Comparing Designed Experiment Techniques to Improve DoD Flight Test Efficiency

The flight test enterprise in the Department of Defense (DoD) has traditionally maximized safety rather than the efficient selection of flight test matrices, leading to excessive test points, longer sortie lengths, and multiple flights required to complete a test. The DoD has encouraged the use of Design of Experiments (DOE) techniques in flight testing to reduce costs and ensure technical and statistical rigor. This research analyzed six DoD vibration flight test reports planned using One-Factor-At-a-Time (OFAT) methods. The vibration data were analyzed to re-accomplish each test matrix using three different DOE methods: (1) Full Factorial; (2) Plackett-Burman; and (3) Box-Behnken. The number of test points yielded by each DOE method and the coefficient of determination generated for each method were then compared to the original test matrices from the flight test reports. The results showed that DOE methods significantly reduced the size of the test matrix for several different aircraft types while maintaining a respectable level of statistical rigor. The Plackett-Burman and Full Factorial DOE methods offered the best combination of test point reduction and high R-sq (adj) values across the six examples analyzed. The fighter aircraft examples showed significant reductions in test point matrices, while the cargo aircraft examples showed increases in numbers of test points. The Full Factorial DOE method produced the highest R-sq (adj) values, with the Plackett-Burman and Box-Behnken methods not far behind (~10%). There was not a significant difference overall between the cargo and fighter aircraft examples in terms of average R-sq (adj) values across the three DOE methods.

Author

• Stephens, Michael J.

Language

• en

Keyword

• Plackett-Burman
• Design of Experiments
• Box-Behnken
• Military Aircraft
• Full Factorial
• Flight Test

Date created

• 2023

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Engineering Management

Degree

• D.Engr.

Advisor

• Etemadi, Amir H.

Committee Member(s)

• Young, Stephen D.
• Widener, Donald V.

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/v979v376q