. » . . v .‘.....\...‘,, .p, .u... “Hwy-nu. .

 

   

 

 

   

 

 

 

_," - u
' . - , i
‘ . ‘ ,
. ' .

 

 

.. ‘ /
. ‘f‘ . ( ‘

 

 

 

 

 

 

 

,w‘l

)
36504303

This is to certify that the
thesis entitled

 

THE USE OF RELIABILITY ANALYSIS FOR DETERMINING
THE LIFE EXPECTANCY OF PREVENTIVE MAINTENANCE
FIXES IN APSHALT SURFACED PAVEMENTS

presented by
JASON PAUL BAUSANO

has been accepted towards fulfillment
of the requirements for the

MS. degree in CIVIL ENGINEERING

" I
' Major Professor’s Signa‘tUlle I T

(79464 ZNZOQ
t/ / ’
Date

 

 

 

MSU is an Affinnative Action/Equal Opportunity Institution

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRCIDateDue.p65-p.15

THE USE OF RELIABILITY ANALYSIS FOR DETERMINING THE LIFE
EXPECTANCY OF PREVENTIVE MAINTENANCE FIXES IN ASPHALT
SURFACED PAVEMENTS
By

Jason Paul Bausano

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Civil and Environmental Engineering

2003

ABSTRACT
THE USE OF RELIABILITY ANALYSIS FOR DETERMINING THE LIFE
EXPECTANCY 0F PREVETNIVE MAINTENANCE FIXES IN APSHALT
SURFACED PAVEMENTS
By

Jason Paul Bausano

The research in this thesis used a large data set of pavement performance
parameters in the form of Distress Index (DI) and Ride Quality Index (RQI) to conduct a
reliability-based analysis for the purpose of determining the life expectancy and
evaluating the guidelines for preventive maintenance (PM) fixes.

The model developed in this research is a probabilistic model that uses pavement
performance parameters, (DI and RQI). Probability distributions were developed for the
following fixes over time: Non-structural bituminous overlay, surface milling with a non-
structural bituminous overlay, single chip seal, multiple course micro-surfacing, and
bituminous crack seal.

Reliability tables were then developed expressing the probability that a given PM
treatment will not reach the performance threshold afier n years. These tables provide the
pavement life expectancy for a given PM treatment at various reliability levels. A
highway agency can then use these tables to select PM strategies based on the expected
life of the fix. In addition, the PM guidelines (in terms of D1 and RQI) were evaluated

using a series of two-sample t-tests as well as using the reliability approach.

ACKNOWLEDGEMENTS

First, I would like to extend my thanks and gratitude to Robert and Ellen
Thompson for providing the fellowship at Michigan State University (MSU) so I could
further my education and improve the research in the field of asphalt pavements.

A special thanks goes to my advisor, Dr. Chatti, for his guidance and help during
my research project at MSU and Michigan Technological University (MTU). I would
also like to thank him for allowing me to attend MTU during the fall semester of 2002.

I would also like to thank Dr. Baladi and Dr. Wolff for sitting on my master’s
committee and providing me with support and guidance throughout my education.

A special thanks also goes to Dr. Williams for taking on “one” more graduate
student while I was at MTU, providing me with additional insight, and guidance for my
research project, and sitting on my masters thesis committee.

Finally, I extend a sincere thanks to my parents, family, and friends for their never

ending support and encouragement throughout all my years of study.

iii

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. vii
LIST OF FIGURES ................................................................................. ix
CHAPTER 1 - INTRODUCTION ...................................................................................... 1
1.1 Problem Statement .............................................................................................. 1

1 .2 Maintenance ........................................................................................................ 2
1.2.1 Preventive Maintenance .............................................................................. 2
1.2.2 Corrective Maintenance .............................................................................. 3
1.2.3 Routine Maintenance .................................................................................. 3

1.3 Research Objective and Organization ................................................................. 4
1.3.1 Research Objective ..................................................................................... 4
1.3.2 The Proposed Model ................................................................................... 4

1 .3 .3 Organization ................................................................................................ 5
CHAPTER 2 - LITERATURE REVIEW ........................................................................... 7
2. 1 Introduction ......................................................................................................... 7
2.2 Facts about Preventive Maintenance .................................................................. 8
2.3 Asphalt Pavement Distresses .............................................................................. 9
2.4 Preventive Maintenance Treatments ................................................................. 10
2.5 Selecting Appropriate Treatment and Timing .................................................. 20
2.6 Performance Findings from Previous Studies ................................................... 26
2.6.1 Survival Modeling .................................................................................... 27
2.6.2 Regression Modeling ................................................................................ 29
2.6.3 Statistical Modeling .................................................................................. 31
2.6.4 Probabilistic Modeling-Markovian Chains ............................................... 32
2.6.5 Engineering Field Review of SPS-3 Projects ........................................... 33
2.6.6 Engineering Field Review of MDOT’S CPM Program ............................ 33

2.7 Conclusions ....................................................................................................... 34
CHAPTER 3 - THE MDOT PAVEMENT PERFORMANCE MEASURES .................. 36
3. 1 Introduction ....................................................................................................... 36
3.2 Surface Distress Data ........................................................................................ 36
3.3 Longitudinal Pavement Profile Data ................................................................. 38
3.4 Rut Depth .......................................................................................................... 39
3.5 Friction Data ..................................................................................................... 39
3.6 MDOT’s Definition of Life Extension ............................................................. 39
CHAPTER 4 - EXTRACTION AND DEVELOPMENT OF DATABASE .................... 41
4. 1 Introduction ....................................................................................................... 41
4.2 Extraction of Data ............................................................................................. 41
4.3 Development of Database ................................................................................. 43
4.4 Compilation of the Data .................................................................................... 45

iv

CHAPTER 5 - RELIABILITY BASED MODEL AND ANALYSIS ............................. 46

5. 1 Introduction ....................................................................................................... 46
5.2 Reliability-Based Model ................................................................................... 47
5.3 Reliability Analysis ........................................................................................... 50
5.4 Two-Sample t-test for Significance .................................................................. 51
5.4.1 Pavement Condition Using Distress Index (DI) ....................................... 52
5.4.2 Pavement Condition Using Ride Quality Index (RQI) ............................. 53
CHAPTER 6 - RESULTS ................................................................................................. 55
6.1 Introduction ....................................................................................................... 55
6.2 Determination of Life Expectancy of the Various PM Fixes ........................... 55
6.2.1 Non-Structural Bituminous Overlay ......................................................... 57
6.2.2. Surface Milling with a Non-Structural Bituminous Overlay .................... 59
6.2.3. Single Chip Seal ........................................................................................ 60
6.2.4. Multiple Course Micro-Surfacing ............................................................. 62
6.2.5. Bituminous Crack Seal ............................................................................. 64
6.2.6. Overall Comparison .................................................................................. 66
6.3 Evaluation of the Guideline Values for the Treatment Types .......................... 70
6.3.1 Two-Sample t-test Approach .................................................................... 70
6.3.1.1 Non-Structural Bituminous Overlay ..................................................... 71
6.3.1.2 Surface Milling with a Non-Structural Bituminous Overlay ................ 72
6.3.1.3 Single Chip Seal .................................................................................... 72
6.3.1.4 Multiple Course Micro-Surfacing ......................................................... 72
6.3.1.5 Bituminous Crack Seal ......................................................................... 73
6.3.2. Reliability Approach ................................................................................. 76
6.3.2.1 Non-Structural Bituminous Overlay ..................................................... 76
6.3.2.2 Surface Milling with a Non-Structural Bituminous Overlay ................ 78
6.3.2.3 Single Chip Seal .................................................................................... 80
6.3.2.4 Multiple Course Micro-Surfacing ......................................................... 81
6.3.2.5 Bituminous Crack Seal ......................................................................... 83

6.4 Evaluation of Ride Quality Guidelines ............................................................. 86
6.4.1 Non-Structural Bituminous Overlay ......................................................... 87
6.4.2 Surface Milling with a Non-Structural Bituminous Overlay .................... 87
6.4.3 Single Chip Seal ........................................................................................ 88
6.4.4 Multiple Course Micro-Surfacing ............................................................. 88
6.4.5 Bituminous Crack Seal ............................................................................. 88
CHAPTER 7 - SUMMARY OF FINDINGS AND RECOMMENDATIONS ................ 91
7.1 Summary of Findings ........................................................................................ 91
7.2 Recommendations for Future Research ............................................................ 94
APPENDIX A ................................................................................................................... 97

HISTOGRAMS USED TO DETERMINE THE LIFE EXPECTANCY FOR THE
FOLLOWING PREVENTIVE MAINTENANCE (PM) TREATMENTS BASED ON
THE DISTRESS INDEX (DI)

APPENDIX B ................................................................................................................. 118
HISTOGRAMS USED TO EVALUATE THE DISTRESS INDEX (DI) PREVENTIVE
MAINTENANCE (PM) GUIDELINE VALUES FOR THE FOLLOWING PM
TREATMENTS

APPENDIX C ................................................................................................................. 152
HISTOGRAMS USED TO EVALUATE THE RIDE QUALITY INDEX (RQI) PM
GUIDELINE VALUES FOR THE FOLLOWING PREVENTIVE MAINTENANCE
(PM) TREATMENTS BASED ON THE DISTRESS INDEX (DI)

APPENDIX D ................................................................................................................. 184
TWO SAMPLE T-TEST RESULTS FROM MINI-TAB USED TO EVALUATE THE
DISTRESS INDEX (DI) PREVENTIVE MAINTENANCE (PM) GUIDELINE
VALUES FOR THE FOLLOWING PM TREATMENTS

APPENDIX E ................................................................................................................. 195
TWO SAMPLE T-TEST RESULTS FROM MINI-TAB USED TO EVALUATE THE
RIDE QUALITY INDEX (RQI) PREVENTIVE MAINTENANCE (PM) GUIDELINE
VALUES FOR THE FOLLOWING PM TREATMENTS BASED ON THE DISTRESS
INDEX

LIST OF REFERENCES ......................................................................... 207

vi

LIST OF TABLES

Table 2.1 — MDOT Performance Threshold Values for Non-Structural Bituminous

Overlay (MDOT, 1998) ............................................................................................ 11
Table 2.2 - Estimated Life Extension for Non-Structural Bituminous Overlay (MDOT,
1998) ......................................................................................................................... 11
Table 2.3 -— MDOT Performance Thresholds for Surface Milling with a Non-Structural
Bituminous Overlay (MDOT, 1998) ......................................................................... 12
Table 2.4 - Estimated Life Extension for Surface Milling with a Non-Structural
Bituminous Overlay (MDOT, 1998) ......................................................................... 13
Table 2.5 — MDOT Performance Thresholds for Chip Sealing (MDOT, 1998) ............... 13
Table 2.6 - Estimated Life Extension for Chip Seals (MDOT, 1998) .............................. 14
Table 2.7 — MDOT Performance Thresholds for Micro-Surfacing (MDOT, 1998) ......... 15
Table 2.8 - Estimated Life Extension for Micro-Surfacing (MDOT, 1998) ..................... 15

Table 2.9 — MDOT Performance Thresholds for Bituminous Crack Treatment (MDOT,
1998) ......................................................................................................................... 16

Table 2.10 - Estimated Life Extension for Bituminous Crack Treatment (MDOT, 1998)17

Table 2.11 — MDOT Performance Thresholds for Overband Crack Filling (MDOT, 1998)

................................................................................................................................... 17
Table 2.12 - Estimated Life Extension for Overband Crack Filling (MDOT, 1998) ....... 18
Table 2.13 — MDOT Performance Thresholds for Ultra-Thin Bituminous Overlay

(MDOT, 1998) .......................................................................................................... 19
Table 2.14 - Estimated Life Extension for Ultra-Thin Bituminous Overlay (MDOT, 1998)

................................................................................................................................... 20
Table 2.15 - Appropriate Maintenance Strategies for Various Distress Types (Hicks, eta],

1998) ......................................................................................................................... 22
Table 2.16 - Factors Affecting Preventive Mainenance Treatments ................................ 23

Table 2.17 - Results of 1999 Southern Region SPS-3 Survival Analysis (Elthahan et. a1.)
................................................................................................................................... 28

vii

Table 3.1 - Pavement Distresses Collected by the MDOT ............................................... 37

Table 3.2 - Distress Index Categories ............................................................................... 38
Table 3.3 - Ride Quality Categories ................................................................................. 39
Table 5.1 - Suggested Levels of Reliability for Various Functional Classifications
(AASHTO, 1986) ...................................................................................................... 50
Table 5.2 - Number of Data Points for Each Fix Type ..................................................... 51
Table 6.1 - First Estimate of Reliability Values over time for the Various PM Fixes ...... 67
Table 6.2 - Estimated Reliabilities of Life Expectancy for Various Fixes Based on
MDOT’S PM Guidelines ........................................................................................... 68
Table 6.3 — Estimated Reliabilities of Life Expectancy for Various Fixes Based on
MDOT’S Rehabilitation Threshold ........................................................................... 69
Table 6.4 — Two Sample t-test Results Using DI .............................................................. 74
Table 6.5 — Summary Table for Reliability Analysis on Evaluating the Guideline Values
for the Treatment Types ............................................................................................ 85
Table 6.6 — Two Sample t-test Results Based on RQI ...................................................... 89

viii

LIST OF FIGURES

Figure 2.1 - Properly Applied Preventive Maintenance (Galehouse, 1998) ..................... 24
Figure 2.2 - Delayed Preventive Maintenance (Galehouse, 1998) ................................... 25

Figure 2.3 - Conceptual Relationship for Timing of Various Maintenance and
Rehabilitation Treatments (Hicks et. al., 1998) ........................................................ 26

Figure 4.1 - Number of Preventive Maintenance Projects with Distress Index Data ....... 42

Figure 4.2 - Number of Preventive Maintenance Projects with Ride Quality Index Data 42

Figure 4.3 — Schematic drawing showing the above calculation ...................................... 44
Figure 5.1 - Schematic Showing the Reliability Model (Pendleton, 1994) ...................... 47
Figure 5.2 - Example Calculation for Reliability Analysis ............................................... 49
Figure 6.1 -— Average DI after Preventive Maintenance versus Time ............................... 56

Figure 6.2 - Non-Structural Bituminous Overlay Reliability versus Time Based on the
PM Guidelines (95% Confidence Interval) ............................................................... 58

Figure 6.3 - Non-Structural Bituminous Overlay Reliability versus Time Based on the
Rehabilitation Threshold (95% Confidence Interval) ............................................... 58

Figure 6.4 — Surface Milling with a Non-Structural Bituminous Overlay Reliability
versus Time Based on the PM Guidelines (95% Confidence Interval) .................... 59

Figure 6.5 — Surface Milling with a Non-Structural Bituminous Overlay Reliability
versus Time Based on the Rehabilitation Threshold (95% Confidence Interval) 60

Figure 6.6 - Single Chip Seal Reliability versus Time Based on the PM Guidelines (95%
Confidence Interval) ................................................................................................. 61

Figure 6.7 — Single Chip Seal Reliability versus Time Based on the Rehabilitation
Threshold (95% Confidence Interval) ...................................................................... 62

Figure 6.8 — Multiple Course Micro-Surfacing Reliability versus Time Based on the PM
Guidelines (95% Confidence Interval) ..................................................................... 63

Figure 6.9 — Multiple Course Micro-Surfacing Reliability versus Time Based on the
Rehabilitation Threshold (95% Confidence Interval) ............................................... 63

ix

Figure 6.10 — Bituminous Crack Sealing Reliability versus Time Based on the PM

Guidelines (95% Confidence Interval) ..................................................................... 65
Figure 6.11 — Bituminous Crack Sealing Reliability versus Time Based on the
Rehabilitation Threshold (95% Confidence Interval) ............................................... 65
Figure 6.12 — Reliability of Life Expectancy for Non-Structural Bituminous Overlay
When PM is done before the DI Guideline Value .................................................... 77
Figure 6.13 — Reliability of Life Expectancy for Non-Structural Bituminous Overlay
When PM is done after the DI Guideline Value ....................................................... 78
Figure 6.14 - Reliability of Life Expectancy for Surface Milling with a Non-Structural
Bituminous Overlay When PM is done before the DI Guideline Value ................... 79
Figure 6.15 — Reliability of Life Expectancy for Surface Milling with a Non-Structural
Bituminous Overlay When PM is done after the DI Guideline Value ..................... 79
Figure 6.16 — Reliability of Life Expectancy for Single Chip Seal When PM is done
before the DI Guideline Value .................................................................................. 80
Figure 6.17 — Reliability of Life Expectancy for Single Chip Seal When PM is done after
the DI Guideline Value ............................................................................................. 81
Figure 6.18 — Reliability of Life Expectancy for Multiple Course Micro-Surface When
PM is done before the DI Guideline Value ............................................................... 82
Figure 6.19 -— Reliability of Life Expectancy for Multiple Course Micro-Surface When
PM is done after the DI Guideline Value .................................................................. 82
Figure 6.20 - Reliability of Life Expectancy for Bituminous Crack Seal When PM is
done before the DI Guideline Value ......................................................................... 83
Figure 6.21 — Reliability of Life Expectancy for Bituminous Crack Seal When PM is
done after the DI Guideline Value ............................................................................ 84
Figure A.1 - Histogram of D1 1 Year after PM ................................................................. 98
Figure A.2 - Histogram of D1 2 Years afier PM ............................................................... 98
Figure A.3 - Histogram of D1 3 Years after PM ............................................................... 99
Figure A.4 - Histogram of D1 4 Years after PM ............................................................... 99
Figure A.5 - Histogram of D1 5 Years after PM ............................................................. 100

