DURABILITY TESTS: statistical analysIs for variable amplitude loads

Abstract

The reduced time available for product evolution has forced original equipment manufacturers and their suppliers to develop new components and subsystems more rapidly, while taking into consideration the reliability of the final product in terms of its durability. Although durability is nowadays improved through virtual testing, it is mandatory to perform experimental tests for the final release of a product. The complexity of this kind of testing has increased. In the early 20th century, evaluations began to represent a real life time history with standardized load–time histories. This coincided with the introduction of the closed-loop test system to reproduce more realistic time histories. Most tier-1 suppliers perform such testing at their own development centers or at consultant test facilities; however, design engineers and product management need to understand the requirements of the testing and be able to interpret the results to understand how to improve a mechanical component. Most papers published on durability give experimental results for constant-amplitude loads and analytical predictions. However, information on tests with time histories and statistical analysis is very often missing. The present paper reviews the main aspects of durability tests for variable-amplitude loads, including the main statistical analysis conducted to reproduce damage to vehicles traveling on roads and proving grounds in the laboratory.