Abstract

Here, we develop a statistical basis for limited adverse testing. This type of testing simultaneously evaluates system performance against minimum requirements and minimizes costs, particularly for large-scale engineering projects. Because testing is often expensive and narrow in scope, the data obtained are relatively limited—precisely the opposite of the recent big data movement but no less compelling. Although a remarkably common approach for industrial and large-scale government projects, a statistical basis for adverse testing remains poorly explored. Here, we prove mathematically, under specific conditions, that setting each independent variable to an adverse condition leads to a similar level of adversity in the dependent variable. For example, setting all normally distributed independent variables to at least their 95th percentile values leads to a result at the 95th percentile. The analysis considers sample size estimates to clarify the value of replicates in this type of testing, determines how many of the independent variables must be set to adverse condition values, and highlights the essential assumptions, so that engineers, statisticians, and subject matter experts know when this statistical framework may be applied successfully and design testing to satisfy statistical requisites.

References

References
1.
Seborg
,
D. E.
,
Edgar
,
T. F.
, and
Mellichamp
,
D. A.
,
1989
,
Process Dynamics and Control
,
1st ed.
,
Wiley
,
New York
.
2.
Du
,
X.
,
2005
, “Probabilistic Engineering Design, Chapter 7 First Order and Second Reliability Methods,”
University of Missouri-Rolla
,
Rolla, Missouri
, accessed Oct. 21, 2019, http://web.mst.edu/∼dux/repository/me360/me360_lecture7.html
3.
Li
,
C.
, and
Mahadevan
,
S.
,
2017
, “
Robust Resource Allocation for Calibration and Validation Tests
,”
J. Verif., Validation Uncertainty Quantif.
,
2
(
2
), p.
021004
.10.1115/1.4037313
4.
Hu
,
K. T.
, and
Paez
,
T. L.
,
2016
, “
Why Do Verification and Validation?
,”
J. Verif., Validation Uncertainty Quantif.
,
1
(
1
), p.
011008
.10.1115/1.4032564
5.
Flierl
,
G.
, and
Ferrari
,
R.
,
2006
, “12.820 Turbulence in the Ocean and Atmosphere, Chapter 3: A Statistical Description of Turbulence,”
MIT OpenCourseWare, Massachusetts Institute of Technology
,
Cambridge, MA
, accessed Oct. 21, 2019, http://ocw.mit.edu/courses/earth-atmospheric-and-planetary-sciences/12-820-turbulence-in-the-ocean-and-atmosphere-spring-2006/lecture-notes/ch3.pdf
6.
Wackerly
,
D. D.
,
Mendenhall
,
W.
, III
, and
Scheaffer
,
R. L.
,
2008
,
Mathematical Statistics With Applications
,
7th ed.
,
Brooks/Cole
,
Belmont, CA
.
7.
Levenspiel
,
O.
,
1972
,
Chemical Reaction Engineering
,
2nd ed.
,
Wiley
,
New York
.
You do not currently have access to this content.