Figure A.6 - Histogram of D1 6 Years after PM ............................................................. 100

Figure A.7 - Histogram of DI 7 Years after PM ............................................................. 101
Figure A.8 - Histogram of D1 8 Years afier PM ............................................................. 101
Figure A.9 - Histogram of DI 1 Years after PM ............................................................. 102
Figure A.10 - Histogram of DI 2 Years after PM ........................................................... 102
Figure A.11 - Histogram Of DI 3 Years afier PM ........................................................... 103
Figure A.12 - Histogram of D1 4 Years afier PM ........................................................... 103
Figure A.13 - Histogram of D1 5 Years after PM ........................................................... 104
Figure A.14 - Histogram of D1 6 Years after PM ........................................................... 104
Figure A.15 - Histogram of D1 7 Years after PM ........................................................... 105
Figure A.16 - Histogram of DI 1 Years after PM ........................................................... l06
Figure A.17 - Histogram of D1 2 Years after PM ........................................................... 106
Figure A.18 - Histogram of DI 3 Years afier PM ........................................................... 107
Figure A.19 - Histogram of D1 4 Years afier PM ........................................................... 107
Figure A.20 - Histogram of D1 5 Years after PM ........................................................... 108
Figure A.21 - Histogram of D1 6 Years after PM ........................................................... 108
Figure A.22 - Histogram of D1 7 Years alter PM ........................................................... l09
Figure A.23 - Histogram of D1 1 Years afier PM ........................................................... 110
Figure A.24 - Histogram of D1 2 Years after PM ........................................................... 110
Figure A.25 - Histogram of D1 3 Years after PM ........................................................... 111
Figure A.26 - Histogram of D1 4 Years after PM ........................................................... 111
Figure A.27 - Histogram of D1 5 Years afier PM ........................................................... 112
Figure A.28 - Histogram of D1 6 Years after PM ........................................................... 112

xi

Figure A.29 - Histogram of D1 7 Years after PM ........................................................... 113

Figure A.30 - Histogram of D1 1 Years after PM ........................................................... 114
Figure A.31 - Histogram of D1 2 Years after PM ........................................................... 114
Figure A.32 - Histogram of D1 3 Years after PM ........................................................... 115
Figure A.33 - Histogram of D1 4 Years after PM ........................................................... 115
Figure A.34 - Histogram of D1 5 Years after PM ........................................................... 116
Figure A.35 - Histogram of D1 6 Years afier PM ........................................................... 116
Figure A.36 - Histogram of D1 7 Years after PM ........................................................... 117

Figure 8.1- Histogram of DI 1 Year after PM when the pre-existing condition is less than
the guideline value .................................................................................................. 119

Figure B.2 - Histogram of D1 1 Year afier PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 119

Figure 8.3 - Histogram of D1 2 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 120

Figure 8.4 - Histogram of D1 2 Years after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 120

Figure 8.5 - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 121

Figure B.6 - Histogram of D1 3 Years afier PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 12l

Figure B.7 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 122

Figure B.8 - Histogram of D1 4 Years afierPM when the pre-existing condition is greater
than the guideline value .......................................................................................... 122

Figure 8.9 - Histogram of DI 5 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 123

Figure B.10 - Histogram of D1 5 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 123

xii

Figure 3.11 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 124

Figure B.12 - Histogram of D1 6 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 124

Figure B. 1 3 - Histogram of DI 8 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 125

Figure B.14 - Histogram of D1 8 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 125

Figure B. 1 5 - Histogram of DI 1 Year after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 126

Figure B.l6 - Histogram of DI 1 Year after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 126

Figure 3.17 - Histogram of D1 2 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 127

Figure B.18 - Histogram of D1 2 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 127

Figure B.19 - Histogram of D1 3 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 128

Figure 8.20 - Histogram of D1 3 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 128

Figure 3.21 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 129

Figure 8.22 - Histogram of DI 4 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 129

Figure 3.23 - Histogram of DI 5 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 130

Figure 8.24 - Histogram of DI 5 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 130

Figure B.25 - Histogram of D1 7 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 131

xiii

Figure B.26 - Histogram of DI 7 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 131

Figure 8.27 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 132

Figure B.28 - Histogram of D1 1 Year after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 132

Figure B.29 - Histogram of DI 2 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 133

Figure B.30 - Histogram of D1 2 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 133

Figure B.31 - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 134

Figure B.32 - Histogram of D1 3 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 134

Figure 3.33 - Histogram of DI 4 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 135

Figure B.34 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 135

Figure B.35 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 136

Figure B.36 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 136

Figure B.37 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 137

Figure B.38 - Histogram of D1 6 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 137

Figure 8.39 - Histogram of DI 7 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 138

Figure B.40 - Histogram of D1 7 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 138

xiv

Figure 3.41 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 139

Figure 8.42 - Histogram of DI 1 Year after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 139

Figure B.43 - Histogram of D1 2 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 140

Figure 3.44 - Histogram of DI 2 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 140

Figure 8.45 - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 141

Figure 8.46 - Histogram of DI 3 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 141

Figure B.47 - Histogram of DI 4 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 142

Figure B.48 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 142

Figure B.49 - Histogram of DI 5 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 143

Figure 8.50 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 143

Figure B.51 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 144

Figure 3.52 - Histogram of DI 6 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 144

Figure 8.53 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 145

Figure B.54 - Histogram of D1 1 Year after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 145

Figure 8.55 - Histogram of DI 2 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 146

XV

Figure B.56 - Histogram of D1 2 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 146

Figure B.57 - Histogram of D1 3 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 147

Figure B.58 - Histogram of D1 3 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 147

Figure 8.59 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 148

Figure 8.60 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 148

Figure 8.61 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 149

Figure 8.62 - Histogram of DI 5 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 149

Figure B.63 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 150

Figure 8.64 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 150

Figure 8.65 - Histogram of DI 7 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 151

Figure B.66 - Histogram of D1 7 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 151

Figure C.1 - Histogram of DI 1 Year after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 153

Figure C.2 - Histogram of DI 1 Year after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 153

Figure C.3 - Histogram of DI 2 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 154

Figure C.4 - Histogram of D1 2 Years after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 154

xvi

Figure C.5 - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 155

Figure 06 - Histogram of D1 3 Years after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 155

Figure C.7 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 156

Figure C.8 - Histogram of D1 4 Years after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 156

Figure 09 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 157

Figure C.10 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 157

Figure C.ll - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 158

Figure C. 12 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 158

Figure C.13 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 159

Figure C. 14 - Histogram of DI 1 Year after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 159

Figure C. 15 - Histogram of D1 2 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 160

Figure C.16 - Histogram of D1 2 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 160

Figure C.17 - Histogram of DI 3 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 161

Figure C. 1 8 - Histogram of D1 3 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 161

Figure C. 19 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 162

xvii

Figure C.20 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 162

Figure C.21 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 163

Figure C.22 - Histogram of D1 5 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 163

Figure C.23 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 164

Figure C.24 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 164

Figure C.25 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 165

Figure C.26 - Histogram of D1 1 Year after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 165

Figure C.27 - Histogram of D1 2 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 166

Figure C.28 - Histogram of D1 2 Years alter PM when the pre-existing condition is
greater than the guideline value .............................................................................. 166

Figure C.29 - Histogram of DI 3 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 167

Figure C.30 - Histogram of D1 3 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 167

Figure C.31 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 168

Figure C.32 - Histogram of D1 4 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 168

Figure C.33 - Histogram of DI 5 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 169

Figure C.34 - Histogram of D1 5 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 169

xviii

Figure C.35 - Histogram of DI 6 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 170

Figure C.36 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 170

Figure C.37 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 171

Figure C.38 - Histogram of D1 1 Year after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 171

Figure C.39 - Histogram of D1 2 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 172

Figure C.40 - Histogram of D1 2 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 172

Figure C.41 - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 173

Figure C.42 - Histogram of DI 3 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 173

Figure C.43 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 174

Figure C.44 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 174

Figure C.45 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 175

Figure C.46 - Histogram of D1 5 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 175

Figure C.47 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 176

Figure C.48 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 176

Figure C.49 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 177

xix

Figure C.50 - Histogram of DI 1 Year after PM when the pre-existing condition is greater
than the guideline value .......................................................................................... 177

Figure C.51 - Histogram of D1 2 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 178

Figure C.52 - Histogram of D1 2 Years afier PM when the pre-existing condition is
greater than the guideline value .............................................................................. 178

Figure C.53 - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 179

Figure C.54 - Histogram of D1 3 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 179

Figure C.55 - Histogram of D1 4 Years afier PM when the pre-existing condition is less
than the guideline value .......................................................................................... 180

Figure C.56 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 180

Figure C.57 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 181

Figure C.58 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 181

Figure C.59 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 182

Figure C.60 - Histogram of DI 6 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 182

Figure C.61 - Histogram of D1 7 Years after PM when the pre-existing condition is less
than the guideline value .......................................................................................... 183

Figure C.62 - Histogram of D1 7 Years after PM when the pre-existing condition is
greater than the guideline value .............................................................................. 183

CHAPTER 1 - INTRODUCTION

1.1 Problem Statement

The highway system in the United States represents the single largest engineering
investment ever made in this nation. Maintaining the highway system is necessary in
order for this nation’s economy to continue to thrive and advance. Without efficient and
modern transportation systems, farm products would spoil in fields and industrial goods
would remain in the factories [57].

Unfortunately, highway systems are not built to last forever. They deteriorate and
disintegrate at an accelerating rate unless they are properly and continually maintained,
rehabilitated, redesigned, and reconstructed. The focus of road construction has shifted
from constructing new pavements to maintaining and rehabilitating the existing pavement
system since the highway system is basically complete [57].

The Michigan Department of Transportation (MDOT) has implemented a
successful preventive maintenance program that is aimed at maintaining the pavement
network by slowing the rate of deterioration and correcting minor pavement deficiencies.
This is accomplished by using treatments that correct pavement surface defects caused by
the environment and the paving materials [13]. Applying preventive maintenance
treatments to a pavement at the appropriate time can extend its service life for a number
of years at a relatively low cost as compared with the cost of rehabilitation.

The MDOT has existing guidelines for the various preventive maintenance fixes
in the form of limiting values of the Distress Index (DI), Ride Quality Index (RQI), and
Rut Depth beyond which, preventive maintenance fixes should not be applied. The

guidelines also include the expected life extension for the different preventive

maintenance fixes. The expected life extension values are based on the past experience
and engineering judgment of the MDOT. The research conducted in this study will add
engineering analysis to confirm the life expectancy values and guidelines for the various

preventive maintenances fixes.

1.2 Maintenance

Maintenance activities are generally divided into three categories: (1) Preventive,
(2) corrective, and (3) routine. Preventive maintenance includes those activities that
protect the pavement structure, hinder the rate of pavement deterioration, and provide a
smooth ride while not increasing the structural capacity of the pavement [1 l]. Corrective

maintenance are those activities performed to correct a specific pavement failure or areas

of distress [11].

1.2.1 Preventive Maintenance

Preventive maintenance includes such fixes as thin overlays, surface seals, and
crack sealing. Thin non-structural overlays and surface seals are those activities that
consist of asphalt and aggregate or asphalt alone applied to the pavement surface. “They
(1) rejuvenate or retard the oxidation of the asphalt surface; (2) restore skid resistance; (3)
seal fine cracks which have appeared at the surface; (4) prevent the intrusion of water
into the pavement structure through cracks; and (5) retard raveling [4].” The most
common types of overlays are non-structural bituminous overlays, surface milling with
non-structural bituminous overlay, and micro-surfacing. The most common type of

surface seals are chip seals and micro-surfacing. The most common type of crack sealing

is crack treatment and overband crack sealing. Each preventive maintenance activity will

be discussed further in Chapter 2.

1.2.2 Corrective Maintenance

Corrective maintenance mainly consists of patching, chip seals, and thin
bituminous overlays [4]. Only patching will be discussed in the following section, since
the latter treatments have been discussed above.

Patching is one of the most common methods of repairing localized areas of
intense cracking whether the cracking is load associated (fatigue), non-load associated
(environmental-transverse or construction-longitudinal) [4]. If the cracks have
deteriorated to the point of spalling then the spalled material must be removed and
replaced. Patching usually consists of full-depth patches.

Partial depth patches involve removing the surface layer and replacing it with Hot
Mix Asphalt (HMA). They are used to fix slippage cracking due to poor bond between
the HMA surface and the underlying layer or for shoving and corrugations [4]. Full-
depth patches involve replacing the entire asphalt concrete layer. They are used to fix

fatigue cracking and potholes.

1.2.3 Routine Maintenance

AASHTO defines routine maintenance as “the day-to-day maintenance activities
that are scheduled or whose timing is within the control of maintenance personnel [54].”
Some examples of routine maintenance include filling cracks on the pavement surface,

painting pavement markings or cleaning ditches.

1.3 Research Objective and Organization

1.3.1 Research Objective

The research objective is to use the wealth of Pavement Management System
(PMS) performance data gathered on two hundred forty (240) preventive maintenance
(PM) projects since 1992 to perform a reliability-based analysis for the purpose of
determining the life expectancy and the evaluation of the limiting values for the different
PM fixes. The pavement performance parameters that MDOT gathers are distress index
(DI), ride quality index (RQI), and rut depth. These performance parameters will be

discussed further in Chapter 3.

1.3.2 The Proposed Model

Reliability analysis is a statistical tool stemming from the fact that there is some
uncertainty and variability in modeling the behavior of a system. The objective of the
analysis is to insure some level of reliability in predicting performance. There are
numerous levels of uncertainties in the prediction of pavement deterioration. Reliability
analysis addresses three basic questions about the reliability of a system:

0 What are the possible outcomes?
0 What is the likelihood of each outcome?
0 What are the consequences of decisions based on the knowledge of the probability

of each outcome [55]?

The reliability-based model which uses the distress index as the main
performance measure allows for estimating the life expectancy for the following MDOT

fixes: Non-structural bituminous overlay, surface milling with a non-structural

bituminous overlay, single chip seal, multiple course micro-surface, and bituminous
crack seal. Once the life expectancy values are determined they can be compared to the
life extension values that the MDOT uses.

The next goal is to evaluate the effectiveness of the guideline values for the
preventive maintenance fixes. This is accomplished by knowing the pre-existing
condition of the pavement before the preventive maintenance fix was applied. The
evaluation of the DI guidelines was tested by looking at the mean distress indices of two
populations:

0 Population one: DI values less than or equal to the DI guideline value at
the time of the preventive maintenance.
0 Population two: DI values greater than the guideline value at the time of

the preventive maintenance.

1.3.3 Organization

This thesis is organized in seven chapters as follows:

Chapter 1 gives some background information on the research undertaken and
provides the problem statement and objectives of the research study.

Chapter 2 presents a literature review of existing preventive maintenance models
for selecting the appropriate fix, the timing of the fix, the life extension of the fix, and the
cost benefits from preventive maintenance.

Chapter 3 gives a brief overview of the MDOT Pavement Management System
(PMS) focusing on the type of data collected and the indexes used, while Chapter 4
describes the data extraction process and the development of the database used in this

research.

Chapter 5 describes the use of reliability analysis and hypothesis testing to
develop estimates of pavement life expectancy and evaluate the guideline values for the
preventive maintenance fixes.

Chapter 6 presents the results of this research and includes a comparison with
MDOT’s life expectancies and an evaluation of their PM guidelines.

Chapter 7 presents the summary of findings and provides recommendations for

future research.

CHAPTER 2 - LITERATURE REVIEW

2.1 Introduction

The objective of a preventive maintenance program is to protect the pavement
structure, reduce the rate of pavement deterioration and/or correct pavement surface
deficiencies [13].

Pavement maintenance can be organized into three activity groups: preventive,
corrective, and routine maintenance. According to AASHTO, “preventive maintenance
is a planned strategy of cost-effective treatments to an existing roadway system and its
appurtenances that preserves the system, retards future deterioration and maintains or
improves the functional condition of the system without (significantly) increasing
structural capacity [22].” Corrective maintenance are activities that must be done in
response to events beyond the control of the highway agency. Some events require
immediate response for safety concerns and hence, they cannot be scheduled [22]. Some
examples of corrective maintenance activities include pothole patching, removing and
patching pavement blowups, or unplugging drainage facilities. AASHTO defines routine
maintenance as “the day-to-day maintenance activities that are scheduled or whose
timing is within the control of maintenance personnel [54].” Some examples of routine
maintenance include filling cracks, painting pavement markings or cleaning ditches. If
delays in preventive maintenance occur, pavement defects and their severity level
increase so that when corrected the cost is much greater. As a result, the life-cycle costs

of the pavements will be considerably increased.

2.2 Facts about Preventive Maintenance

Below are some facts about preventive maintenance that have been extracted from

the literature:

0 Pavement Management Systems (PMS) have been around since the early
1970’s in the United States and internationally [11].

o The policy of preventive maintenance did not catch on in the US. until the
early 1990’s [11].

0 France in 1969 began reconditioning its 28,000 km of roads after adopting a
preventive maintenance policy [2].

o In 1991, the US. Congress amended Section 119 of Title 23, of the United
States Code with the Intermodal Surface Transportation Equity Act. This act
provided for federal-aid fund eligibility for preventive maintenance type
projects [13].

o In Michigan there is approximately 9,580 lane-miles of highway constructed
of flexible, rigid and composite pavements [l3].

0 Between 1992 and 1998, approximately 2,650 miles of trunk line pavement
have been treated with preventive maintenance [13].

0 Rehabilitation and Reconstruction costs 14 times more per lane-mile than
preventive maintenance projects [13].

0 Since 1997 the annual budget for the capital preventive maintenance program

has grown from $3,000,000 to $60,000,000.

