A Scientific Approach for Designing Conservative Tests in Vehicle Development 2008-01-0848
This paper suggests a scientific approach to designing conservative tests based on computer simulation of the influence of the sources of variations. The idea is to design the conservative test so that, even in the presence of variation, there is a high probability that a random test will have a better result than the conservative test. Therefore, if the conservative test meets the requirement, one has a scientific reason to believe that any random test would have a high probability of meeting it. This new approach is illustrated for FMVSS301 80 kph 70% rear offset deformable barrier impact.
SAE International Journal of Materials and Manufacturing-V117-5EJ, Reliability and Robust Design in Automotive Engineering, 2008-SP-2170, SAE International Journal of Materials and Manufacturing-V117-5