Paper
25 June 2002 Bayesian model for the failure rate estimation of photonic components
Chang W. Kang, George E. Soltysik, William Denson
Author Affiliations +
Abstract
Estimating the reliability of new products is an important part of new product development. There are several ways in which this reliability can be estimated, including the use of: field data, empirical reliability prediction models, test data, and reliability physics based models. Field data is usually the most desirable for estimating the reliability of a product since it represents the deployment conditions and therefore represents all applicable stresses to which the item is exposed. There is typically not enough operational field data to estimate the reliability of products in a statistically robust manner, and therefore using field data alone leads to low statistical confidence inferences on the measure of failure rates. This paper presents a practical and structured approach to estimating the reliability of components used in photonic systems. This approach consists of a generic model that allows for the fusion of various data types (e.g., tests, models, simulations), and merges available data with a Bayesian approach to form the best estimate of component reliability. This paper explains the theoretical robustness of the method, and demonstrates the Bayesian method. The approach is particularly suitable for photonic components given the short development times typical in the telecommunications industry and the resulting relative scarcity of field data.
© (2002) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Chang W. Kang, George E. Soltysik, and William Denson "Bayesian model for the failure rate estimation of photonic components", Proc. SPIE 4639, Optical Fiber and Fiber Component Mechanical Reliability and Testing II, (25 June 2002); https://doi.org/10.1117/12.481333
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Failure analysis

Reliability

Data modeling

Data fusion

Humidity

Statistical analysis

Systems modeling

RELATED CONTENT

Embedded network risk model based on random forest algorithm
Proceedings of SPIE (October 20 2022)
Fiber reliability standards
Proceedings of SPIE (March 01 1994)
Modeling connector degradation
Proceedings of SPIE (June 25 2002)
Life cycle cost of military displays
Proceedings of SPIE (August 16 1999)

Back to Top