0 An estimated savings of more than $700 million has been realized since
Michigan’s preventive maintenance program was implemented in 1992 as
Opposed to rehabilitating the pavements[l 3].

0 Highway agencies are now in a maintenance/rehabilitation mode of operation

since no new hi ghways/interstates are being constructed.

2.3 Asphalt Pavement Distresses

Some of the most common types of distresses in flexible pavements are listed
below:

0 Transverse Cracking,

0 Longitudinal Cracking,

0 Longitudinal Joint Deterioration,

o Alligator Cracking,

- Block Cracking,

o Rutting,

o Raveling & Weathering/ Segregation,

o Patching, and

o Roughness.
The causes of these distresses will not be discussed here, but they can be found in several

publications from various sources including the National Center for Asphalt Technology

[4]-

2.4 Preventive Maintenance Treatments

The MDOT uses a mix of fixes to benefit the highway network. The guidelines
for each preventive maintenance fix were obtained from the MDOT’S Capital Preventive
Maintenance Program Guide [22]. The following lists the MDOT’s preventive
maintenance fixes:

a Non-Structural Bituminous Overlay,

- Surface Milling with Non-Structural Bituminous Overlay,

0 Chip Seal,

0 Micro-Surfacing,

0 Crack Treatment,

0 Overband Crack Filling,

0 Bituminous Shoulder Ribbons, and

o Ultra-Thin Bituminous Overlay.
Non-Structural Bituminous Overlay
Description: A dense-graded bituminous mixture limited to 90-kg/m2-application rate.
Purpose: A non-structural bituminous overlay is the highest type of surface treatment
fix available in the Capital Preventive Maintenance Program (CPMP). It will provide
some protection to the pavement structure, slow the rate of pavement deterioration,
correct many pavement surface deficiencies, improve the ride quality, and add some
strength to the existing pavement structure.
Existing Pavement Condition: The existing pavement condition should exhibit a good
base condition and a uniform cross section. The visible surface distress may include

moderate raveling, longitudinal and transverse cracks and small amounts of block

10

cracking. Low associated distress may be present. The pavement should only have some
minor base failures and depressions. Table 2.1 gives performance criteria for distress
index, ride quality index, and rut depth as to when the non-structural bituminous overlay

should be applied.

Table 2.1 — MDOT Performance Threshold Values for Non-Structural Bituminous
Overlay (MDOT, 1998)

 

 

 

 

Pavement Minimum RSL
Tine (years) D.I. R.Q.I. Rut Depth
Flexible 3 <40 <70 <12mm
Composite 3 <25 <70 <l2mm

 

 

 

 

 

Performance: This treatment performs best on flexible pavement structures, but is also
applicable to composite pavements depending on the extent of the reflective cracking.
Table 2.2 gives the estimated life extension for the non-structural bituminous overlay for

flexible and composite pavements.

Table 2.2 - Estimated Life Extension for Non-Structural Bituminous Overlay
(MDOT, 1998)

 

 

 

 

Pavement Type Years:
Flexible 5 to 10
Composite 4 to 9

 

 

'The time range is the expected life extending benefit given to the pavement, not the anticipated longevity
of the treattnent.

Surface Milling with Non-Structural Bituminous Overlay

Description: The removal of an existing bituminous surface by the cold milling method
and the placement of a dense-graded bituminous mixture limited to 90-kg/m2 application
rate.

Purpose: In the CPMP, the cold milling operation has been used to: (1) correct specific

existing surface deficiencies, (2) correct the shape of the existing cross section, and (3)

ll

 

 

produce a more economical project as compared to a non-structural bituminous overlay
project. The non-structural bituminous overlay replaces the bituminous material that is
removed.

Existing Pavement Condition: The existing pavement should exhibit a good base
condition. The visible surface distress may include: severe surface raveling, multiple
longitudinal and transverse cracking with slight raveling, a small amount of block
cracking, patching in fair condition, debonding surface and slight to moderate rutting.
The cold milling operation is used to correct rutting in the existing bituminous surface
layer where the rutting is not caused by a weak base and when the condition of the
existing pavement has deteriorated to a point where it is not practical to correct the
rutting problem by a more economical treatment. The cold milling operation is also used
to remove an existing bituminous course that is debonding. Table 2.3 gives performance
criteria for distress index, ride quality index, and rut depth as to when the surface milling

with a non-structural bituminous overlay should be applied.

Table 2.3 — MDOT Performance Thresholds for Surface Milling with a Non-Structural
Bituminous Overlay (MDOT, 1998)

 

 

 

 

Pavement Minimum RSL
Type (years) D.I. R.Q.I. Rut Depth
Flexible 3 <40 <80 <25mm
Composite 3 <30 <80 <25mm

 

 

 

 

 

Performance: This type of type of treatment will protect the remaining pavement
structure, slow the rate of deterioration and improve the ride quality. Table 2.4 gives the
estimated life extension for surface milling with a non-structural bituminous overlay for

flexible and composite pavements.

12

 

Table 2.4 - Estimated Life Extension for Surface Milling with a Non-Structural
Bituminous Overlay (MDOT, 1998)

 

 

 

 

Pavement Type Years.
Flexible 5 to 10
Composite 4 to 9

 

 

'The time range is the expected life extending benefit given to the pavement, not the anticipated longevity
of the treatment.

Chip Seal

Description: A chip seal is the application of a polymer modified asphalt emulsion with
a cover aggregate. A single or a double chip seal can be used in the CPMP.

Purpose: A chip seal will seal and/or retard the oxidation of an existing pavement
surface, improve skid resistance of the pavement surface, seal fine surface cracks in the
pavement thus reducing the intrusion of water into the pavement structure, and will retard
the raveling of aggregate from a weathered pavement surface.

Existing Pavement Condition: The existing pavement condition should exhibit a good
cross section and a good base. The visible surface distress may include slight raveling
and surface wear, longitudinal and transverse cracks with a minor amount of secondary
cracking and slight raveling along the crack face, first signs of block cracking, slight to
moderate flushing or polishing and/or occasional patch in good condition. Table 2.5
gives performance criteria for distress index, ride quality index, and rut depth as to when

single and double chip seal should be applied.

Table 2.5 — MDOT Performance Thresholds for Chip Sealing (MDOT, 1998)

 

 

 

 

Pavement Minimum RSL
Type (years) D.I. R.Q.I. Rut Depth
. 5 (double) <30 (double)
Flexrble 6 (single) <25 (single) <54 <3mm
Composite Sldouble) <15 (double) <54 <3 mm

 

 

 

 

 

13

 

 

Performance: Since chip seals are used to seal the cracks and construction joints in the
pavement in lieu Of only extensive overband crack fill, the life expectancy may vary
based on reflective cracking. Table 2.6 gives the estimated life extension for single and

double chip seals for flexible and composite pavements.

Table 2.6 - Estimated Life Extension for Chip Seals (MDOT, 1998)

 

 

 

 

Pavement Type Years.
Flexible:
Single Seal 3 to 6
Double Seal 4 to 7
Composite:
Double Seal 3 to 6

 

 

rThe time range is the expected life extending benefit given to the pavement, not the anticipated longevity
of the treatment.

Micro-Surfacing

Description: Micro-Surfacing is a mixture of polymer modified asphalt emulsion,
mineral aggregate, mineral filler, water, and other additives, properly proportioned,
mixed, and placed on a paved surface.

Purpose: A single course of micro-surfacing will retard oxidation and improve skid
resistance in the pavement surface. Multiple courses of micro-surfacing is used to correct
certain pavement surface deficiencies including severe rutting, minor surface profile
irregularities, polished aggregate or low skid resistance and light to moderate raveling.
Micro-surfacing is typically used on flexible or composite pavements and can perform
satisfactory under all traffic volumes.

Existing Pavement Condition: The existing pavement condition should exhibit a
uniform cross section and a good base. The visible surface distress may include slight

raveling, rutting, minor surface irregularities, flushed or polished surface and/or moderate

14

 

raveling. Table 2.7 gives performance criteria for distress index, ride quality index, and

rut depth as to when single and multiple micro-surfacing should be applied.

Table 2.7 — MDOT Performance Thresholds for Micro-Surfacing (MDOT, 1998)

 

 

 

 

 

 

 

 

Pavement Minimum RSL
Type (years) D.I. R.Q.I. Rut Depth
. 5 (multiple) <30 (multiple)
Flexrble 10 (single) <15 (single) <54 <25mm
Composite 5 (multipfle) <15 <54 <25mm

 

 

Performance: This treatment corrects rutting, flushing and low fi'iction. A Micro-
Surface performs well on high volume roadways to correct the pavement surface
conditions described above. Table 2.8 gives the estimated life extension for single and

multiple micro-surfacing courses for flexible and composite pavements.

Table 2.8 - Estimated Life Extension for Micro-Surfacing (MDOT, 1998)

 

 

Pavement Type Years.
Flexible:
Single Course 3 to 5
Double Course 4 to 6

 

 

 

Composite: MDOT acknowledges that micro-surfacing will provide a life extension to a
composite pavement; however data is not available to quantify the life extension.

 

'The time range is the expected life extending benefit given to the pavement, not the anticipated longevity
of the treatment.

Crack Treatment

Description: Crack treatment consists of both crack sealing and crack filling. Crack
scaling is attained by the Cut and Seal Method. Crack filling is attained by the Overband
Crack Fill Method. The Cut and Seal Method consists of cutting the desired reservoir
shape at the working crack in the existing bituminous surface, cleaning the cut surfaces
and placing the specified materials into the cavity to prevent the intrusion of water and

incompressible into the crack. The Overband Crack Fill Method consists of cleaning the

15

 

non-working crack into the bituminous pavement surface and placing the specified
materials into and above the crack to substantially reduce the infiltration of water and to
reinforce the adjacent pavement.

Purpose: The purpose of sealing and filling cracks and construction joints in the flexible
pavement surface is to prevent water and incompressible materials from entering the
pavement structure.

Existing Pavement Condition: The existing bituminous surface should be a relatively
newly placed surface on a good base and with a good cross section. On a flexible base,
the bituminous surface should be two to four years old and on a composite pavement, one
to two years old. The visible surface distress may include: fairly straight open
longitudinal and transverse cracks with slight secondary cracking and slight raveling at
the crack face, and no patching or very few patches in excellent condition. Table 2.9
gives performance criteria for distress index, ride quality index, and rut depth as to when

a bituminous crack treatment should be applied.

Table 2.9 — MDOT Performance Thresholds for Bituminous Crack Treatment
(MDOT, 1998)

 

 

 

 

Pavement Minimum RSL
Type (years) D.I. R.Q.I. Rut Depth
Flexible 10 <15 <54 <3mm
Composite 10 <5 <54 <3mm

 

 

 

 

 

Performance: The effectiveness of the seal will greatly depend upon the width of the
crack being sealed and the movement of the pavement structure at the crack. Table 2.10
gives the estimated life extension for bituminous crack treatment for flexible and

composite pavements.

l6

 

Table 2.10 - Estimated Life Extension for Bituminous Crack Treatment (MDOT, 1998)

 

 

 

 

Pavement Type Years'
Flexible Up to 3
Composite Up to 3

 

 

jThe time range is the expected life extending benefit given to the pavement, not the anticipated longevity
of the treatment.

Overband Crack Filling

Description: Overband Crack Filling consists of cleaning the crack in the bituminous
pavement surface and placing the specified materials into and above the crack to
substantially reduce infiltration of water and to reinforce the adjacent pavement.
Purpose: The purpose of overband filling the cracks in the surface Of the bituminous
pavement is to prevent water and incompressible materials from entering the pavement
structure. This treatment is commonly used as a surface preparation for the Micro-
Surface and Chip Seal treatments. Use as a stand-alone preventive maintenance
treatment, due to excess wear or failure shall be limited to older pavements where the cut
and seal method is suitable.

Existing Pavement Condition: The condition of the existing bituminous surface
depends upon the other preventive maintenance treatment that Overband Crack Filling
treatment will be used as a surface preparation. Overband Crack Filling should be used
to fill all non-working cracks. Table 2.11 gives performance criteria for distress index,

ride quality index, and rut depth as to when overband crack filling should be applied.

Table 2.11 -— MDOT Performance Thresholds for Overband Crack Filling (MDOT, 1998)

 

 

 

 

 

 

 

Pavement Minimum RSL
Type (yea LS) DJ. R.Q.I. Rut Depth
Flexible 7 <20 <54 <3mm
Composite 7 <10 <54 <3mm

 

 

 

l7

 

Performance: This treatment will help extend the service life of the surface treatment it
is being used with as a pretreatment. Stand-alone overband crack filling will also extend
the life of the pavement structure. Table 2.12 gives the estimated life extension for

overband crack filling for flexible and composite pavements.

Table 2.12 - Estimated Life Extension for Overband Crack Filling (MDOT, 1998)

 

 

 

 

Pavement Type Years”
Flexible Up to 2
Composite Up to 2

 

 

'The time range is the expected life extending benefit given to the pavement, not the anticipated longevity
of the treatment.

Bituminous Shoulder Ribbons

Description: This work includes the construction of a new bituminous shoulder ribbon
where gravel shoulders exist or the removal and replacement of a deteriorated bituminous
shoulder ribbon.

Purpose: The purpose of a bituminous shoulder ribbon is: (1) to accommodate an
increasing encroachment of traffic, (2) to expedite runoff water from the traveled lane
pavement, (3) to provide other usage such as bicycle paths, (4) to reduce edge stresses
and edge and corner deflections by increased lateral support, and (5) to reduce the
development of pavement edge drop-offs.

Existing Pavement Condition: In order for this treatment to be used in the CPMP, the
condition of the adjacent pavement structure must meet the CPMP’s pavement condition
criteria. The design life of the shoulder ribbons should be equal to or less than the
Remaining Service Life (RSL) of the main line pavement.

Performance: Most shoulder deterioration is attributable to truck encroachment, water

intrusion in the longitudinal joint, use of lower quality materials, and inadequate

18

 

structural thickness. Field Observations have shown that shoulder distress is primarily
concentrated within 0.6 m from the traveled lane. The extension of pavement life will be

up to 3 years.

Ultra-Thin Bituminous Overlay '

Description: A dense-graded bituminous mixture limited to 49-kg/m2 application rate
and a maximum average thickness of 20 mm.

Existing Pavement Condition: The existing pavement condition should exhibit a good
base condition and a uniform cross section. The visible surface distress may include
slight raveling, minor surface irregularities, and slight polished surface. The cross
sections should be free of ruts or distortions. Table 2.13 gives performance criteria for
distress index, ride quality index, and rut depth as to when an ultra-thin bituminous

overlay should be applied.

Table 2.13 — MDOT Performance Thresholds for Ultra-Thin Bituminous Overlay
(MDOT, 1998)

 

 

 

 

Pavement Minimum RSL
Type (years) D.I. R.Q.I. Rut Depth
Flexible 7 <20 <54 <3mm
Composite 7 <10 <54 <3mm

 

 

 

 

 

Performance: This treatment performs best on surfaces that are distortion free and
exhibit very little crack sealing material that may bleed through the mat. Table 2.14
gives the estimated life extension for the ultra-thin bituminous overlay for flexible and

composite pavements.

19

 

Table 2.14 - Estimated Life Extension for Ultra-Thin Bituminous Overlay (MDOT, 1998)

 

 

 

Pavement Type Years“
Flexible 3 to 5"
Composite 3 to 5"

 

 

 

"It is acknowledged that an ultra-thin bituminous overlay will provide a life extension to
a pavement; however data is not available to quantify the life extension.

 

The time range is the expected life extending benefit given to the pavement, not the anticipated longevity
of the treatment.

2.5 Selecting Appropriate Treatment and Timing

Three factors affect the future performance of a highway; they are the
applicability of the treatment, application time, and quality of the maintenance it receives
[10]. It is understood that inexpensive preventive maintenance activities are more
economical than high cost rehabilitation/reconstruction treatments; engineers must
recognize the causes of pavement deterioration and apply appropriate treatments at the
right time during the pavement life.

The first factor that should be looked at is the applicability of the treatment. Each
preventive maintenance fix serves a certain purpose. If the appropriate fix is not applied,
the life extension of the preventive maintenance fix will be drastically shortened. Hicks,
Dunn, and Moulthrop have developed decision trees to determine the appropriate fix for
various distresses [11]. These distresses include roughness, rutting, cracking, structural
condition, and bleeding. The decision trees take into account several factors that cause a
pavement to deteriorate over time. Some of these variables are traffic, stability of the
pavement structure, types of cracking, and material related distresses. Thus a decision
maker can identify the distresses in the asphalt pavement and from that select an
appropriate maintenance strategy. Hicks, Moulthrop, and Daleiden developed a table for

selecting the appropriate treatment based on distress type, as can be seen in Table 2.15

20

 

[10]. From this table an engineer can identify the distress in the field and then select the

appropriate fix for the distress.

21

63358 35:8:me 338:8 x mm mar—s

303.3 oamtaoaqmux ADM

 

X

dam—96¢

 

XX

XX

BEBE

 

web—8.5 385:8...
a Escaacfi

 

Imago—Oahu oswwmm

 

Hausa

 

BEE
£58m - managed

 

X

X

Beam £3882
- mmofiwnom

 

«EEO
no ES

 

chm—BO

 

Em
EEO

 

Hum
230

 

Rom
bus—m

 

”Bantam
-822

 

Rom mom

 

BEES
Esau

mmobmma uGOEO>Mnm

 

 

mafiOflh—flOuF oogfiofiummz o>flflo>ohn~

 

 

 

A33 430 .335 $qu mmobma 32.5» How momwofibm ocean—8:82 oumtnoaaac - m L :0. 03a...

