Estimation of Lifetime Parameters Using Warranty Data Consisting only of Failure Information
-
- ALAM Md. Mesbahul
- Department of Information Management Science, The University of Electro-Communications
-
- SUZUKI Kazuyuki
- Department of Systems Engineering, The University of Electro-Communications
-
- YAMAMOTO Watalu
- Department of Systems Engineering, The University of Electro-Communications
Bibliographic Information
- Other Title
-
- 保証期間中の市場故障データのみに基づく信頼性特性値の推定
Search this article
Abstract
A warranty database typically contains valuable field performance information obtained under actual operating conditions and is automatically updated whenever a customer requests repairs on a unit during the warranty period. However, it contains data on only those units that have been repaired under warranty; data (e.g., usage time) is not available for those units that have not failed during the warranty period. The current methods for estimating the lifetime parameters of products thus require gathering supplementary information for those units by, for example, conducting a follow-up study or by estimating the usage time distribution (such as mileage distance or copy volume) per specified calendar time. A method has now been developed that does not require such information, making it possible to estimate the reliability characteristics accurately using only failure information. It uses the "type" of the usage time distribution, which comes from previous literatures and the results of analyzing actual data sets. Maximum likelihood estimation is used for single-failure mode, and an iterative method is used for multiple-failure modes. This iterative method is applicable to analyze a general field data set, which consists of a lot of failure modes formulating a specified failure mode as a Weibull distribution and all other failure modes to which attention is not paid as an exponential distribution.
Journal
-
- Journal of The Japanese Society for Quality Control
-
Journal of The Japanese Society for Quality Control 39 (3), 353-363, 2009
The Japanese Society for Quality Control
- Tweet
Details 詳細情報について
-
- CRID
- 1390001204500492672
-
- NII Article ID
- 110007338362
-
- NII Book ID
- AN00354769
-
- ISSN
- 24321044
- 03868230
-
- NDL BIB ID
- 10383101
-
- Text Lang
- en
-
- Data Source
-
- JaLC
- NDL
- CiNii Articles
-
- Abstract License Flag
- Disallowed