Prediction intervals for the lifetime of products in field
Currently manufacturers feel intense pressure to develop new and better products in shorter time, ensuring reliability in the hands of the customer and high quality to be competitive in the market. Many of these requirements need the implementation of Accelerated Life Tests (ALT) in the laboratory to provide short-term assessments of the reliability of the materials, components and subsystems of the product and to predict their performance under normal conditions of use. However, it is more appropriate to combine the results of laboratory tests with field data by formulating an appropriate statistical model to establish the relationship between both types of data and use them in the identification and estimation of the model of life then project or predict the effects of changes in product design for its future performance. In this paper we seek to construct prediction intervals for the lifetime of products in field from the failure time model proposed by Meeker et al. (2009), which is a mixture of distributions which combines information of the cycles to failure distribution in an ALT with a use rate model of product in the field.
| Main Authors: | , |
|---|---|
| Format: | Digital revista |
| Language: | spa |
| Published: |
Universidad Nacional de Colombia - Sede Medellín - Facultad de Ciencias
2012
|
| Online Access: | https://revistas.unal.edu.co/index.php/rfc/article/view/49097 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Currently manufacturers feel intense pressure to develop new and better products in shorter time, ensuring reliability in the hands of the customer and high quality to be competitive in the market. Many of these requirements need the implementation of Accelerated Life Tests (ALT) in the laboratory to provide short-term assessments of the reliability of the materials, components and subsystems of the product and to predict their performance under normal conditions of use. However, it is more appropriate to combine the results of laboratory tests with field data by formulating an appropriate statistical model to establish the relationship between both types of data and use them in the identification and estimation of the model of life then project or predict the effects of changes in product design for its future performance. In this paper we seek to construct prediction intervals for the lifetime of products in field from the failure time model proposed by Meeker et al. (2009), which is a mixture of distributions which combines information of the cycles to failure distribution in an ALT with a use rate model of product in the field. |
|---|