22

Hicks, Moulthrop, and Daleiden also developed a decision matrix to include the

effects of several variables in the selection process of the appropriate fix [10]. These

rating variables are determined based on the level of significance and then multiplied by a

scoring factor based on level of importance to give a total score for that variable. The

total scores are then summed together for each variable, and the same procedure is

repeated for different fix types. The fix type with the highest score will be the one to use

for that specific project.

Selection of the appropriate fix/treatment is a difficult task as numerous factors

affect the selection of the appropriate maintenance according to Table 2.16 [24].

Table 2.16 - Factors Affecting Preventive Mainenance Treatments

 

 

 

Type and extent of distress Traffic loading
Climate Existing pavement type
Cost of treatment Expected life

 

Availability of qualified contractors

Availability of quality materials

 

Time of year of placement

Pavement noise

 

 

Facility downtime

 

Surface friction

 

Once one understands the above factors and how they influence each potential treatment,

then the selection of the most cost effective treatment can be done.

The next critical element in a preventive maintenance program is the

determination of the application time of a given fix. This is illustrated in Figure 2.1 and

Figure 2.2 according to the MDOT [13]. A distress value of fifiy (50) points or less

indicates a pavement in “satisfactory” condition, while a distress value of more than fifty

(50) indicates a pavement in “unsatisfactory” condition. Figure 2.1 shows the pavement

life extension when a preventive maintenance treatment is applied at the appropriate time

to a pavement in “satisfactory” condition. Figure 2.2 shows the result when a preventive

 

maintenance treatment is applied to a pavement in “unsatisfactory” condition. From
these figures certain conclusions can be drawn. Treatments applied to severely distressed
pavements receive little benefit. However, if treatments applied to pavements with light
to moderate distress, the same treatment provides substantial benefit by significantly
extending the life of the pavement. Note that while the curves for Figure 2.1 and 2.2 are
linear, this may not be the case for most pavements. The figures are therefore just for

illustration purposes.

   

 

50

    

j-p————_—

urrent
‘ . —_———.
(ondtttor

  

Distress Index

 

Time

I
I I Life Extended I

Fix Life

L Design Life I

Figure 2.1 - Properly Applied Preventive Maintenance (Galehouse, 1998)

 

24

    
   

Current
, .—"T——"'
Condition

 

 

Distress Index
vi
3

A
\

 

0

 

Time
Fix Life

Figure 2.2 - Delayed Preventive Maintenance (Galehouse, 1998)

Figure 2.3 is an example of a decision making process that can be used in order to
determine the timing of a treatment for a specific project. Using the data from a
pavement condition survey on a scale of one to one hundred, threshold limits can be

determined to define when a treatment type should be optimally applied.

25

Routine
Mainlvnnnr'e

 

‘mvmm Maintenance
Defer Action

Rulmbilitaliun

 

 

 

 

 

 

 

 

 

 

 

7.....7...---..._w..-__._---_-__-__-.._.1--_-....,-._-__... L......

20-9"“

L Ti_me

Figure 2.3 - Conceptual Relationship for Timing of Various Maintenance and
Rehabilitation Treatments (Hicks et. al., 1998)

Pavement. Condition Index

 

In summary, factors affecting the performance of a pavement that is subjected to
preventive maintenance include the timing of the preventive maintenance activity,
selection of the appropriate maintenance fix, and the construction problems and quality of
materials. If all these factors are taken into account, then the appropriate fix will provide

a smooth, high-quality facility that the public expects.

2.6 Performance Findings from Previous Studies

Various studies have been done to determine the life extension and optimal timing
of preventive maintenance fixes. The models used to determine the life extension and
Optimal timing include: survival modeling, probabilistic modeling, regression and other

statistical analyses, and field reviews.

26

2.6.1 Survival Modeling

A survival analysis of SPS-3 sites in the southern Long Term Pavement
Performance (LTPP) region was conducted in 1999 [9]. The Obj ectives of the study were
to determine the following:

o The fix treatment life expectancy of the treatment (i.e. median survival
time),

o The timing of the treatment application (i.e. was the pre-existing condition
of the pavement in poor, good, fair, or excellent condition), and

o The benefit of the treatment in terms of life extension as compared with
the “do nothing” option.

The Kaplan-Meier method, which is a nonparametric survival analysis technique,
was used in determining the life expectancy and life extension. The survival probability
for each fix can be estimated by using the calculated survival time on those pavements in
which a fix was applied and then allowed to deteriorate until a poor condition is assessed.
The failure probability is calculated as 1-P[survival]. Therefore the failure probability is
a function of age only. The six year failure probability and the median survival times for
the different treatments are shown in Table 2.17. The added life is calculated by taking
the difference between the median survival time for the treatment and the median

survival time for the control curves corresponding to the same original condition.

27

$2535 on 8 voflnfioo
2:: 3233 5:92: 05 2 at? 30:33 a who» no 59:5: 05 m_ 82:89: 2. 3 coax—Boo “meson 5:82..

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B25
tongue 83 Eon—«RE 05 £033 9 8288 05 mo $8 :8: who» he Bun—Es: 05 mm 08: Etta 8:32.
o 32 $2 mm Loom
m._ 32 «<2 mm ham Rom QED
Wm 32 $2 mm @000
o mud mud 2: Boom
mg Em NS 3 ham Bow x080
Wm L We cm @000
o Wm Wm 2: Doom
m4 m.m m hm BBQ Hum Ea
Wm N m6 xv Boom
o Wm Na 9: Loom
2 3. 3 cm can 335
m m N m m 5 mm vooO
:mmnotoom
.95:er .AEaSC
30839;. 2.253;. on AJMHWWEHEH b AR;
. m Bannock 5.56.50
oz .53 3 vegan—BOO E: o 9:. 3 «Em... «5.5.3:.
anaem— aaGo—z .u S _. h _ . . O
25,—. _a>_>.=_m . «uncut». 03.3.5. aao>é
5:52 <

 

 

 

 

 

 

 

3.. .s afiéa Baas... 335m 3% SEE 805:8 92: co 338m - SH 2x3

28

Elthahan et. al showed that after six years of treatment, sections with a poor
condition had a probability of failure of 83% whereas those with a fair or good original
condition had a probability of failure of 38% and 37%, respectively. The median survival
times (life extension) for thin overlays, slurry seals, and crack seals were 7.0, 5.5, and 5.0
years, respectively. The chip seal sections did not reach the 50 percent failure probability
after eight years of experiment. Therefore chip seals were concluded to have

outperformed thin overlays, slurry seals, and crack seal treatments.

2.6.2 Regression Modeling

Morian et. al. used SPS-3 performance data and applied regression analyses in a
1998 study [51]. The data analyzed included distress, deflection, roughness, rut depth,
and friction data. Prediction models using multiple regression were developed for
cracking, rutting, roughness, friction, and an index called the Pavement Rating Score
(PRS). The purpose of this study was to:

0 Determine the effective timing of the various treatments,

0 Evaluate the effectiveness of the different treatments, and

0 Share information and experience among the profession.

Results of this study are summarized below:

0 The structural adequacy was not found to have a significant effect on the
performance of the SPS-3 treatments. Structural adequacy is “the actual
structural number of the test section divided by structural number
requirements to carry the section traffic volume.” This means that “the
pavements with inadequate structure performed as well, or as poorly, as

those with adequate structure.”

29

c There was an immediate reduction in rutting for the thin overlay treatment.
Slurry seal sections rutted at a lower rate and chip seal sections at a
slightly faster rate compared to the control section.

0 Thin overlay achieved significant reductions in roughness; chip seals and
slurry seals had a slight reduction, and crack seals did not reduce
roughness.

0 Composite regression curves (good, fair, and poor curves combined into
one curve) were developed for each fix type and climatic region. This will
allow for the determination of the life extension for each fix based on a
threshold for the PRS.

Raj agopal et. al. looked at how timing and level of maintenance affect the
deterioration rate and pavement life-cycle for three maintenance treatments [15]. They
developed nonlinear regression models using pavement age, cumulative equivalent
single-axle loads, composite structural number, life-cycle before overlay and thickness of
the proposed treatment to determine the pavement condition rating (PCR) at some time
(t). A relationship that included the preexisting condition of the pavement was also
developed. From the regression equation, if one enters the PCR before the preventive
maintenance action then one can estimate the PCR after the preventive maintenance
action. Performance curves can then be developed showing the PCR versus age, and
from these one can make a decision as to when the preventive maintenance should be
applied.

Sebaaly et. al. developed a model that relates the present serviceability index

(PS1) to age, material properties, traffic loading, and climate [18]. These performance

30

models were for the flexible pavement maintenance treatments commonly used by the
Nevada Department of Transportation (N DOT). For each NDOT district and fix type the
actual PSI was compared with the PSI calculated by the developed model for validation
since the model is only applicable to the range of values it was developed from. The
majority of the models had R2 values greater than 70% which indicates a good fit
between the model and the data. Once the model is validated it can be used as a
preventive maintenance tool to plan ahead for fiiture activities.

Kuo et. al. looked at the projected rate of deterioration of pavement segments in
Florida due to traffic, speed limit, rainfall, temperature, work-mix, historical condition
ratings, distress types, pavement thickness, age, and most recent year-to-year rating
decline [56]. An analysis was made to determine the appropriate response variable and
predictor variables in the prediction model with statistical tests conducted to finalize the
predictor variables for the model. The validity of the prediction model was compared
with the predicted condition ratings. The model solves for the number of years until
failure based on cracking given the amount of cracking at the current year. Once the
model is validated it can be used as a preventive maintenance tool to plan ahead for

future activities.

2.6.3 Statistical Modeling

Hall et. al. used Dunnet’s method of Multiple Comparison with Control (MCC) to
analyze LTPP data from the SPS-3 experiment. The following objectives were sought
[47]:

o The initial effects, if any, of the fix type on the pavement condition,

31

o The long term effects, if any, of fix type on the performance of the
pavement,

o The influence, if any, of the pre-existing condition of the pavement, and

o The relative effectiveness of the different preventive maintenance fixes.
The pavement condition measures were roughness (IRI), rutting, and fatigue cracking.
The Dunnet’s MCC method is a statistical test that defines a confidence interval for the
difference between the observed values in the treated section versus the observed value in
the control section. If zero is in the confidence interval then there is a significant effect
of the treatment on the change in the pavement condition. This analysis was used for
each fix type and for analyzing each objective using the pavement condition measures
outlined above. The treatments used in this experiment were thin overlay, chip seal,
slurry seal, and crack seal.

This study resulted in the following conclusions. In terms of roughness, rutting,
and fatigue cracking, the most effective maintenance treatment in the SPS-3 experiment
was a thin overlay followed by the chip seal treatment and then the slurry seal treatment.
The thin overlay treatment was the only fix type to produce a small reduction in
roughness and the only one to have a significant effect on long-term roughness relative to
the control sections. For very rough pavements, chip seal and slurry seal had a small
effect on long-term roughness relative to the control sections. Crack seals did not have a

significant effect on long -term rutting, roughness, or fatigue cracking.

2.6.4 Probabilistic Modeling-Markovian Chains

The Markov Model is a very powerful probabilistic method that can be used as a

decision making tool for determining optimal maintenance and repair strategies. Butt et.

32

al developed a Markov model that related pavement condition index (PCI) to pavement
age [39]. Li et. al. looked at the deterioration of a pavement (pavement condition state
(PCS)) over time due to traffic and climate [17].

The general process of a Markov model is as follows: There is a certain number
of pavement sections in different “states/conditions” over time. These sections
deteriorate into different “states” each year. This results in a Transition Probability
Matrix (TPM). Each matrix element is the probability of a pavement being in that “state”
the ensuing year. The validity of the model can be accomplished by plotting the actual
PCI curve compared with the predicted PCI curve. The Markov process can be used in

determining optimal pavement strategies for all pavement sections in the network.

2.6.5 Engineering Field Review of SPS-3 Projects

In the summer and fall of 1995, four Expert Task Groups (ETG’S), one in each
LTTP Region, visited and performed site reviews on fifty-seven SPS-3 sites [38]. The
ETG members rated the pavements based on overall pavement condition irrespective of
fix type, the overall condition of the fix types, the overall effectiveness of the fix types (in
terms of the amount of distresses present), and whether or not the fix was appropriate.
After analyzing the data the ETG’s concluded that afier five years of field performance
the treatments have reduced the presence of cracking. The best performing treatments

were thin overlays and chip seals as compared to the slurry seal and crack seal.

2.6.6 Engineering Field Review of MDOT’s CPM Program

In 1999, 2000, and 2001 B.T. Bellner and Associates conducted a field review

and data base study on the MDOT’s Capital Preventive Maintenance (CPM) Program

33

[50]. The purpose of this study was to assess the overall effectiveness of the CPM
Program and to make any improvements, if necessary. The field review was conducted
to see if the CPM work met the warranty criteria and to evaluate the overall condition of
the pavement surface such that a comparison can be made with the DI calculated in the
MDOT’s PMS database.

The consulting firm reviewed and reported on the following flexible preventive
maintenance treatments: Single chip seal, sealing cracks in bituminous pavements, non-
structural bituminous overlay, surface milling with non-structural bituminous overlay,
micro-surfacing, overband crack sealing, double chip seals, ultra-thin bittnninous
overlays, flexible designed micro-surfacing and hot in-place bituminous recycling. Each
project for each fix type was reviewed; field data were collected independently, and a
report was written. After assessing all the projects, conclusions were drawn for each fix
type.

B.T. Bellner and Associates reported that, overall the CPM Program had been
successful in extending the life of the pavements. The majority of the projects were
constructed prior to the CPM Program Guidelines in 1998. The projects were found to be
cost-effective by extending the life of pavements at lower costs. The report did not rate

the different fixes relative to each other.

2.7 Conclusions

The objective of this research is to determine the life expectancy of the preventive
maintenance fixes and to evaluate the PM guidelines based on PMS data from the
MDOT. From this literature review it can be concluded that the majority of the pavement

performance models predict some future performance of the pavement, which can then be

34

used to determine whether or not a preventive maintenance action should be taken. Only
the survival analysis model by Elthahan, et. al. determined the life extension for chip seal,
crack seal, slurry seal, and thin overlay treatments. It has been determined that
preventive maintenance is cost effective and extends the life of pavements. The majority
of the authors in the bibliography agree that the Optimal timing of preventive
maintenance treatments is when the pavement is in good to fair condition. When the
pavement is in poor condition, the increased risk of failure is two to four times higher
according to Elthahan et. al.

Preventive maintenance is a tool that the highway agencies can use to improve the
quality of the pavement network while reducing expenditures. This is based on the fact
that preventive maintenance is more economical than the higher cost treatments of
rehabilitation/reconstruction. If the do-nothing option is chosen the pavement will slowly
deteriorate beyond the point where a maintenance treatment will be of any economical
benefit. By applying the appropriate maintenance at the right time (pavement in
relatively good condition) during the pavement life, it will increase the quality Of the

pavement network by improving its serviceability level.

35

CHAPTER 3 - THE MDOT PAVEMENT PERFORMANCE MEASURES

3.1 Introduction

In order for highway agencies to manage their pavements and to provide a smooth
ride, anticipate future routine maintenance, and correct pavement deficiencies, they must
collect pavement condition data of the entire pavement network.

The MDOT collects pavement surface distress data, longitudinal pavement profile
data, and rut depth for every tenth-of-a-mile along the pavement network under its
jurisdiction. Friction data are collected on a per need basis and are kept in the
appropriate region. The data are collected on one-half of the network every year.
Therefore data for the entire network are collected every two years. The data are

described in the ensuing sections.

3.2 Surface Distress Data

The pavement distress survey in Michigan is accomplished by videotaping fifty
percent of the network each year. The videotaping is carried out by a private contractor
and administered by the MDOT central office. The videotapes are reviewed one frame
(each frame contains ten feet of pavement) at a time in the central office. During the
review, pavement distresses are documented for each frame. Hence, the MDOT distress
data are detailed and can be used at both the network and project levels. The pavement

distresses collected by the MDOT are listed in Table 3.1.

36

Table 3.1 - Pavement Distresses Collected by the MDOT

 

 

 

 

Pavement Type
Flexible Rigid Composite
Transverse tears Transverse joints Transverse tears

Transverse cracks
Longitudinal cracks
Alligator cracks
Block cracks
Patches

Raveling

Flushing

Intensive miscellaneous
cracks

Rut depth

Ride Quality

 

Transverse Cracks
Intensive transverse cracks
Longitudinal cracks
Delamination
Reactive aggregates
High steel

Mudj acked areas
Patches

Corner Breaks
Popouts

Scaling

Ride Quality

 

Transverse cracks
Longitudinal cracks
Block cracks

Patches

Raveling

Flushing

Intensive reflective cracks
Rut depth

Ride Quality

 

The distress data are than grouped into 0.1-mile long unit sections. The pavement

management system (PMS) databank contains detailed data for each type of pavement

distress, severity, and extent for each 0.1-mile section.

The MDOT pavement management group has developed a rating system whereby

each type of principal distress and its associated distress are ranked and assigned

‘Distress Points’ (DP) based on their impact on pavement performance and on

experience. For any pavement section, the Distress Index (DI) can be calculated as the

sum of the distress points along a section normalized to the section length as stated in

equation 3.1.

_ ZDP
L

DI

where: D1 = Distress index,

(3.1)

)3 DP = Sum of the distress points along the pavement section, and

L =Number of 0.1-mile pavement sections.

37

 

The DI scale starts at zero for a pavement in perfect condition and increases (without
bound) as the pavement condition worsens. The MDOT categorizes DI into three levels

as shown in Table 3.2.

