Recent Research on Classes of Life Distributions Useful in Maintenance Modelling.

Abstract

Several physically motivated broad classes of life distributions have been introduced recently and shown to be natural and useful in the study of maintenance policies, such as age replacement, block replacement, and replacement at failure only. These classes include the 'new better than used' class, the 'new better than used in expectation' class, and their duals. Recent papers studying these classes are surveyed, and results are summarized concerning the following: (a) replacement policy comparisons, (b) models giving rise to these classes, (c) preservation of classes under standard reliability operations, (d) bounds on reliability and mean life for both individual components and systems, and (e) statistical inference for these classes. Additional classes of life distributions such as the 'decreasing mean residual life' class, the 'decreasing percentile residual life' class, and their duals are briefly discussed. Inclusion relationships among all of these classes and reliability classes studied earlier are exhibited. (Modified author abstract)

Document Details

Document Type
Technical Report
Publication Date
Apr 01, 1974
Accession Number
AD0779938

Entities

People

  • Frank Proschan

Organizations

  • Florida State University

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Data Science
  • Inclusions
  • Information Science
  • Maintenance
  • Reliability
  • Residuals
  • Standards
  • Statistical Inference

Readers

  • Graph Algorithms and Convex Optimization.
  • Life Cycle Cost Analysis
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms