Bill Parr
Professor and Head, Department of Statistics
332 Stokely Management Center
University of Tennessee
Knoxville, Tennessee 37996-0532
E-mail Address:
Web Address:
Phone: 423-974-1631
Fax: 423-974-2490
Address: Department of Statistics
332 Stokely Management Center
University of Tennessee
Knoxville, Tennessee 37996-0532

Title: Sources of Variation: How SHOULD You Set Limits for an Xbar Chart?
Duration: 1 - 3 hours, as desired.
Audience: Those in the audience should have at least a nodding familiarity with the basic control charts. It will be most helpful if a good number in the audience have themselves used control charts. Industries where the talk has been particularly interesting have included semiconductor manufacturing, pharmaceuticals, and other continuous or batch processing applications.
Abstract: We discuss the implications of which sources of variation are allowed to enter into the limit calculation for Xbar charts, in terms of i)   ability of the user to learn from the chart, ii) ability of the chart t aid in detection of process changes, and iii) the substantial (and substantially misleading) statistical literature discussing control charts. In particular, we examine how most common graduate and undergraduate textbooks in common use encourage bad misconceptions about the appropriate application of the control chart.
Honorarium: Not expected. I would expect my travel expenses to be picked up, and would hope for the opportunity to talk with a variety of people from the sponsoring organization during my visit.

Title: How do World Class Organizations Use Statistics?
Duration: This talk is 1 - 1.5 hours in length, with ample opportunity for discussion during or following the talk.
Audience: Audience:
Abstract: I review the results of interviews with representatives of over 300 different organizations. Based on these interviews, I have determined those practices in the use of statistics and statistical thinking which best differentiate between world class performers and others. Here, world class performers are defined as organizations with a track record of obtaining high impact from use of statistical organizations over a period of 5 - 10 years or more. The targeted audience can be either executives and managers who employ statisticians, or statisticians who wish to review their own organization's strategy for use of statistics.
Honorarium: Honorarium is not required, but travel expenses must be covered.
Honorarium: Expected.