Table 3.2 - Distress Index Categories

DI Level
Low < 20

Medium 20-40
' >40

 

A pavement with a D1 of fifiy or higher is considered to be in unacceptable
condition and has therefore exhausted its service life. This DI-threshold value is based

on historical pavement performance.

3.3 Longitudinal Pavement Profile Data

The longitudinal pavement profile is collected by the MDOT using a rapid inertial
profiler. The system, which is composed of laser sensors mounted on a utility vehicle,
measures the longitudinal profile of the road and records the data in three-inch
increments. An onboard computer analyzes the data for each 0.1-mile of pavement and
calculates the International Roughness Index (IRI) and the Ride Quality Index (RQI).
The former is calculated to satisfy the Federal Highway Administration requirements and
the latter is for use by the MDOT. The RQI is a weighted sum of the variances in
elevation from three wavelength bands (short, intermediate, and long waves). The RQI
has been calibrated to relate with the human perception of ride. The RQI scale starts at
zero for a perfectly smooth pavement and increases (without bound) as the pavement gets
rougher. Higher RQI implies lower ride quality. The RQI scale is subdivided into

various categories as shown in Table 3.3.

38

Table 3.3 - Ride Quality Categories

 

 

 

 

 

 

RQI Ride Quality Categories
0 to 30 Excellent
31 to 54 Good
55 to 70 Fair

>70 Poor

 

 

 

3.4 Rut Depth

Pavement rutting is also collected by the MDOT. The average rut depth for each
0.1-mile section is calculated and stored in the data bank. Unfortunately rut depth will
not be used in the analysis since the MDOT database has only two years of data for this

performance measure (1998 and 1999).

3.5 Friction Data

Pavement friction data are collected as requested by the different regions. The

friction data are then kept in that region. Friction data will not be used in the analysis.

3.6 MDOT’s Definition of Life Extension

According to the MDOT, the life extension from a preventive maintenance
treatment is defined as follows:

Life Extension = RSLz - RSL] (3.2)
Where RSL; is the remaining service life of a pavement section at the time of the
preventive maintenance fix and RSL; is the remaining service life of the pavement
section after being treated by the preventive maintenance fix.

The remaining service life (RSL) of a pavement section at a given time is defined
by the MDOT as the time duration for a pavement section to reach the threshold distress

index value of fifty. This can be estimated by extrapolation using some performance

39

model with time. The MDOT uses a logistic growth ftmction as their DI performance

curve.

40

CHAPTER 4 - EXTRACTION AND DEVELOPMENT OF DATABASE

4.1 Introduction

The distress index and ride quality index data were acquired from the MDOT
PMS database. The data were given in 0.1-mile increments for each project. A list of
preventive maintenance projects was compiled by the PMS group at the MDOT from
1992 to 2001. The projects chosen for this research received a PM action sometime
between 1993 and 1996. These years were selected because it would be possible to get
some insight as to how the pavement performed prior to the fix and how the pavement
performed after the fix. The pavement performance several years after the fix would
allow us to compare the life expectancies calculated by the reliability analysis with

current MDOT’S life extension estimates for different preventive maintenance fixes.

4.2 Extraction of Data

The different projects were grouped according to fix type: 1) Non-Structural
Bituminous Overlay; 2) Surface Milling and Non-Structural Bituminous Overlay; 3)
Single Chip Seal; 4) Multiple Course Micro-Surface; and 5) Bituminous Crack Seal.
Pavements where the D1 or RQI decreased with time were not used in the analysis. The
projects where the D1 or RQI decreased with time were excluded because pavements
performance should decrease with time, therefore it was assumed that some sort of action
was taken to improve the condition of the pavement surface. This analysis looks at
pavements that have had only one PM action taken. Also, the DI for pavements that had
a subsequent preventive maintenance fix after the initial PM fix were not used because

this project looks at the effect of single PM treatments and not multiple fixes. There were

41

not enough data available to look at the effect of applying a mix of PM treatments to
pavement surfaces. Figure 4.1 and 4.2 illustrate pie charts with the number of projects

used for each fix type for DI and RQI, respectively.

 

 

D Non-structural bituminous
overlay

ISurface milling and non-
structural bituminous
overlay

I Single Chip Seal

I Multiple Course Micro-
surfacing

III Bituminous Crack Seal

 

 

 

 

 

 

Figure 4.1 - Number of Preventive Maintenance Projects with Distress Index Data

 

 

El Non-structural bituminous
overlay

ISurface milling and non-
structural bituminous
overlay

I Single Chip Seal

I Multiple Course Micro-
surfacing

El Bituminous Crack Seal

 

 

 

 

 

 

Figure 4.2 - Number of Preventive Maintenance Projects with Ride Quality Index Data

42

The automated PMS data collection at the MDOT originated in 1992. Therefore
using projects from 1993 to 1996 would give the pre-existing condition of the pavement
as well as the condition one to eight years afier the preventive maintenance fix. Projects
from 1997 and later would not provide additional insight into the life expectancy of the
preventive maintenance fixes due to the lack of data available after the preventive
maintenance action.

The first step in the research was to obtain a list of the preventive maintenance
fixes that were done from 1992 to 2001. This database included all relevant information
including year of preventive maintenance, route number, control section, beginning mile
post, ending mile post, job number, cost of activity, and fix type. The data were filtered
in order to obtain a set of projects from 1993 to 1996 that have had only one preventive

maintenance treatment applied to them.

4.3 Development of Database

The next step was to develop a database by extracting the appropriate data from
an individual project data file. This was done according to the beginning and ending mile
posts. Distress Index and Ride Quality Index data are in separate data files; therefore two
separate databases were compiled for each project. The surveyed distress index did not
necessarily match up from one survey year to the next and also did not necessarily line up
according to the project mileposts. Therefore the data had to be rearranged to correct for
this, and a new distress index had to be calculated based upon the actual project limits.

This new distress index was formulated based on some assumptions. The first
assumption is that the 0.1-mile distress index value is an average value over that 0.1-mile.

The second assumption is that the variation of the distress index between measured points

43

is piece-wise linear. This would allow us to use linear interpolation to calculate the
distress index at a new point along the project length, as can be seen in the schematic

diagram of Figure 4.3.

 

Dlxz D1x3

 

 

xz X3

 

DI'

 

 

 

XI

 

 

 

Figure 4.3 — Schematic drawing showing the above calculation

The equation used is as follows:

(x2 ‘x1)X(DIx2 - D1x3)

(x2 - x3)

 

131': D1,, - (4.1)

where:

DI’=New distress index calculated at a new milepost,
DIx2=Distress Index at milepost 2,

DIx3=Distress Index at milepost 3,

x1= Section Length (at every 0.1 mile),

xz=Survey length 2 (at every 0.1 mile), and

x3=Survey length 3 (at every 0.1 mile).

44

This calculation was not needed for the ride quality index because the ride quality

index was taken at the same point from year to year.

4.4 Compilation of the Data

The final step was to determine the DI and RQI values prior to the preventive
maintenance fix at year one, year two, year three, year four, year five, year six, year
seven, and year eight after the preventive maintenance fix. Determination of the
condition of the pavement was calculated by taking the average DI or RQI over 0.1-mile
sections within the project length for each survey year. This would allow us to
determine if the D1 or RQI increased or decreased from one survey year to the next. If
the pre-existing condition was unknown for the year the preventive maintenance was
done or decreased down to zero (i.e. the road was surveyed after the preventive
maintenance) the DI from the previous year would be used as the pre-existing condition.
The future performance of the preventive maintenance fix was calculated by using the DI
or RQI at year one, year two, year three, year four, year five, year six, year seven, and
year eight after the preventive maintenance fix. For example, if the preventive
maintenance action occurred in 1994 and the pavement was surveyed in 1992, 1994,
1996, 1998, and 2000, then one would have the pre-existing condition along with the DI

two years afier the fix, four years after the fix, and six years after the fix.

45

CHAPTER 5 - RELIABILITY BASED MODEL AND ANALYSIS

5.1 Introduction

The word reliability infers dependability, trustworthiness, and steadiness. The
mathematical definition of reliability is the probability that a system will not fail.
Therefore, a system is considered reliable until it fails. The reliability value can be
calculated as:

R=1-Pr(f) (5.1)

where Pr(f) is the probability of failure.

In the area of pavement performance, the American Association of State Highway
and Transportation Officials (AASHTO) defines reliability as “the probability that the
pavement system will perform its intended function over its design life and under the
conditions (or environment) encountered during operation. . .the probability that any
particular type of distress will remain below or within a permissible level. . .during the
design life [54].”

The term failure in engineering reliability refers to any occurrence of an adverse
event such as a column buckling during an earthquake or a pavement that is distressed
beyond a threshold value set by a highway agency. In this research study, the limiting
value for probability of failure is given in the MDOT capital preventive maintenance
program guidelines, which were described in Chapter 2. The use of reliability analysis
stems from the fact that there is uncertainty/variability in the distresses, climate,

construction, structural capacity, etc.

46

5.2 Reliability-Based Model

According to Pendleton, reliability-based models contain three sources of

variability [53]:

o Variability in the output variables,

0 Variability in the input variables, and

o Uncertainty in the model.
Figure 5.1 is a schematic diagram showing a general reliability modeling process. The
input side of the model may consist of known equations developed through mechanics or
physical laws of nature. The input model may also include a hypothesized probability or
model or it could be totally unknown in which nothing is known about the mathematical
model or the input variables. Whatever model a researcher chooses to use, one must
always realize that there is some error in the modeling process. The goal is to determine

what kind of inputs will produce the lowest amount of error and thus give reliable outputs

[53].

 

 

 

Output Input Error

 

 

 

 

 

 

 

 

 

Figure 5.1 - Schematic Showing the Reliability Model (Pendleton, 1994)

In order to explain the developed model, pavements that have been treated with
different preventive maintenance fixes will be used. Basically, an examination of how
the various treatments perform based on the distress index (DI) and ride quality index

(RQI) growth over time will be done. Some sources of variability include: traffic,

47

environment, subgrade, pavement thickness, pre-existing condition, drainage,
construction and material variability within fix types. For the most part, traffic (i.e.
equivalent Single axle loads, ESAL’s) is less variable than the other variables since the
preventive maintenance fixes are used mainly on low volume roads. One more source of
variability is in calculating the DI since individual interpretation of the severity and
extent of pavement distresses can vary. The goal is to have enough data where the
variability will be minimized for those variables that may lead to an increase in the error.

The first step in doing a reliability-based model is the need for a large data set.
The larger the data set is, the more defined the distribution becomes. In this research, the
data used were obtained from the MDOT pavement management system (PMS). The
MDOT uses four pavement performance parameters: Distress index, ride quality index,
rut depth, and friction. This research project Specifically looks at the distress index and
ride quality index on pavements that have had a preventive maintenance activity done
during the period from 1993 to 1996. The MDOT has distress index data from 1992 to
2001 and ride quality index data from 1992-1999. This will allow the examination of the
pre—existing condition to see if it has a significant effect on future pavement condition. It
will also provide up to eight years worth of performance data afier the treatment.

Step two is to separate the projects into different preventive maintenance fix types
so their effectiveness can be evaluated separately. The distress index and ride quality
index for different years after the fix are used to monitor the performance of the fix. The
distress index and ride quality index prior to the treatment are used to look at the effect of

pavement condition prior to the fix on pavement performance after the fix.

48

Step three is to plot probability distributions of D1 for each year after the
treatment for each fix type. Frequency distributions of DI for each year after the
treatment were plotted using the actual data. The reliability can then be calculated based
on the MDOT’S preventive maintenance guidelines. These distributions can be seen in
Appendix A. Figure 5.2 is an example showing how the reliability is calculated
graphically. The reliability based on a D1 threshold of forty (40) is the area under the

curve to the left of the limiting value.

 

 

 

 

 

 

DI Threshold]

 

 

 

 

 

 

1 0 10 20 30 40 50 60 70 80 90 100110120
I DIAROI'PMatt.

 

 

Figure 5.2 - Example Calculation for Reliability Analysis

The final step is to calculate the reliability for each year after the treatment. A
table can then be produced showing the reliability values for each fix type at different
years after the treatment. This will allow an engineer to estimate the life expectancy of a
given fix using an acceptable reliability level. For example, the AASHTO recommends

different reliability levels depending on the type of road and traffic volume, as shown in

Table 5.1.

49

Table 5.1 - Suggested Levels of Reliability for Various Functional Classifications
(AASHTO, 1986)

 

 

 

 

 

 

 

Recommended level
Functional classification of reliability (%)
Urban Rural
Interstate and other
freeways 85-999 80-999
Principal arterials 80-99 75-95
Collectors 80-95 75-95
Local 50-80 50-80

 

 

 

 

5.3 Reliability Analysis

The projects that were selected for the analysis were those preventive
maintenance projects that were completed in 1993, 1994, 1995, and 1996. All the
information (e. g. control section, project mileposts, and fix type) was given by MDOT so
the appropriate project could be selected for the analysis. As described in Chapter 3,
performance data in the form of D1 and RQI values were available at 2-year intervals for
0.1-mile pavement sections. The preventive maintenance fixes that were looked at
included: Non-structural bituminous overlay, surface milling and non-structural
bituminous overlay, single chip seal, multiple micro-surfacing, and bituminous crack
treatment.

A reliability analysis was performed on each fix type in order to determine the life
expectancy for each treatment. Frequency distributions of D1 at different years after the
preventive maintenance activity were plotted for each fix type. Based on the actual
distributions of D1 values, the reliability was calculated as the number of the sections that
have distress indices that are less than the DI guideline value divided by the total number

Of sections surveyed for that year. Table 5.2 shows the number of data points

50

corresponding to 0.1-mile pavement sections for each fix type at different years after the

preventive maintenance activity.

Table 5.2 - Number of Data Points for Each Fix Type

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Time After PM Fix
Year Year Year Year Year Year Year Year
Fix Type 1 2 3 4 5 6 7 8
Bituminous
Overlay 945 1261 879 862 439 344 159 105
Mill & Fill 210 566 169 407 88 17 9 0
Chip Seal 1775 1276 1312 823 1269 559 306 0
Micro-Surface 1015 1363 913 976 742 539 1 1 1 0
Crack Seal 1877 2596 1360 2472 1370 1094 628 83

 

 

5.4 Two-Sample t-test for Significance

Knowing when to apply the preventive maintenance activity is critical for a
pavement management team. If the treatment is applied too early, its effect on pavement
performance may not be as beneficial. On the other hand, if it is applied too late, the fix
will not last as long, thus reducing the economic benefit. Therefore there is a need for
evaluating the guideline limits at which a preventive maintenance fix should be applied.
In order to accomplish this, a two sample t-test was used on two populations:

0 Population One: Those pavements whose pavement performance
parameter was less than the limiting value at the time of the preventive
maintenance activity.

0 Population Two: Those pavements whose pavement performance
parameter was greater than the limiting value at the time of the preventive
maintenance activity.

The goal of the two sample t-test is to compare the response in two groups

assuming that the response in each group is independent of that in the other group. In

51

order to use the t-test on sample sizes that are not equal, the data should be approximately
normally distributed and meet the following criteria for sample size populations [27]:

0 Sum of Sample Sizes is less than 15: Use t-test only when the data is
normally distributed. If the data is clearly non-normal or outliers are
present, do not use this statistical test.

0 Sum of Sample Sizes is greater than 15 and less than 40: t-test can be
used except when there are strong outliers or strong skewness.

0 Sum of Sample Sizes greater than or equal to 40: t-test can be used even

for distributions that are skewed.

5.4.1 Pavement Condition Using Distress Index (DI)

A two sample t-test was conducted in order to evaluate the DI guidelines that the
MDOT imposes for the different preventive maintenance fixes. Under consideration will
be the mean distress index for two populations:

0 Population One: DI values less than or equal to the DI guideline values at the

time of the preventive maintenance activity.

0 Population Two: DI values greater than the guideline values at the time of the

preventive maintenance activity.
For each treatment, mean distress indices at subsequent years were analyzed using this
method. Below are the formulated null and alternative hypotheses:

0 Ho: Mean DI at time (ti) of population one equals mean DI at time (ti) of

population 2 or

0 HA: Mean DI at time (ti) of population one does not equal mean DI at time (ti)

of population 2.

52

The next step is to select the appropriate level of significance (1 -oc). An alpha
value (or) of five percent (5%) was chosen in this case, which corresponds to a 95% level
of significance. This level of significance is the most common one used in statistics.
This value is then compared with the computed p-value from the t-test. If the computed
p-value is less than the level of significance we then state that the data is significant at
that level (l-a). The smaller the p-value, the stronger evidence against the null
hypothesis is provided by the data.

One can either accept or reject the null hypothesis based on comparing the
computed estimate of the test statistic with the test statistic (from the p-value) estimated
by the degrees of freedom and level of significance. If the computed value lies within the
region of rejection defined by the alternative hypothesis (the critical value) then the null
hypothesis is rejected. The distributions for each fix type and year after the preventive

maintenance activity are given in Appendix B along with the population sample size.

5.4.2 Pavement Condition Using Ride Quality Index (RQI)

A two sample t-test was conducted in order to determine the relevance of the RQI
guideline values that the MDOT imposes for the different preventive maintenance fixes.
Under consideration will be the mean distress indices at subsequent years for two
populations:

0 Population One: RQI values less than or equal to the RQI guideline values at

the time of the preventive maintenance activity.

0 Population Two: RQI values greater than the guideline values at the time of

the preventive maintenance activity.

