Predictive analysis for joint progressive censoring plans: a Bayesian approach

Area of Science:

• Reliability Engineering
• Statistical Inference
• Survival Analysis

Background:

• Comparative lifetime experiments are crucial for assessing product reliability in production.
• Progressive Type-II censoring is a common method for efficiently collecting lifetime data.
• Estimating remaining lifetimes is vital for planning subsequent experiments, especially with non-homogeneous samples.

Purpose of the Study:

• To develop Bayesian methods for predicting failure times of surviving units under joint progressive Type-II censoring.
• To analyze non-homogeneous samples, particularly those from industrial storages.
• To illustrate inferential procedures with real-world data sets.

Main Methods:

• Utilizing a joint progressive Type-II censoring plan.
• Applying Bayesian prediction techniques for exponential parent populations.
• Analyzing two real data sets to demonstrate the proposed methods.

Main Results:

• The study provides a detailed discussion on Bayesian prediction of failure times.
• Inferential procedures are illustrated effectively using practical examples.
• The developed methods are shown to be useful for estimating remaining lifetimes in specific industrial contexts.

Conclusions:

• The Bayesian approach offers a robust method for predicting product reliability under progressive censoring.
• The findings are particularly valuable for industries dealing with non-homogeneous samples.
• This research enhances the ability to make informed decisions regarding product lifecycle management and future testing.