In automotive and other industries, reliability validation involves a series of individual reliability tests, each targeting a subset of product overall failure modes. The design of each individual test is based on the predefined reliability target and confidence level, generally set as standards rather than customized per component and test. This method has been very useful. However, there are two major questions frequently asked about the validity of this approach. First, it is not clear how the reliability requirements of these individual tests are determined and how they relate to the product's overall reliability performance. Consequently, evaluating product overall reliability based on all individual test results is not straightforward. Second, it is not clear how the individual tests directly impact the warranty cost and the overall customer experience at the product level. If customer experience needs be improved or if the warranty cost needs to be reduced, it is difficult to determine what the new reliability requirement for these individual tests ought to be to achieve the necessary improvement at the product level.In this paper, a system approach to reliability testing is presented. The product reliability goal must first be determined by establishing a product field reliability performance requirement. Then the individual tests are linked to field performance through distributional usage and stress. Lastly, three methods are proposed to allocate the overall reliability goal to the reliability targets of each individual test. Each method utilizes different information to provide a possible quantitative way for such an allocation. This methodology bridges the gap between the individual tests and the product-level reliability performance, allowing for an easier risk assessment and statistically sound decision making. FIELD RELIABILITY GOAL AND ITS REQUIREMENT TO TEST DESIGNReliability engineers often have to answer questions about where reliability specification, such as X% reliability with Y% confidence, for test comes from and what the real consequences in the field are if the given specification is not met. Generally, specifications come from industry or company standards or traditions, but this explanation is becoming increasingly unsatisfactory to both reliability engineers and the engineers they work with. If design changes are needed to improve customer experience or warranty cost, this explanation is not even relevant. The problem lies with the lack of relationship between the reliability targets for an individual test and the overall product reliability expectation. To solve this problem, one first needs to define a product reliability performance requirement under real field usage and stress conditions. By basing all individual tests on known usage and stress conditions in the field, it is possible to establish a linkage between performance in the test and performance in the field and thus customer experience and warranty cost.Without loss of generality, the term "product" referred to in this p...
& CO CLUSIO SLife tests are generally very time-consuming and expensive. For a highly reliable product, even an accelerated life test takes too much time to meet business and product development needs. This is particularly true in automotive component testing. Traditional life test design with classical statistics does not consider prior knowledge or previous test results. In real applications, one typically has some prior knowledge before designing a new testing. This knowledge can come from previous test, expert opinion, engineering analysis, past similar product performance, or combination of them. Utilizing this prior knowledge can reduce the test sample size or increase the confidence, and more importantly, help to make decisions quicker. Some studies in the area of utilizing prior knowledge are developed through Bayes approach. [1 -4] are good examples in automotive industry application. However, the acceptance of Bayesian statistics proves to be much more challenging than the acceptance of classical statistics by management as well as by the engineering community and even by many reliability engineers. In this paper, we present a method that utilizes prior knowledge for the life test design but within a classical statistics framework.A Weibull time-to-failure distribution with a known shape parameter is used throughout the discussion. The prior knowledge of reliability and its confidence is translated into an imaginary or virtual life test to obtain an equivalent total time of testing and an equivalent number of failures. They are then quantitatively taken as the prior knowledge for a classical confidence bound based new life test design. This method will be compared with Bayes method. The prior knowledge determination and applications of the method developed here are discussed with examples. OPTIMISTIC PRIOR K OWLEDGEFor a two-parameter Weibull distribution, then the reliability within the specified time, t, is t t e e t R ) ( ) ( (1) Let t z , above equation becomes z e z R ) ( (2) This means z follows an exponential distribution with a meantime-to-failure of in z-space. Assume n units in a life test with testing time of n t t t , , 2 1 . For a time-censored test, the one-sided lower and upper confidence limits, respectively, on are given by 2 2 2 ; 1 2 r L Z (3) and 2 2 ; 1 1 2 r U Z (4) 1 = confidence level. 2 ;is percentage point of 2 distribution with the degree of freedom . r is the number of failures. Z is the total time of testing or test size in z-space, or n t t t Z 2 1 (5) Let us say we have an optimistic opinion about product reliability. We have a confidence of p 1 for a reliability of at least ) ( s L t R . Subscript p indicates prior knowledge. s t is the service life under test condition. For a time-to-failure Weibull distribution, we can incorporate this into an imaginary life test with a confidence calculated from 2 2 2 ; 2 p p r p L Z (6) p z is an imaginary total time of testing in z-space. Reliability for the service life, s t , is ) ( ) ( s t s e t R (7) Therefore, ) ( ln s s t R t ...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.