53

For each treatment, mean distress indices at subsequent years were analyzed using this
method. Below are the formulated null and alternative hypotheses:
0 Ho: Mean DI at time (ti) of population one equals mean DI at time (ti) of
population 2 or
0 HA: Mean DI at time (ti) of population one does not equal mean DI at time (t;)
of population 2.
The distributions for each fix type and year after the preventive maintenance

activity are given in Appendix C along with the population sample size.

54

CHAPTER 6 - RESULTS

6.1 Introduction

A reliability analysis was performed using distress index (DI) data in order to
determine the life expectancy for the various preventive maintenance treatments. For
each fix type, distributions of DI of 0.1-mile sections were generated for various years
following the preventive maintenance. For a given year, the reliability was calculated as
the number of sections with distress indices that are less than the guideline value divided
by the total number of sections for that year. The same procedure was done for

subsequent years (i.e. one year after the PM action, two years after the PM action, etc.)

6.2 Determination of Life Expectancy of the Various PM Fixes

The life expectancy of a given fix is the time (in years) it takes for a pavement
section that was subjected to the fix to reach a limiting DI value. The life expectancy for
each fix type was determined by plotting the distributions of D1 values at subsequent
years afier the fix and calculating the area under the curve to the lefi of the DI guideline
value. The reliability values were then plotted over time; a quadratic function was fitted
to the data points; and the 95% confidence intervals over the mean values were then
plotted. From these graphs, tables can be developed for each treatment showing the
reliability, plus or minus some error 11 years after the initial treatment.

It should be noted that the life expectancy of a fix as defined above is different
from the life extension provided by the fix as defined by the MDOT. The definition of

MDOT’S life extension is provided in Chapter 3.

55

In order to check the validity of reliability values, the average distress index for
each fix type was plotted over time. Because the analyzed pavement sections had only
one preventive maintenance treatment applied to them, the distress index is expected to
grow over time. Figure 6.lshows that the distress index for crack seal, micro-surface,
mill & fill, and bituminous overlay increase with time except for year six and year seven.
The distress index for chip seal for the most part increases with time, but has some
variability which could be due to a number of factors. Given that these values represent
the average over multiple pavement sections, and that the general trend is upward, it can

be inferred that the data are reasonable for further analysis.

    
  

 

  
 
 
  
 
 

100’ . ICreck Seal
90' IMicro-Surface
80 0 Chip Seal
23 n Mill 3. Fill
Average DI After 50 lIBituminous Overlay
PM 40
30
20
_ B. .
10 Militgrglilious Overlay
0 a Chip Seal
Micro-Surface
3 4 5 6 Crack Seal FIX Type
7
Time (Years) 8

Figure 6.1 — Average DI after Preventive Maintenance versus Time

56

6.2.1 Non-Structural Bituminous Overlay

Figure 6.2 and Figure 6.3 show the reliability of the non-structural bituminous
overlay versus time based on the PM guidelines and rehabilitation threshold, respectively.
Figure 6.2 shows that the reliability is relatively constant up to year three and then starts
to drop at a faster rate yet the reliability is still rather high at year six (78%). This agrees
reasonably well with the MDOT’s preventive maintenance guidelines, which state that
the life extension for a non-structural bituminous overlay is five to ten years. Therefore
one would not expect to see a very distressed pavement from year one to year five.
Figure 6.3 shows that reliability based on the rehabilitation threshold over time is quite
high up to year six (90%). Both of these figures show a small amount of variability
around the mean value. Figure 6.2 will allow a highway agency to plan a preventive
maintenance program for a pavement section when a pavement section gets close to the
end of the preventive maintenance cycle than the reliability based on the rehabilitation

threshold (Figure 6.3) can be used.

57

 

Reliability

.3

.2:

.1 I
0.0

 

 

 

 

qu = 0.6811

 

_LI

NI

#1
CI
0)

5

Time (Years)

Figure 6.2 - Non-Structural Bituminous Overlay Reliability versus Time Based on the
PM Guidelines (95% Confidence Interval)

 

Reliability

.40 1

.20 r

.10 I
0.00

 

 

 

qu = 0.5833

 

O
dbl

3

#1
on
0)

Time (Years)

Figure 6.3 - Non-Structural Bituminous Overlay Reliability versus Time Based on the
Rehabilitation Threshold (95% Confidence Interval)

58

6.2.2. Surface Milling with a Non-Structural Bituminous Overlay

Figure 6.4 and Figure 6.5 Show the reliability of surface milling with a non-
structural bituminous overlay treatment versus time based on the PM guidelines and
rehabilitation threshold, respectively. The reliability is close to 100% for the first four
years for both figures and the variability around the mean values is quite small. One
would expect this since the MDOT life extension is from five to ten years. Therefore one
would not expect to see a pavement that is very distressed from year one to year four
since this treatment mills off the existing surface and a new bituminous overlay is placed

to correct the existing distresses.

1.10

 

 

LOOP "

CI

 

I:

.90 .
.80 I
.70 'l

.601

Reliability

.30-
.204

.10 I

0.00 _ _ qu = 0.3333
0 1 2

 

 

 

 

one
A

Time (Years)

Figure 6.4 — Surface Milling with a Non-Structural Bituminous Overlay Reliability
versus Time Based on the PM Guidelines (95% Confidence Interval)

59

 

1.200
1.100'I
1.0mm
.900ll
.800I
.7001
.600-l

Reliability

.400 I
.300 u
.200 r

.100 I
0.000 qu = 0.3333

fi T

0 i 2 3 4

 

 

 

Time (Years)

Figure 6.5 — Surface Milling with a Non-Structural Bituminous Overlay Reliability
versus Time Based on the Rehabilitation Threshold (95% Confidence Interval)

6.2.3. Single Chip Seal

Figure 6.6 and Figure 6.7 show the reliability of the single chip seal treatment
versus time based on the PM guidelines and rehabilitation threshold respectively. Figure
6.6 shows that there is quite a bit of variability around the mean reliability values. The
overall trend for the mean values is to decrease with time with the rate of deterioration
increasing afier year three. Figure 6.7 shows the reliability over time based on the
rehabilitation threshold. The variability around the mean value decreases with the overall
trend showing that the reliability decreases with time. For a given year, the reliability
value based on the rehabilitation threshold is higher than that from the preventive
maintenance guideline, since the PM value is much more restrictive than the

rehabilitation threshold. The results indicate that single chip seals have a reliability of

60

about 80% after three years and about 60% after six years (21:20%). The reliability values
increase to about 90% after three years and about 70% afier six years (:l:10%) when the
rehabilitation threshold is used. This agrees reasonably well with the MDOT’S

preventive maintenance guidelines, which state that the expected life extension is three to

six years.

1.2

 

1.1ll
1.00
.9' D
.81
.7-

.61

Reliability

.5I
.4!
.3-
.21

.11

0.0 - _ qu = 0.5307
0 1 2 3

 

 

 

A:
OH
0:

Time (Years)

Figure 6.6 — Single Chip Seal Reliability versus Time Based on the PM Guidelines (95%
Confidence Interval)

 

1.2
1.1'l
1.0‘I

 

Reliability

.5-
.41
.31
.21

.11

0.0 f i - qu = 0.7470
0 1 2 3

 

 

 

AI
01:
CD

Time (Years)

Figure 6.7 — Single Chip Seal Reliability versus Time Based on the Rehabilitation
Threshold (95% Confidence Interval)

6.2.4. Multiple Course Micro-Surfacing

Figure 6.8 and Figure 6.9 Show the reliability of the multiple course micro-
surfacing treatment versus time based on the PM guidelines and rehabilitation threshold
respectively. Both figures show the reliability over time decreasing at a slightly
increasing rate with time. Also, the variability around the mean value is quite small. The
results indicate that multiple course micro-surfacing have a reliability of about 80% after
four years and about 65% after six years (i5%). The reliability values increase to about
85% after four years and about 75% after six years (15%) when the rehabilitation
threshold is used. This agrees reasonably well with the MDOT’s preventive maintenance

guidelines, which state that the expected life extension is four to six years.

62

 

 

Reliability

 

 

.0 ' qu = 0.9375
3

 

C
All
N1
#1
OH
0)

Time (Years)

Figure 6.8 — Multiple Course Micro-Surfacing Reliability versus Time Based on the PM
Guidelines (95% Confidence Interval)

1.2

 

1.1'
1.0‘
.9' o
.81
.71
.6I

Reliability

.5I
.4-
.31
.21

.1 1
0.0 qu = 0.8582

- i i

0 1 2 3

 

 

 

 

.51
011
O)

Time (Years)

Figure 6.9 — Multiple Course Micro-Surfacing Reliability versus Time Based on the
Rehabilitation Threshold (95% Confidence Interval)

63

6.2.5. Bituminous Crack Seal

Figure 6.10 and Figure 6.10 Show the reliability of the bituminous crack treatment
versus time based on the PM guidelines and rehabilitation threshold, respectively. Figure
6.10 shows that the reliability based on the preventive maintenance guideline decreases
steadily with time, dropping to 80% after one year, 65% after two years, and 55% after
three years (:ES%). On the other hand, the reliability based on the rehabilitation threshold
remains above 90% after four years, then starts dropping Sharply after five years. The
large discrepancy in the reliability values corresponding to the preventive maintenance
and rehabilitation thresholds is mainly due to the fact that the distress points assigned to
sealed cracks are about half of those for unsealed cracks. The limiting value based on
preventive maintenance appears to be too restrictive. This agrees reasonably well with
the MDOT’s preventive maintenance guidelines, which state that the expected life

extension of bituminous crack seals is up to three years.

64

1.2

 

Reliability

 

 

 

0.0 _ ' ' qu = 0.9615
0 1 2 3 4

 

011
OH
\I

Time (Years)

Figure 6.10 — Bituminous Crack Sealing Reliability versus Time Based on the PM
Guidelines (95% Confidence Interval)

1.2

 

1.1‘
1.0'I

33x

.61

Reliability

.5-l
.4-
.31
.21

.11

0.0 _ _ qu = 0.9412
0 1 3 4

 

 

 

NI
OH
(”I
N

Time (Years)

Figure 6.11 - Bituminous Crack Sealing Reliability versus Time Based on the
Rehabilitation Threshold (95% Confidence Interval)

65

6.2.6. Overall Comparison

Table 6.1 is a first estimate of the reliability values corresponding to the different
life expectancies for each preventive maintenance fix. Table 6.2 provides a summary of
the reliability values of the life expectancies. The DI guidelines for preventive
maintenance were used for detennining these reliability values since the MDOT uses
them for initiating preventive maintenance actions. The results indicate that the
guidelines are reasonable. Table 6.3 provides a summary of reliability values of the life
expectancies based on the rehabilitation threshold. Both of these tables will allow a
highway agency to select the appropriate fix depending on the life expectancy that they
are looking for along with the variability associated with certain fixes. The reliability
values based on the rehabilitation threshold can be used when certain preventive
maintenance projects are towards the end of their life-cycle and a rehabilitation project is
planned for the future. A highway agency will then know the life expectancy of the

subject treatment.

66

monm 0388 380. 03065 230 @0325 “032

w 30 > 3 308$ 2 m

n 30 > 3 $00.85 m w

n 30 > 3 $00.85 w n

N. 30 > 3 308$ 2 23 o 30 > 3 30085 N .m 30> 3 $00.85 v N
w 30 > 3 30.8.5 ~ 93 N. 30> 3 $00.85 N _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$3 $3 $3 $3 $3 $3 $3 $3 3
....$~._ $3. $3 $3 $3. $3 $5 $3 2 mam 286
$2 $8 $3 $3 $3 $3 $3 $3 3 833-922
32 .,$3., $8 $3 $3 $3 $3 $3 3 . .
$2 $3 $3 $3 $3 $3 $3 $3 3

<2 $3 $2. $3 $3 $3 $3 $3 3 28m 26
$2 $2: $2: $3 $3 $2: $3 $2: 3

$2 $2: $2: $2 $3 $2: $3 $2: 3 2232:
$3 $3 $3 $3 $3 $2: $3 $2: 3 2235305533
$3 $3 $3 $3 $3 $2: $3 $2: 2. _ . .
w b 0 m e m N _ amaawwywmwwfin 2:2 E

3.33 E 2.: .2: 08:

 

 

 

 

was”: 2.: Sou; 2: 5 0:5 as 33> £320: me 353 mi - S 035

67

 

0N3 03:30 :95 mo 004303 03303253 “oz “<\ZN
$02: 000020 8:30 5:53:00 02:. ”Bo?

 

 

 

 

 

 

fiemfimm Xmfiam $.mflmm $.33 Xmflmm AXmHS fomflow m 8 AD 30m 0:080
N32 .333 $9”? $0?”on gmfimw $925 3?? 04 003.2:W822
N<\Z $838 $833 .xamnnt. xxbmfimw uxeomflmw 23833 9m 30m QED
N<\Z N<\Z N<\Z R3333 :gmfiao atom“? 3:333 on Em 0% ES

23 00>
N32 .3332. $611.» 3330“? $.20"; .fibiog x3032: on machmgmm
$380
2 0 m a N N : 5:88:— 25. NE
a: HOG—Z

 

 

 

 

 

 

 

2:86 a: s: 082 2.5

 

 

 

 

3:35 88380 $3
82025 2.: “.892 8 880 35 35> 50 35333 .23 00 8232...: 3350mm - N0 222

68

 

0N6 03:80 2080 .«0 003003 0300230 8 Z ”<\Z~
$2: 00088 8580 3:30:00 0.: ”0007:

 

 

 

 

 

 

 

 

 

 

 

 

 

$fi$ $3: $32 $38 $003 $2": $302 meg: new 08.6
N$2 $9"? $3; $9§ $38 $$3 $33 3 8050-8sz
N$2 $235 $206” $2an $208 $203 $Sur0m 0-». new 020
N32 "<2 N.32 $32 $32: $32 $18. own 5002::
32 $38 $33 $33 $38 $m+.oo_ $380 oi. Smufim
9.836
b 0 m a m N 0 5:535. 25‘ NE
a: son:
A2306 5,.— 20 .35.. 2:;

 

 

 

 

@235 85380 $30
2285 85:30:00 5002 8 830 85 msofi> :0 %“$825 a: 00 305023 0250mm u 3 2E

69

6.3 Evaluation of the Guideline Values for the Treatment Types

Most preventive maintenance engineers agree that the timing of the preventive
maintenance treatment is very critical. If a treatment is applied to very distressed
pavements then the treatment will not give the pavement the added benefit it needs. If a
treatment is applied too soon, then money is being poorly spent on a pavement that really
does not need the maintenance. Therefore there is a real need to evaluate the guideline
values at which a given preventive maintenance fix should be applied so that the life
extension of the pavement is maximized while minimizing the cost to the highway
agency.

This section will look at the effect of applying a treatment to a pavement that has
a distress index that is lower than the preventive maintenance guideline value compared
with applying it when the distress index is greater than the preventive maintenance
guideline value. Two approaches will be looked at: (1) two sample t-test, and (2)

reliability analysis.

6.3.1 Two-Sample t-test Approach
A two sample t-test was conducted in order to evaluate the DI guidelines that the
MDOT imposes for the different preventive maintenance fixes. Under consideration will
be the mean distress indices for two populations:
0 Population One: DI values less than or equal to the DI guideline values at the
time of the preventive maintenance activity.

0 Population Two: DI values greater than the guideline values at the time of the

preventive maintenance activity.

70

The mean distress indices at subsequent years for each preventive maintenance treatment
were analyzed using the hypotheses formulated below:
0 Ho: Mean DI at time (ti) of population one equals mean DI at time (ti) of
population 2, or
0 HA: Mean DI at time (ti) of population one does not equal mean DI at time (ti)
of population 2.
The next step is to select the appropriate level of significance (l-a). An alpha
value (on) of five percent (5%) was chosen in this case, which corresponds to a 95% level

of significance. The results are summarized in Table 6.4.

6.3.1.1 Non-Structural Bituminous Overlay

The PM guideline for a non-structural bituminous overlay calls for a D1 value less
than forty (<40). The first three years of extended life for a non-structural bituminous
overlay are relatively insignificant since the expected life extension according to the
MDOT is from five to ten years. The results of the two sample t-test for this treatment
can be seen in Table 6.4. The p-values from year five to year six are below the 5%
significance level and the mean DI for population two is significantly higher than that of
population one for these years. This analysis reasonably confirms that the MDOT
guideline value for a non-structural bituminous overlay (DI<40) is valid. Data in year
eight and beyond has a very limited population and thus analysis was not done for these

years.

71

6.3.1.2 Surface Milling with a Non-Structural Bituminous Overlay

The PM guideline for surface milling with a non-structural bitmninous overlay
calls for a DI value less than forty (<40). The results of the two sample t-test for this
treatment can be seen in Table 6.4. These results show that no statement can be made
about the guideline value at this time since the difference in the mean DI values for both
populations is not statistically different. The results are therefore inconclusive, and more

performance data for later years are needed.

6.3.1.3 Single Chip Seal

The PM guideline for single chip seal calls for a D1 value less than twenty-five
(<25). The results of the two sample t-test for this treatment can be seen in Table 6.4.
The p-values from year one to year six are below the 5% significance level, except for
year seven (p-value=0.77). This anomaly can be ignored since the anticipated life
extension for a single chip seal according to the MDOT is three to five years. Therefore
one would not expect this treatment to last seven years. The mean values of DI after the
fix are statistically different, with those from population one being lower than those from
population two. This analysis confirms that the MDOT guideline value for a single chip
seal (DI<25) is reasonable. Year seven was ignored due to the fact that only two projects

had DI data seven years after the treatment.

