Skip to content

Teaching Tip: Using an SPC Chart to Examine American Airlines’ Pilots “Sick Out”

September 26, 2012

Looking for a current business issue to illustrate statistical process control when you are covering Supplement 6? The Wall Street Journal (Sept.24,2012) notes how “American Airlines continued to rack up high numbers of flight delays and cancellations, blaming a dispute with its pilots union. The union, meanwhile, denied that pilots disrupted flights unnecessarily.”

Percent of Pilots Sick

Can the same set of data be used to make opposite points in an argument?  It’s not that statistics lie, it is more in how we present all of the available data points, as can see in this timely example regarding the alleged “sick out” of American Airlines pilots. Here is a 13 month “snapshot” of percent of pilots out sick at American that you can use in class:  9/18/11, 4.9% ; 10/18/11, 9.5% ; 11/18/11, 5.0% ; 12/18/11, 6.5%, ; 1/18/12, 5.4% ; 2/18/12 , 6.6% ; 3/18/12, 6.6% ; 4/18/12, 6.0% ; 5/18/12, 7.0% ; 6/18/12, 7.4%; 7/18/12, 6.1%; 8/18/12, 6.6%; and 9/18/12, 7.5%. (The number of pilots dropped a few percent during this period in American’s financial struggles, from a high of 7,840 to a current 7,563.)

“By my calculations,” writes a Dallas Morning News(Sept. 20, 2012) reporter, “the number of pilots on sick leave was 45.7% higher on Sept. 18, 2012, than on Sept. 18, 2011, up 177 pilots. That seems like an increase in sick leave usage”. (See the bar chart graph above used to make this point).  American’s spokesman adds that sick leave “has been up more than 20 percent year over year and has been elevated for months.”

13 Month SPC Chart

Counters the union: “Contrary to claims by management, we have confirmed that pilot sick rates have not deviated from historical norms”. (Here we see the SPC p-chart showing percent of sick days being within p-chart control limits).

What happens to the p-chart if the 1st two months last year are excluded? Ask your students to recompute the control limits and draw conclusions.

One Comment leave one →
  1. July 20, 2015 8:23 pm

    I just received this email from Professor Albena Ivanova at Robert Morris University in Pittsburgh. She very carefully points out that the control limits are wrong, and too wide. The following is her analysis:

    “The limits are wrong. I tried to calculate them with the actual data provided in the article (you will see it in the data below) and it confirmed that they are wrong. I think that they are wrong on purpose. Remember that in the article the management accuses the pilots that they are calling in sick with increasing rates, however, they provide only comparison for two points in time, November this year versus November last year, maybe because they did not think/know of using control charts. The union replies with the control charts, but they are computed (and I think intentionally) wrong. If we compute the limits the percentage goes out of control for two months in the last year. You get the same results when you remove the first two months of historical data. I am surprised that the management did not do their own computations, then it would have been an easy win for them. The pilots are indeed calling in sick more over time.

    So I think this is a good example that we can use in class to show students that not everything that is published in the news is computed correctly. And very often people intentionally misrepresent data to prove their point of view.”

    Date Total number Number sick Percentage P-bar UCL LCL
    9/18/2011 7840.00 387.00 0.049362245 0.065512402 0.073996079 0.057028726
    10/18/2011 7807.00 740.00 0.09478673 0.065512402 0.073996079 0.057028726
    11/18/2011 7705.00 388.00 0.050356911 0.065512402 0.073996079 0.057028726
    12/18/2011 7642.00 498.00 0.065166187 0.065512402 0.073996079 0.057028726
    1/18/2012 7629.00 411.00 0.053873378 0.065512402 0.073996079 0.057028726
    2/18/2012 7630.00 502.00 0.065792923 0.065512402 0.073996079 0.057028726
    3/18/2012 7644.00 502.00 0.065672423 0.065512402 0.073996079 0.057028726
    4/18/2012 7623.00 458.00 0.060081333 0.065512402 0.073996079 0.057028726
    5/18/2012 7615.00 533.00 0.069993434 0.065512402 0.073996079 0.057028726
    6/18/2012 7636.00 566.00 0.074122577 0.065512402 0.073996079 0.057028726
    7/18/2012 7604.00 467.00 0.061415045 0.065512402 0.073996079 0.057028726
    8/18/2012 7583.00 504.00 0.06646446 0.065512402 0.073996079 0.057028726
    9/18/2012 7563.00 564.00 0.074573582 0.065512402 0.073996079 0.057028726
    p-bar 0.065512402
    Nr pilots – average 7655.46
    sigma 0.002827892

    I am using the book and the lab and your blog. I think that you and your colleagues have done wonderful job over the years. I especially like your videos with your comments at the end. The students like to see the authors of the book “in person” – it is more “human”.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Supply Chain Management Research

Andreas Wieland’s supply chain management blog for academics and managers

better operations

Thoughts on continuous improvement: from TPS to XPS

%d bloggers like this: