%40تخفیف

NEW AGING CLASSES OF LIFE DISTRIBUTIONS UNDER SHOCK MODELS

تعداد89 صفحه در فایل word

Ph.D. DISSERTATION IN

MATHEMATICAL STATISTICS

 

NEW AGING CLASSES OF LIFE DISTRIBUTIONS UNDER SHOCK MODELS

ABSTRACT

In this thesis we introduce a new aging class of life distributions when a device is operating in a realistic environment. We study the behavior of such life distributions through the mean residual life notion, when a device is experiencing number of shocks. Due to these shocks the lifetime of such device has become shortened or prolonged. These tempered events are governed by a homogenous Poisson process. A moment inequality which characterizes this new aging class, namely renewal increasing mean residual life (RIMRLshock), is derived. We propose a new U-statistic test procedure to address the problem of testing exponentiality against such class of life distributions. It is shown that the proposed test enjoys a superior power for some commonly used alternatives.  Two sets of real data are used as the examples to elucidate the use of our proposed test statistic.

Table of Contents

Content                                                                                                               Page

List of Figures                                                                                                                    xi

List of Tables                                                                                                                     xii

 

Chapter 1: Introduction

1.1 Aims . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .    1

1.2 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   1

            1.3 Notation and Definitions  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   2

            1.4 Increasing (Decreasing) Failure Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .    3

1.5 Mean Residual Lifetime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .    4

1.5.1 Construction of the test Statistic   . . . . . . . . . . . . . . . . . . . . . . . . .  .   4

1.5.2 Decreasing (Increasing) MRL    . . . . . . . . . . . . . . . . . . . . . . . . . .  . .  5

            1.5.2.1 Testing against DMRL alternatives  . . . . . . . . . . . . . . . . . .  6

1.6 Variance Residual Lifetime   . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

            1.6.1 Decreasing (Increasing) VRL  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  7

1.6.1.1 Testing DVRL (IVRL) against Exponentiality  . . . . . . . . .  8

1.7 Increasing Failure Rate Average  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . .   8

1.7.1 Testing against IFRA alternatives   . . . . . . . . . . . . . . . . . . . . . .  . . .   9

1.8 New Better than Used . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . . .  10

1.9 New Better than Used in Expectation   . . . . . . . . . . . . . . . . . . . . . . . . . . . .  .  10

1.10 New Better than Used in Convex Ordering  . . . . . . . . . . . . . . . . . . . .  . . .  . 11

1.10.1 Testing against NBUC alternatives  . . . . . . . . . . . . . . . . . . . .  . . .  11

1.11 Net Decreasing VRL  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  12

1.11.1 Testing NDVRL against Exponentiality . . . . . . . . . . . . . . . . . . . . . 12

1.12 Implication between Classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

1.13 Increasing Mean Doubly Truncated   . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . 14

1.14 Increasing Variance Doubly Truncated   . . . . . . . . . . . . . . . . . . . . . . . . . . .   14

1.15 Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  15

Chapter 2: Shock and Damage Models in Reliability Theory

2.1 Introduction  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   18

2.2 Renewal Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   21

                        2.2.1 Renewal Process   . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   22

                        2.2.2 Poisson Process   . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  .   24

2.3 Shock Processes   . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  26

2.4. Damage Models  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  27

2.4.1 Cumulative Damage Models   . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  29

2.4.2 Independent Damage Models  . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  34

Chapter3: Renewal Increasing Mean Residual Life Distributions: An age Replacement Model with Hypothesis Testing Application

3.1 Introduction   . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.2 Basic definitions and properties   . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  39

3.3 Moment inequality  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

3.4 Testing exponentiality against RIMRLshock alternatives . . . . . . . . . . . . . . . .  48

3.5 Power Simulation and numerical examples      . . . . . . . . . . . . . . . . . . . . . . .  53

3.5.1. Simulated Power Studies  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

3.6 Numerical examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Appendix A:

R Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . . . .  . .. .. . . . . . . . . . . . . . . . . . 58

  • Simulated percentile . . . . . . . . . . . . . . . . . . . .  . … . . .. . . . . . . . . . . . . . . . . . . 58

  • The empirical power estimate . . . . . . . . .  . . . .  . .. .. . . . . . . . . . . . . . . . . . . . . 59

    • Simulation Weibull . . . . . . . . . . . . .  . .. .. . . . . . . . . . . . . . . . . . . .  59

    • Simulation Gamma . . . . . . . . . . . . .  . .. .. . . . . . . . . . . .  . . . . . . . .  61

1.2.3 Simulation Linear Failure Rate . . . . . . . . .  . . . .  . .. .. . . . . . . .  . .  ..  62

References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  64

List of Figures

Figure 1.1. The hazard function h (t) bathtub shape function   . . . . . . . . . . . . . . . . . . . . . . . . .    3

Figure 2.1. Total number of failed units over the time    . . . . . . . . . . . . . . . . . . . . . . . . .  21

Figure 2.2. Standard cumulative damage model  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  28

Figure 2.3. Independent damage model  . . . . . . . . . . .  .  . . . . . . . . . . . . . . . . . . . . . . . . . 34

Figure 3.1. Cumulative damage model . . . . . . . . .   . . . . . . . . . . . . . . . . . . . . . . . . . . .  .  40

Figure 3.2. Relation between critical value, sample size and confidence level. . . . . . . .   54

List of Tables

Table 4.1.Critical values for the upper percentiles of  . . . . . . . . . . . . . . . . . . . . . . . . . .  60

Table 4.2. Power estimates for RIMRLshock using α=0.05 . . . . . . . . . . . . . . . . . . . . . . . . . 62

قبلا حساب کاربری ایجاد کرده اید؟
گذرواژه خود را فراموش کرده اید؟
Loading...
enemad-logo