6.3.1.4 Multiple Course Micro-Surfacing

The PM guideline for multiple course micro-surfacing calls for a D1 value less
than thirty (<30). The results of the two sample t-test for this treatment can be seen in

Table 6.4. The p-values from year one to year six are below the 5% significance level,

72

suggesting that the means of population one and two are statistically different, with those
for population one being lower than those for population two, at all years. This analysis

confirms that the MDOT threshold for a multiple course micro-surfacing (DI<30) is

valid.

6.3.1.5 Bituminous Crack Seal

The PM guideline for bituminous crack sealing calls for a D1 value less than
fifteen (<15). The results of the two sample t-test for this treatment can be seen in Table
6.4. The p-values from year one to year six are below the 5% significance level, with the
mean DI values for population two being clearly higher than those from population one.
The p-value for year seven is higher (0.69). This can be ignored since the anticipated life
extension for a bituminous crack seal according to the MDOT is up to three years only.
Therefore one would not expect this treatment to last six to seven years. This analysis
confirms that the MDOT guideline value for a bituminous crack seal (DI<15) is

reasonable.

73

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E mam: 338m as-“ 038mm 25 1 to 2.5

55 ~83. . o o o o o 85¢
2 .v- N .5. m5- .2- a. s- v .x gem;
:3. a.» new 3N VS Md. ”3 9% 3m 5.: as 2d ”am as >8 .3
8+ 86 5a SN gm ”.2 m: 8.5 Nam w: 5.2 NE EN _.2 5...: sum 35
mm % a. 8m 02 ”mm 5 a mom .5 as mm. 5 9% z
.5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5
H32 28mm co¢< 20.3m— 5¢< 280m 3¢< 280m 8% 280m $¢< 28cm St< Beam
Born Eo>c Eo>m Eo>v §o>m Eo>m 58>.
m8 one who 86 Ed 9o 83¢
a: as ”No- :5 2a N2 35
2+ a: Sm 3m 3: m; w am one Rs a; 3m >5 ém
N3 5 gm 2:. we $2 :2 Re 3 2 a2 ME :82 E a a:
a N _ _ m S mom 8 on 5 OS an t. z
.5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5
noc< 280m n32 280m $¢< oaomom ho¢< ohomom coc< 880m cot/x 88cm 53?. vacuum
Eo>n Eo>m $0er 30>m So>m 30>—
md «.85 . o Sod @5 ~85 c 85-;
:g- am- m3. m2- 2.“- gm ”8- 5.25
am Ea :_ SN 3: was we SN 3m NE a: 2: 8+ 3:. >8 .3 E55
SN v3 5 a: 0.8 8.” EN 2 mg a: ma 2: an and :82 255525
La 3 _ _ : _ 2 5 5H 5 we 5 5 5 a _ 8m
.5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5
5c< 280m new? 280m noc< 280m coc< Baum Hot/x 280m Bases 280m coa< 280m
Eo>w Eo>o Eo>m 58>v Eo>m “moxrm “mo>_ 2*»th
E 2; ER 25

74

.xh 55 mmo. mos—«>5 88:2: 260 woumnm.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$5 58.0 We a, c c , . c 35¢
:5- 3. mm? .28. W5. 9.2- 5.“. .55
SN m: N8 9% EN _.: we. 2: we we SN a: ”.3 Rd so .3
m8 2% 3m 3m 5m 3m 2m v.2 5 2: 5m «:2 EN Va: 502 sum #86
mm 5 a 5 ow v8 5 :2 5 oz 5 28 a: 98 z
.5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5
uo¢< Bowen uo¢< 280m— uo¢< vacuum uo¢< chomom uo¢< Begum uo¢< vuomom u0¢< 20.3mm
Horn ufi0>© ao>m 30>? umo>m umo>N 30>—
Scod 585 c o o . , o 35¢
2.- mi... 5.8. $2- ta. 2- .55
N: Sm Sm EN 5 _.2 3m 2 E. a.” 9m. .3 >5 .3 Beam
2: Em a 3: mm m: «.5 2 EN 2: Em 5.2 :82 -222
a? E. M: 3m :2 as a. OS 5. so 8 5 z .
.5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5
uo¢< 280m uo¢< Ouomom u0d< vuomom— uo¢< 280m uo¢< 280m uo¢< 0.5me uu¢< 280m
30> c 30> m 30> v .50? m 30> N 53> _ 09:. XE
x: 5 he? 25

 

 

@3580an HQ wfimb 338% 58; 038mm 03,—. I v.0 033.

 

 

75

6.3.2. Reliability Approach

A reliability approach was also done using the same concept of having two
populations and looking at how the distress index changes over time. Distributions were
plotted for the two populations and at a specified point in time (i.e. 1 year, 2 years, etc.
after the preventive maintenance activity). This will show the performance of the
pavement when the treatment is applied correctly and when it is applied incorrectly
according to the distress index criteria. The results are summarized in Table 6.5, and are
discussed below. The life expectancy of a given fix type for the two populations was
determined by plotting reliability versus time, fitting a quadratic function to the data
points, and then plotting the 95% confidence intervals over the mean values. The figures
can be used to determine the reliability in performance when a treatment is applied before

or after the guideline values.

6.3.2.1 Non-Structural Bituminous Overlay

Reliability values for this treatment type along with the average distress index
before the preventive maintenance treatment was applied are shown in Table 6.5 for both
populations. The shaded cells indicate very small sample sizes and should therefore be
ignored. Figure 6.12 and Figure 6.13 show a graph of reliability versus time for the two
populations.

The results show that while the reliability values for population two are slightly
lower than those of population one, and this difference increases with time, the difference
is too small to make any statement on the guideline value. No conclusions can be drawn

about the guideline value since the reliability level is still high for both populations to

76

make a firm statement about the guideline value (DI<40). Future performance data are

needed in order evaluate the guideline value afier five years or more.

1.2

 

1.1¢
1.0a

Reliability

 

 

 

1 i 5

Time (Years)

#-

(”I

qu = 0.7185

Figure 6.12 — Reliability of Life Expectancy for Non-Structural Bituminous Overlay

When PM is done before the DI Guideline Value

77

 

Reliability

 

 

0.0 _ j ESQ = 0.5031
2 3

 

O
A:
#1
0!

Time (Years)

Figure 6.13 — Reliability of Life Expectancy for Non-Structural Bituminous Overlay
When PM is done after the DI Guideline Value

6.3.2.2 Surface Milling with a Non-Structural Bituminous Overlay

The reliability values for this treatment type along with the average distress index
before the preventive maintenance treatment was applied are shown in Table 6.5 for both
populations. Figure 6.14 and Figure 6. i 5 show a graph of reliability versus time for the
two populations.

The sample sizes for year three and later are too small to make any inferences.
For the first two years, the reliability values for both populations are practically identical;
this is expected since it would be too early for this fix type to show any signs of serious
distress. Therefore, no conclusions can be drawn about the validity of the guideline value

(DI<40) for surface milling with non-structural bituminous overlay.

78

 

1.10‘
.0017

.90‘

.80'

.70 I

.601

.50ll

Reliability

.40!I

.30-

.20 I

.10 I
0.00

 

 

 

l3

 

qu = 0.4857

 

 

2 3 4

Time (Years)

Figure 6.14 — Reliability of Life Expectancy for Surface Milling with a Non-Structural
Bituminous Overlay When PM is done before the DI Guideline Value

1.10

 

1.00'

.90‘

.801

.70 I

.60!

.50-

Reliability

.40lI

.30-

.20 I

.10 I
0.00

 

 

 

0.0

”0,1

I I'

1.0 1.5 2.0

Time (Years)

Figure 6.15 - Reliability of Life Expectancy for Surface Milling with a Non-Structural
Bituminous Overlay When PM is done after the DI Guideline Value

79

6.3.2.3 Single Chip Seal

The reliability values for this treatment type along with the average distress index
before the preventive maintenance treatment was applied are presented in Table 6.5 for
both populations. Figure 6.16 and Figure 6.17 show a graph of reliability versus time for
the two populations.

The results clearly show that the reliability values for population one are higher
than those for population two, with the difference increasing with time. Therefore this
method reasonably confirms that the MDOT guideline value (DI<25) for single chip seal

is valid.

1.2

 

1.1I

1.01) 0

.9‘
o
.81
/0—\
.7i

Reliability
9

 

 

.0 - qu = 0.3726
3

 

O
—u
N
#1
O'l
0)

Time (Years)

Figure 6.16 — Reliability of Life Expectancy for Single Chip Seal When PM is done
before the DI Guideline Value

80

 

Reliability

 

 

0.0 . qu = 0.5831
0 1 2 3

 

 

#1
ON
a:

Time (Years)

Figure 6.17 — Reliability of Life Expectancy for Single Chip Seal When PM is done after
the DI Guideline Value

6.3.2.4 Multiple Course Micro-Surfacing

The reliability values for this treatment type along with the average distress index
before the preventive maintenance treatment was applied are presented in Table 6.5 for
both populations. Figure 6.18 and Figure 6.19 show a graph of reliability versus time for
the two populations.

Despite the small sample sizes in population two for most of the years, the results
clearly show that the reliability values for population two are significantly lower than
those of population one. Therefore this method confirms that the MDOT guideline value

(DI<30) for a multiple course micro-surfacing is reasonable.

81

 

 

Reliability

 

 

 

.5 d
.4 -
.3 ll
.2 I
.1 l
.0 _ _ r _ qu = 0.9028
0 1 2 3 4 5 6
Time (Years)

Figure 6.18 — Reliability of Life Expectancy for Multiple Course Micro-Surface When
PM is done before the DI Guideline Value

 

1.00
.90 "
.80 ‘
.70 l
.60 I

.50-

Reliability

.40 I
.30 'l
.20 1

.10 I m

 

 

0.00 _ - _
0.0 1.0 2.0 3.0 4.0

 

Time (Years)

Figure 6.19 — Reliability of Life Expectancy for Multiple Course Micro-Surface When
PM is done afier the DI Guideline Value

82

6.3.2.5 Bituminous Crack Seal

The reliability values for this treatment type along with the average distress index
before the preventive maintenance treatment was applied are shown in Table 6.5 for both
populations. Figure 6.20 and Figure 6.21 show a graph of reliability versus time for the
two populations.

The results show clearly that for year one to year five there is a significant
difference in the reliability values when performing the treatment before the guideline
value then when doing it after the guideline value. Therefore this method reasonably

confirms that the MDOT guideline value (DI<15) for a bituminous crack seal is valid.

 

Reliability

 

 

 

qu = 0.8902

 

0

AI

MI
011
O)!
\l

5 5

Time (Years)

Figure 6.20 — Reliability of Life Expectancy for Bituminous Crack Seal When PM is
done before the DI Guideline Value

83

1.1

 

1.0'

Reliability

 

 

U
0.0 qu = 0.9525

Time (Years)

0
in
”1
co
.5
01
m
w

Figure 6.21 - Reliability of Life Expectancy for Bituminous Crack Seal When PM is
done afier the DI Guideline Value

84

BR 038mm :95. 88:65 303 E cocmnm “802

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

<\Z $m $om $2 $3 $2 $8 $3 NNN m _A How x85
32 $5 $mm $3 $3 $3 $3 $2. vw m To
$2 ,..,.$0 .. $3 . .$2 $2 $5 $3 , $3 Dov. omA ll 083m
.32 $o3.....m.. $5 $3 $3 $3 $3 $3 9w omé -832
$2 $o3...n $2 $8 $3 $mm $R $wm 52. RA Rom QED
$2 $3 ., $8 $wn $2: $3 $3 $3. ad 36 .
$2 $63 $2 $3 $03 . $2: $3 $2: ode OVA . :5 a. E2
$2 $8 a ,, _ $02 $8 $3 $2: $wo $2: 92 ové . .
. ,. $mN $2 $a $oo $Nw $2: $ma $3 vfi: OVA .8125
$3, $3. $5 $3 $3 $2: $wa $2: 0.3 Ste 323885
w \l o w v m m _ Em 880m Em Roman 09C. xi
5 owfio>< 654 a
meet 5 SE baa. 0E

 

 

89¢. “cogmouh 65 8% 333/ 053230 68 mucus—gm so £9222 banged how 033. Egm l m6 2%...

85

6.4 Evaluation of Ride Quality Guidelines

The ride quality of a pavement is an important aspect to the user of the road. If
pavements are too rough this may accelerate the rate of distress but also accelerate
damage to cars and trucks through spalled off pieces damaging windows, chipping of
paint, increased wear of tires, and damage to the suspension of cars and trucks.

The evaluation of the RQI guideline values was accomplished by checking
whether the RQI value at the time of application of the preventive maintenance fixes
significantly affects the distress index of the pavement after the fix. A two sample t-test
was conducted on the mean distress indices for two populations:

0 Population One: RQI values less than or equal to the RQI guideline value at

the time of the preventive maintenance activity.

0 Population Two: RQI values greater than the guideline value at the time of

the preventive maintenance activity.
The mean distress indices at subsequent years for each preventive maintenance treatment
were analyzed using the hypotheses formulated below:

0 H0: Mean DI at time (ti) of population one equals mean DI at time (ti) of

population 2, or

0 HA: Mean DI at time (t;) of population one does not equal mean DI at time (ti)

of population 2.

The next step was to select the appropriate level of significance (l-a). An alpha

value (a) of five percent (5%) was chosen in this case, which corresponds to a 95% level

of significance.

86

6.4.1 Non-Structural Bituminous Overlay

The results of the two sample t-test shown in Table 6.6 indicate that the p-values
are all greater than the 5% level of significance except for year four. This implies that the
difference in the mean DI after the fix of both populations is not statistically significant.
Therefore the results seem to indicate that the RQI value at the time of the fix does not
seem to affect the future DI values (at least up to five or six years). This is expected
since overlaying a pavement will smoothen its surface so that roughness is minimized
afier overlaying. The anomaly seen at year four where the p-value is less than the 5%
level of significance could be explained by the fact that the ROI threshold (>70) being
very high may reflect serious problems in the pavement foundation layers. In that case,
overlaying with a thin, non-structural layer may not solve the root cause of the pavement

distress.

6.4.2 Surface Milling with a Non-Structural Bituminous Overlay

The results of the two sample t-test shown in Table 6.6 indicate that the p-values
are all greater than the 5% level of significance except for year five. This implies that the
difference in the mean DI after the fix of both populations is not statistically significant.
These results indicate that the RQI value at the time of the fix does not seem to affect the
future DI values. Again, this is expected since this fix involves milling the pavement
surface, and then overlaying the pavement, thus minimizing surface roughness. The
anomaly seen at year five where the p-value is less than the 5% level of significance can
be attributed to the fact that a high RQI value (>70) may be indicative of more serious

foundation problems that can not be fixed by a mill and fill treatment.

87

6.4.3 Single Chip Seal

The results of the two sample t-test shown in Table 6.6 indicate that the p-values
are below the 5% level of significance. However, the mean DI values are higher for
population two only after four years or more. Year six show a p-value above the 5%
level of significance. However the data for year six may be ignored because of small
sample size of pavements with RQI values higher than the guideline value. Therefore the
results seem to indicate some effect of the RQI guideline value (>54) at later years, but

they are not as conclusive as those for the DI guideline value.

6.4.4 Multiple Course Micro-Surfacing

The results of the two sample t-test shown in Table 6.6 indicate that the p-values
are variable from year to year. Years one, three, and five have a p-value higher than the
5% level of significance while years two, four, and six have p-values lower than the 5%
level of significance with the mean DI value for population two being higher than those
for population one. Therefore the results are inconclusive, and more data are needed to
confirm or disprove the hypothesis that the RQI level at the time of applying a multiple

course micro-surfacing will affect the future performance of the pavement after the fix.

6.4.5 Bituminous Crack Seal

The results of the two sample t-test shown in Table 6.6 indicate that while the p-
values are lower than 5%, the mean DI values for population one are higher than those of
population two for years four and seven. Therefore the results are inconclusive, and more
data are needed to support or disprove the hypothesis that the RQI level at the time the

crack sealing is applied will affect the future performance of the pavement.

88

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a; 9 ,. , 8.0... m2 . «.86 02.30
2 0%- e5 2. 3- .093
20 20 q: 2: m. a tam .2 an; EN EN an em
0:. as a: we 2m a: w: 5: SN «.2 as: :00 36
8 mm E m: we 5 as $0 ”a so 2
.E .E .E E .5 .E .E .E .E .E .E .E
n0c< 0080m 00c< 0080m 00¢< 00£0m 000.2 0080mm 00c< 0080m n0c< 280m 00c< 0080m
30> \l 80> w 80> m 30> v 30> m 50> m 50> _
02 0805 m2 :3 Ge 0.8 2.30
$3- 2- a: So. So. :5. new;
03. 9% SN 30 mg as a? MS 3.... 3.” m2 3. an .3
cm as: 25. $3 :2 3.2 n: 8.2 was EN 02 as :82 :5 a a:
e 0 NM 2. a me a E E 2 _ em 3 z
E .5 .E. .E .E .E .E .E .E .E. .E .E .E E
00¢< 0080m 00c< 0080m 00a< 0080m 00¢< onomom 00d< 0.80m .32 0080m 00c< 0.80m
n00>n 80>c 000>m 50>¢ 30>m 30>N 30>~
a; :0 .. o 33 :.o 5.2 83?;
”2 3... 2.0- do NE- woe- same
25 z: 2: :0 :2 :R a2 0; 3H 3.0 NS 8.0 >8 am 3.20
3.: 8.: 2 :0 SN 08 e; a; a; we 80 N3. :82 32.535
e :N : a: :. SN : mm 8 at. 2 N8 2 . .
.E .E .E .E .E .5 .E .E .E .E .E .E .E .E
uo¢< chomom uo¢< 280m ho¢< 880m uo¢< 0.5QO uo¢< 280m hod< Begum hot/x whomum
30> N. 80> c 30> m 50> v 30> m 30> N 30> ~ 09C. XE
E 20 .22 25

 

 

am so Emam 338% 008.0 295m 93. l 0.9 058.

