My Shopping Cart
There are no items in your shopping cart.

Why and How to use Design of Experiments (MFG561X)

Presented by: William A. Levinson, P.E.
(*) Single User Price. For multiple users please call 1-800-223-8720
Pre Recorded Webinar
60 minutes
  •  Thu, May 26, 2016
Event Description

Learn how Design of Experiments can Save Time and Resources While Delivering Actionable Results

Design of Experiments is part of the body of knowledge for ASQ certifications ranging from Certified Quality Engineer to Certified Six Sigma Green or Black Belt. It is also a vital tool for defect reduction and process improvement. It allows us to know beyond a reasonable doubt, rather than guess, that an experimental treatment is better than the control or process of record, and to determine beyond a reasonable doubt that a suspected assignable cause is actually responsible for poor quality.

A complete overview of DOE is a subject for a full-day workshop, and mastery of the standard techniques requires two or more college courses. In this session, our expert speaker William A. Levinson will provide you with the basic introduction, which requires no prior knowledge of statistics, covers fundamentals such as factors and levels, hypothesis testing, the need for random samples, and interpretation of results.

Session Highlights:

  • Understand the fundamentals of hypothesis testing; the foundation of everything we do with industrial statistics:
    • Recognize the null hypothesis as the counterpart of the presumption of innocence in a criminal trial. A statistical test begins with the assumption that the process is in control and does not need adjustment (SPC), a production lot is acceptable for shipment (acceptance sampling), or no difference between an experimental treatment and a control.
    • Know that the alternate hypothesis (the process is out of control, the production lot should be rejected, or the experiment differs from the control) must, like guilt in a criminal trial, be proven beyond a reasonable doubt. This reasonable doubt can be quantified as the Type I, producers, or alpha risk, and it is also known as the significance level. It is the risk of concluding wrongly that the process is out of control, the production lot should be rejected, or there is a difference between the experimental treatment and the control.
    • Know that there is also a risk (Type II, consumer's, or beta risk) of not rejecting the null hypothesis when it should be rejected.
  • Learn how DOE can save enormous amounts of resources and time. The webinar will cite a historical example (prior to the invention of DOE) that shows how it is easily possible to waste millions of dollars in today's money, as well as years, in the absence of DOE, while DOE can on the other hand solve very complicated problems in weeks.
  • Understand the concepts of factors, levels, and interactions (the whole is greater or less than the sum of its parts).
  • Understand the importance of specimen randomization, blocking, and similar techniques to keep extraneous variation out of the experiment.
    • Also recognize the need to obtain enough data (replication) to know beyond a reasonable doubt that the experiment worked.
  • Interpret experimental results in terms of significance level and P value (chance that any observed differences in results are due to luck or random chance).


  1. A comprehensive understanding of hypothesis testing will equip you to understand all applications of industrial statistics, including not only DOE but also SPC and acceptance testing.
  2. You will be able to explain, in the language of time and money, how DOE saves time and money while it delivers actionable results.
  • You will be able to use the concept of interactions to explain why one variable at a time (OVAT) experiments are often inaccurate as well as cost-ineffective. Frederick Winslow Taylor's efforts to optimize a metal cutting operation by adjusting one variable at a time required more than 20 years and consumed half a million dollars in the money of the late 19th century. In addition, a factorial design can eliminate from further consideration factors that do not affect the result, and allow focus on those that do.
  • Provides a sufficient foundation for you to understand the language of DOE as used with regard to workplace improvement or corrective action projects.

Who should attend?

  • Managers/Executives responsible for root cause analysis (part of corrective action) and productivity and quality improvement in manufacturing and potentially service industries.
  • Quality engineers
  • Quality managers
  • Project managers and project quality personnel
  • Risk management personnel
  • Technicians
About Our Speaker(s)

William A. Levinson | Quality and Productivity Management SpeakerWilliam A. Levinson P.E.
William A. Levinson, P.E., is the principal of Levinson Productivity Systems, P.C. He is an ASQ Fellow, Certified Quality Engineer, Quality Auditor, Quality Manager, Reliability Engineer, and Six Sigma Black Belt. He is also the author of several books on quality, productivity, and management, of which the most recent ... More info

Recommend this event to a colleague
  • Your Name:
  • Your E-mail:
  • Colleague's Name:
  • Colleague's E-mail:
    Subject: You might be interested in this event from AudioSolutionz
    Event Title: Why and How to use Design of Experiments
    Presenter(s): William A. Levinson, P.E. Protection Status