5.5 Basic Lean Tools: Histogram

While we used a simple check sheet as our example, a complete sheet might identify more than one problem area. That data can be taken from the check sheet and put into a histogram chart to help visualize the intensity of an element as part of the improvement process. A bar graph depicts the collected data into categories with separate bars.

Figure 5.6 presents a typical histogram with each category based upon a number. The user determines the breaks in classes, which should be based upon some logic for the function or process being measured. For example, the histogram below may be the number of pallets moved each shift.

The histogram’s most basic form visually depicts the results of a tally sheet to allow the user to identify critical points quickly. The shape of the histogram can show if a process has a normal or skewed distribution and can determine if the data is bimodal. It also easily identifies outliers at either end or if one area is vastly overrepresented in the data.

Figure 5.7, Pallets Moved (by day of the week), highlights the specific benefits of using a histogram. Let’s assume that your warehouse workers must load and unload pallets on and off trucks each day. You count and post the average number of pallets moved each day. However, there is a clear outlier. For some reason, Wednesday is significantly different from the other days of the week. While the histogram itself does not identify the root cause, it quickly highlights an area for further examination.

Further examination may find that more trucks arrive on Wednesday due to shipping patterns or more workers on that day of the week. Whatever the cause, the histogram helps identify opportunities to improve the process.

The histograms displayed here are just a starting point in showing their usefulness as a tool. An additional enhancement to a histogram is adding Upper or Lower Specification Limits (USL/LSL). While a complete discussion of this enhancement is beyond this introductory chapter, you should know it exists.

The goal is to see if the process is “in control” or “out of control” during operations. In short, if the outliers are too far left or right of the mean or median in the histogram, they are beyond the USL/LSL, and the process is out of specs. This feature is another helpful view of the business process that the histogram can provide.

As a side note, this is also the origin of the term “Six Sigma” as a control process. The LSL/USL usually are three standard deviations from the norm/mean/median. Therefore, three up and three down equal a total of six, and this can highlight if a process is out of control—hence “Six Sigma”!