C Chart

Quick reference

• 00:04 Hi, I'm Ray Sheen.
• 00:06 Let's dig deeper into the attribute data control chart called the C
• 00:09 chart. >> The C chart gets its
• 00:12 name because it's a chart of counts.
• 00:15 The data being charted is a count of the process attribute,
• 00:18 which is almost defects, in the sample or subgroup.
• 00:22 The C chart is based upon a constant-size sample or subgroup, so
• 00:25 that is why there's no need to normalize the count.
• 00:28 The C chart is often used for tracking categories that occur regularly.
• 00:33 For instance, accidents in the workplace that occur each month, or
• 00:36 unplanned maintenance on equipment each month, the sample size being a month.
• 00:41 Another example is customer complaints each week, the sample size being a week.
• 00:45 Notice that the data is counts, so the data values will always be whole numbers.
• 00:51 The C chart is easy to understand and
• 00:53 it's simple to create, as we'll see when we go to the slides with the calculations.
• 00:58 Just determine the category you wanna count, set the sample size,
• 01:01 and start counting.
• 01:03 Let's look at an example of a C chart.
• 01:06 Of course, it has the standard control chart elements of a mean and
• 01:09 control limits.
• 01:10 As you can see in this case, the Y value is the number of counts.
• 01:14 On this example, we recognize a special cause has occurred at point number 19.
• 01:19 There have been 9 data points in a row above the mean.
• 01:23 The operator should have stopped the process at that point to investigate
• 01:26 what was happening.
• 01:28 Of course that is a process like the receipt of customer complaints,
• 01:31 you can't stop and refuse to take any more complaints.
• 01:35 But management should definitely investigate why complaints have been
• 01:38 consistently higher than average.
• 01:40 One other point about this chart,
• 01:42 the lower control limit can never be less than 0.
• 01:44 It is impossible to have less than 0 counts in the subgroup sample.
• 01:49 So if the lower control limit calculation would indicate that it is below 0,
• 01:53 we stop and cap it at 0.
• 01:55 Recall that the subgroup size was selected so
• 01:57 that the data would have an average above 2.
• 02:00 That will help to minimize the probability of this happening.
• 02:03 Well we went through the steps of creating a control chart in the previous module,
• 02:07 but let's look at some specifics about the C chart.
• 02:11 I assume that you know what attribute you're charting.
• 02:13 One of the key decisions that you must make is the sample size.
• 02:17 There are some ground rules that must be followed.
• 02:19 One is that the size is fixed, it's not constantly varying.
• 02:22 Another is that we want an average or mean of the sample counts to be greater than 2.
• 02:28 Based upon the typical use of this chart, the sample is a calendar criteria like
• 02:32 a week or a month, or it is some physical boundary such as within an airplane.
• 02:37 I've already mentioned the example of customer complaints,
• 02:40 well based upon the number of examples being received
• 02:43 you may want to use a sample of a day, a week, or a month.
• 02:47 Another example could be the number of incorrect answers on a standardized test.
• 02:52 Once your sample is set, count the occurrences in each sample.
• 02:56 When you have enough data points, calculate your control limits.
• 03:00 Once these limits are set you probably don't need to recalculate them with each
• 03:03 new sample, but only when a process change has occurred.
• 03:07 Now plot your data points and the control limits.
• 03:10 And of course if the chart shows that the process is not in control,
• 03:14 take an action to stabilize it.
• 03:17 Let's look at how we do these calculations manually, or in Excel.
• 03:20 As with other modules, formulas are shown on the right of the screen and
• 03:23 the calculation steps are discussed on the left side.
• 03:27 The first step is easy, count the attributes and put it in your spreadsheet.
• 03:31 I normally put all the data in one column.
• 03:33 Once you have enough data points, and we recommended 30,
• 03:36 you can calculate the mean or average for the data set.
• 03:40 Now calculate the control limits.
• 03:42 This is done by multiplying 3 times the square root of the mean of
• 03:45 the subgroup data.
• 03:47 Then add that product to the mean for the upper control limit, and
• 03:51 subtract that product from the mean for the lower control limit.
• 03:54 But remember, the lower control limit cannot be less than 0.
• 03:58 If you're using Excel, you can plot your data, the mean value, and the upper and
• 04:02 lower control limits using the line graph graphing option.
• 04:06 Now let's look at creating this chart in Minitab.
• 04:08 Go to the Start Menu, select Control Charts,
• 04:11 then select the Attribute Charts, and finally select the C chart.
• 04:15 When you do that, you should see the panel that looks like this.
• 04:19 Place your cursor in the variable window to activate the column display.
• 04:23 Highlight the column where your data is located.
• 04:26 Remember, you can cut and paste data directly from Excel into Minitab.
• 04:30 Then click the Select button,
• 04:32 your data column should now be in the Variable window.
• 04:35 Now click on the OK button in the bottom of the panel, and
• 04:38 Minitab will generate your control chart. >> So that's the C Chart, one of
• 04:43 the four attribute data control charts, and the easiest to use and understand.

Lesson notes are only available for subscribers.