift is producing less than the other two shifts. Total production is a performance measure, but without knowing the actual labor hours (or some other similar measure of effort invested), it is difficult to draw any conclusions beyond the obvious difference between shifts in the average production/shift.
Taking into account the direct labor hours used, the same production performance looks like this:
Now what? We see differences between the shifts, but we can’t tell if they are real differences or just a reflection of the noise in the system. The second shift is clearly more productive than the other two shifts.
As Donald Wheeler says (on page 30 of Understanding Variation): “While every data set contains noise, some data sets contain signals. Therefore, before you can detect a signal within a given data set, you must first filter out the noise.” Do we have noise here, or should we be evaluating how we are managing our shift work operations?
First, let’s look at the total production levels between shifts. We know that 3rd shift is clearly lower, but is the first shift performance actually lower than the second shift? If we perform an analysis of variance (ANOVA) between the 1st and 2nd shifts, we find that the answer, with a 5% significance level (the risk that our hypothesis is false, even though it appears to be true using ANOVA), is yes (F > Fcrit). Here’s the ANOVA table:
What about the difference in productivity? Is the first shift more productive than the third shift? Performing an ANOVA analysis on that data yields:
In this case, F<Fcrit, and we cannot conclude that there is a difference in performance between the first and third shifts at the 5% significance level.
What does all this tell us? The second shift is much more productive than the first and third shifts. The first and third shifts, statistically, are equally productive.
Questions that we need to investigate:
- Why is the second shift more productive than the first and third shifts?
- What explains the difference in overall production levels, given the productivity levels?
- Is it the shift schedule a contributing factor?
- How do we get the first and third shift up to the second shift’s productivity level?
Finally, back to our original question about the shift schedule: Is the shift schedule causing productivity to vary in our operation? At this point, we don’t know if it is the shift schedule, but we cannot rule it out. We have more work to do.
We can help you evaluate your shift work operations and help you solve the actual problems that are limiting your operational performance.
Call us today at (435) 200-5566.