In this paper, two kinds of compliance prediction techniques (single test compliance and population compliance) on the vehicle crash testing are discussed. For each one of the above two techniques, the confidence (C) in meeting the government mandated requirement based upon successful test results were derived with demonstrated examples. In Sections 2 and 3, the derivations of confidences are based on the assumption that the population distribution is approximately normal. Actually, exact normality is not always found in the real world. One can use a normal probability plot or a normality goodness-of-fit test to check whether the data are normally distributed. If data are not normal, one can transform the original data in such a way that the transformed data will meet the normal distribution assumption. The Box-Cox power transformation can be used to transform the original non-normal data to normal data. If the data can be transformed to be normal, then compute the confidence values by using the transformed data.
Statistical techniques for vehicle crash testing
International Journal of Materials and Product Technology ; 20 , 1-3 ; 91-100
2004
10 Seiten, 2 Tabellen, 4 Quellen
Article (Journal)
English
TIBKAT | 1971
|Sign Convention for Vehicle Crash Testing
SAE Technical Papers | 2007
Sign Convention for Vehicle Crash Testing
SAE Technical Papers | 2018
SIGN CONVENTION FOR VEHICLE CRASH TESTING
SAE Technical Papers | 1994
Simulation proving superior to vehicle crash testing
Automotive engineering | 1980