89

{on :06 000— 0035-00 000002: 0:00 003%.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2002553 am no 008m 330.0% 0000-0 038% 03... I 0.0 030.0

 

_..... 0000.0 . 0000.0 .. 000.0 _000 .. 0... 0000.0. ,_ 0... 035-0
00.0 00.0- . 0. _- 000 _00. 00.0- 00.0- .085
E 0.: 0.0_ 2 0.2 0.0_ 0.0_ 0.: 00.0 S 00.0 00.0 2: 0.0. >5 .000
0.: 0.00 0.00 0.0_ 00 0.00 0.2 0.0_ 0.00 0.2 _.2 00.0 2 : :82 .80 0220
00 00. 0_ 000 02 _00 00 000 02 000 00 _00 00_ 000 z
.5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5
uo¢< whOmom uo¢< Duomum uoflxx ohomom uo¢< Duouom uu¢< whomom uo¢< ouowum uo¢< Duomom
umo> h umo> 0 53> w 23> V amu> M .30? N 23> _
0 00.0 0 00.0 0 00.0 2.5-0
2.0- 00.0- $0. _00- 0E _- 00.0 .0805
0.00 0.00 E 0.00 0.3. _00 _0.0 00.0 0.00 00.0 : _0.0 So .000 80050
0.00 0.: 0.0_ 0.: 0.00 2.2 00.0 00.0 0.00 00.2 0.0. 00.: :82 -222
0: 000 E 000 02 _00 00 000 00_ 0: 00 0; z .
.5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5
uo¢< Ouomom uo¢< ohomom uo¢< DuOmum uo¢< chemom uo¢< Ouowom uo¢< uuowom ho¢< ohowom
hmo> h umo> 0 23> m 53> .v umo> M .30.? N 53> — ogflh X—nm
0: 20 he? 0:5

 

90

CHAPTER 7 - SUMMARY OF FINDINGS AND RECOMMENDATIONS

7.1 Summary of Findings

The research objective was to use the wealth of PMS performance data collected
on 240 preventive maintenance (PM) projects since 1992 to perform a reliability-based
analysis for the purpose of determining the life expectancy and evaluation of the
guideline values for the different PM fixes. These are: (l) Non-structural bituminous
overlay, (2) surface milling with a non-structural bituminous overlay, (3) single chip seal,
(4) multiple course micro-surfacing, and (5) bituminous crack seal. The pavement
performance parameters that MDOT collects are distress index (DI), ride quality index
(RQI), and rut depth. There were insufficient rut data to perform the analysis, therefore
only DI and RQI data were used.

0 Determination of the Life Expectancy for PM Fixes Using D1

The life expectancy was determined by plotting distributions of the distress index
at subsequent years after the preventive maintenance activity. The reliability was then
calculated as the area under the curve to the left of the MDOT guideline value. The
results were summarized in tables showing reliability values that correspond with the
different life expectancies for each preventive maintenance fix. The life expectancy
values estimated from the reliability-based analysis compare reasonably well with those
suggested in MDOT’s preventive maintenance guidelines. The reliability approach
suggested in this research can be used as a pavement management tool, provided that the

MDOT adopts a minimum level of reliability that would be acceptable before an action is

taken.

91

0 Evaluation of Distress Index (DI) Guidelines

A two sample t-test was conducted in order to evaluate the DI guidelines that the
MDOT imposes for the different preventive maintenance fixes. The mean distress index
was determined for two populations:

0 Population One: DI values lower than or equal to the DI guideline values at

the time of the preventive maintenance activity.

0 Population Two: DI values greater than the guideline values at the time of the

preventive maintenance activity.
For each treatment the mean distress indices at subsequent years after applying the fix
from both populations were compared.

A reliability approach was also done using the same concept of having two
populations and looking at how the distress index changes over time. Distributions of the
distress index were plotted for the two populations at subsequent years after the
preventive maintenance activity. The corresponding reliability values were then
compared.

Based on the above analysis, the distress index guideline values for a non-
structural bituminous overlay (<40), single chip seal (<25), multiple course micro-
surfacing (<30), and bituminous crack seal (<15) were found to be reasonable. The
results for surface milling with a non-structural bituminous overlay (<40) were
inconclusive. This is probably due to the fact that this fix type had the smallest number

of projects compared with the other fix types.

92

0 Evaluation of Ride Quality Index (RQI) Guidelines

The evaluation of the RQI guidelines was done by looking at the distress index at
subsequent years after the fix was applied for two populations:

0 Population One: RQI values less than or equal to the RQI guideline values at

the time of the preventive maintenance activity.

0 Population Two: RQI values greater than the guideline values at the time of

the preventive maintenance activity.

A two sample t-test was then conducted by comparing the mean distress indices at
subsequent years after applying the fix from both populations.

The results for the non-structural bituminous overlay and surface milling with a
non-structural bituminous overlay indicate that the RQI value at the time of the fix does
not seem to affect the future DI values (at least up to five or six years for the bituminous
overlay and up to four years for the mill and fill). This is expected since overlaying the
pavement will smoothen its surface so that roughness is minimized after overlaying. The
results for chip sealing indicate that the RQI guideline value of 54 may be warranted at
later years. The results for micro-surfacing and bituminous crack seal were inconclusive
and more data are needed to confirm or disprove the hypothesis that the RQI level at the

time of PM action is taken will affect future performance of the pavement.

93

7.2

Recommendations for Future Research

The results of this study have led to the following recommendations:

First, additional variables should be investigated. This research studied the
distress index before and after the preventive maintenance action. Additional
variables to be considered would be: Pavement thickness, pavement age,
drainage, construction, materials, and climate. Traffic could possibly be another
variable, although, these fixes are applied mainly on low to medium volume
roads; it may not be a significant factor except for pavement sections with
relatively high traffic levels.

Secondly, more data should be used for plotting the distributions and calculating
the reliability of life expectancies. Additional projects could be added to those
used in this study to increase the database for subsequent years after the initial
treatment. This will increase the dependability of the reliability tables developed
for fixes such as the bituminous crack seal, single chip seal, and multiple course
micro-surfacing. In order to increase the dependability of the reliability tables for
the non-structural bituminous overlay and surface milling with a non-structural
bituminous overlay, additional data are needed at later yearscto evaluate these
treatments from five to ten years.

Thirdly, more data are needed for pavements where the pre-existing condition is
known for the distress and ride quality indices. Having more data will lead to
better results with the two-sample t-test. Also, with the addition of future
performance data, more conclusive results could be reached, especially for the

validity of the RQI guideline values.

94

Fourth, in order to do a project level analysis, one must look at the types of
distresses that predominately occur in preventive maintenance projects and plot
distributions of these different distresses over time while tracking the type of fix
that is performed. From this analysis one can determine the reliability versus time
for particular distresses. Based on some minimum level of reliability one will be

able to estimate life expectancy for a given fix type with a particular distress.

9S

APPENDICIES

96

APPENDIX A

HISTOGRAMS USED TO DETERMINE THE LIFE EXPECTANCY FOR THE
FOLLOWING PREVENTIVE MAINTENANCE (PM) TREATMENTS BASED
ON THE DISTRESS INDEX (DI)

Non-Structural Bituminous Overlay
Surface Milling with a Non-Structural Bituminous Overlay
Single Chip Seal
Multiple Course Micro-Surfacing
Bituminous Crack Seal

97

Non-Structural Bituminous Overlay

 

700

  

 

 

200
5‘
5 100 Sid.DOV'5.38
g- Moan-3.2
u_ o n-mm

 

a w I I V I? I V V V I f V
'0 '0 0.0 qo‘bo fofiofio bofio‘fia fia‘iofio

1 Year after PM

Figure A.1 - Histogram of D1 1 Year after PM

 

1000’

 

5‘ 200

g Std. Dov . 20.58
5- Moon I 10.8

u_ o N - 1232.00

 

a, .,.,..‘3...'.'..-
o (be 42, 00%, a,%,%, 9:, afififi,
2 Years after PM

Figure A.2 - Histogram of DI 2 Years after PM

98

Frequency

 

 

 

 

 

1004
Ski. Dov-8.04
Mom I83

0 N'm.m
00 0’0. Wfié‘t‘b‘fi‘o'qfifi‘fr‘bqro’bfimfifi‘
3 Years after PM

Figure A.3 - Histogram of DI 3 Years afier PM

55‘

: SH. Dev-27.98

8 Mean- 20.7

E ”.0...

 

 

 

o
a re. .5, e ‘0 0,000,009.,
4YearsaflerPM

Figure A.4 - Histogram of D1 4 Years after PM

99

Frequency

 

Std. Dev I 10.77

 

 

Mom I120
N ll440.00

 

 

0 o’qo’a‘b eo%%%%‘%%%%’b%~

'0' Woooooooaooooo

5 Years alter PM

Figure A.S - Histogram of DI 5 Years after PM

Frequency

 

 

Std. Dov-39.77
Moan-28.2
N-345.00

so '0. ., 0,5,0, 0,90, 0,0,0,

 

 

 

6 Years after PM

Figure A.6 - Histogram of D1 6 Years afier PM

100

 

100

 

snow-17.42
Mean . 13.2
‘N -100.00

 

Frequency

 

7 Years after PM

Figure A.7 - Histogram of D1 7 Years after PM

 

7O

 

 

 

Frequency

0, I I e A a
o 60'" 9.7553559350353550 -o -o 332095523320

8 Years afler PM

Figure A.8 - Histogram of D1 8 Years after PM

101

Surface Milling with a Non-Structural Bituminous Overlay

 

 

 

 

3‘

§ Std. Dev I 5.63
8- Mean ' 7.8

u_ N I 210.00

o-oeov-o-o 0,09,70&%¢QV%%

WOOOOOOOOOO

1 Year afier PM

Figure A.9 - Histogram of DI 1 Years after PM

 

 

 

 

 

100
g I sunny-14.99
g- Mean-8.7
u. o NI566.00

4’ 0% 20°00"? 00,000,000‘10‘090‘1b

2 Years efler PM

Figure A.10 - Histogram of D1 2 Years afier PM

102

 

 

SH. Dev- 6.16
Moon I 12.5
N I 160.00

 

 

0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0
2.0 6.0 10.0 14.0 18.0 22.0 28.0 30.0

3 Years efler PM

Figure A.11 - Histogram of D1 3 Years after PM

 

 

100

Std. Dev I 26.02
Mean I 16.7
N I 407.00

”-0 bofiog@%%%%%'q,%o@

.0 so .0 .0 .0 no no .0 no .0

 

 

Frequency

4 Years afler PM

Figure A.12 - Histogram of D1 4 Years afier PM

103

 

Frequency

snow-17.34
Maui-33.0
NI88.00

 

 

 

 

 

Std. Dev 3 4.79
Mom I 19.4
N I 15.00

 

10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5
6 Years after PM

Figure A.14 - Histogram of DI 6 Years after PM

104

 

 

 

 

0.0 5.0 10.0 15.0

7 Years after PM

Figure A.15 - Histogram of D1 7 Years after PM

105

Single Chip Seal

Frequency

Frequency

 

700

200

100

  

Std. Dev I 27.96
Moon I 16.8
N I 1775.00

 

 

1 Year after PM

I

0%ooqo@;‘b;@;°¢;‘bo .fi:‘@%%%%

0'00'0‘0

Figure A.16 - Histogram of D1 1 Years after PM

 

 

Std. Dev I 21.63
Moon I 14.1
N I 1276.00

 

 

0.0%'o%

2 Years after PM

%0%%’%@%%%

'0 '0 '0 '0 '0 '0 '0

Figure A.17 - Histogram of D1 2 Years after PM

106

 

700

      

l

200

100 SH.Dov-43.43
Mean I26.0
N I 1312.00

%%@%%%%R%%%%

'0 '0 '0 '0 '0 '0 '0 '0 '0 ‘0

 

quuency

0

 

3 Year: after PM

Figure A.18 - Histogram of D1 3 Years after PM

 

 

3'

§ 810.0011I 24.29
a- Moon I 15.3

1L 0 N I 623.00

 

a ' I v I v f r— w v v v T
4 Year: after PM

Figure A.19 - Histogram of D1 4 Years after PM

107

 

 

 

 

700

600

500

400

300

200
6‘
g 100 510.0." 130.50
8- Moon I 49.4
“- 0 ........ T - N I 1201’00

q .20 z z a ) )
o '0 920%;2bo‘gfio 030%0 -o -o 552095523920

5 Year: after PM

Figure A.20 - Histogram of D1 5 Years after PM

 

 

SH. Dev I 104.52
Men I 63.0
N I 559.00

 

Frequency

 

 

6 Years after PM

Figure A.21 - Histogram of D1 6 Years afier PM

108

 

180

140

120

100

Frequency
3

0

0.0

 

 

10.0 20.0 30.0 40.0 50.0 80.0 70.0 80.0
5.0 15.0 25.0 35.0 45.0 55.0 85.0 75.0

7 Years after PM

810.0011 I 15.32
Mean I 6.6
N I 306.00

Figure A.22 - Histogram of D1 7 Years afier PM

109

Multiple Course Micro-Surfacing

 

 

 

5‘

5 SM. DOV 3 15.39
g- Mem I 132

LL N .1015“)

 

o%%%%@%e%%%%%%

0'0'.000...000

1 Year after PM

Figure A.23 - Histogram of D1 1 Years afier PM

 

 

 

 

 

5‘
g 3”. Dev 3 16.14
é» Men I 16.2
I: N - 1300.00
Q0 I 1 ) I
Q o‘bo‘bo 00‘50 000 0000-0490: 920 ’0. ago‘b-o'O-o‘bofio

2 Year: after PM

Figure A.24 - Histogram of D1 2 Years afier PM

110

 

k

  

Frequency

0

 

 

86.1.0011 I 27.05
Mom I 19.3
N I 913.00

0_ '2 '2 . .‘3 . - . .
o '00 ‘00 #20 120%., 00%.,“00 320 "be

3 Years after PM

Figure A.25 - Histogram of D1 3 Years after PM

 

 

Frequency

0

 

 

 

Slim-4129
Mom-251
NI976.00

‘-’o @o%o%@'}bebe%‘¢b%'%‘b'%

.0 no no 00 .0 .0 .0 .0 no .0

4 Years utter PM

Figure A.26 - Histogram of D1 4 Years afier PM

111

 

100

Frequency

0

 

Std. Dev I 47.26
Men I 25.3
N I 742.00

 

 

no .0 2». www.mewmfifi ..

5 Years alter PM

Figure A.27 - Histogram of D1 5 Years after PM

 

 

100

Frequency

0

 

$111. Dev I 62.14
Mean I 42.6
N- 539.00

"a “T? 0% 05109122552125 %‘%%%§bo%a

 

 

6 Year: after PM

Figure A.28 - Histogram of DI 6 Years afier PM

112

 

100

  

Frequency

‘.

7 Years after PM

810. DeVI61.65
Mom I 30.6

 

‘N 3111.“)

003.513; {,2 z) ' 'e a) a a)
o ‘0 ‘0 '0 Q’0 ¢o%o"%% 1% qfio‘e’fibo 63:90

Figure A.29 - Histogram of D1 7 Years afier PM

113

Bituminous Crack Seal

 

1000

8004

 

400

 
 

i

 

  

 

5' 200

g Std.DeVI11.55
§ MeanI11.1

m 0 ‘ n- 1077.00

0%%%%055%%%

"000"'000

1 Year after PM

Figure A.3O - Histogram of DI 1 Years after PM

 

1200

 

 

 

6'

g 510. Dev - 19.01
g. Mean . 10.4

11. N - 2590.00

 

402.7% “‘0 “’0’ Q; a; ’0 5b 905009 5030:0500

0000 0000

2 Years afler PM

Figure A.31 - Histogram of D1 2 Years after PM

114

 

700 ’

SH. Dev- 24.32
Mean I 17.6
N I 1360.00

 

 

Frequency
‘8‘

 

%%%%%%%%%%%%

'0 '0 '0 ‘0 '0 '0 '0 '0 '0

3 Years after PM

Figure A.32 - Histogram of D1 3 Years afier PM

 

12“?

1M

 

 

$10. Dev . 45.17
Mean - 20.4
N - 2472.00
Q I I 1 2
. 1% 5.551355525535525.
4 Years after PM

Figure A.33 - Histogram of D1 4 Years after PM

115

 

 

Std. Dev I 31.41
Mean I 24.4
N I 1370.00

 

 

 

5 Years alter PM

Figure A.34 - Histogram of D1 5 Years afier PM

 

 

 

 

6‘

g ‘00 snow-01.15
g- Mean-42.0
u_ o u-1oe4.00

‘-’o 1% ’42., {aflofiaawfififio
6 Years alter PM

Figure A.35 - Histogram of DI 6 Years after PM

116

 

200

 

 

5‘

g snow-00.40
§ Mean-52.3

u‘: o ‘NI828.00

ea 00501512015150" 05151510305030

.0 .0 .0 .0 .0 .0 .0 .0 .0 .0 .0 4
7 Years after PM

Figure A.36 - Histogram of D1 7 Years after PM

 

 

6'
§ Std. Dev I 321.00
3. Mean I 416.0
1.1. N I 63.00
'0 '0 'fio

 

6 Years alter PM

Figure A.37 - Histogram of D1 8 Years after PM

117

APPENDIX B

HISTOGRAMS USED TO EVALUATE THE DISTRESS INDEX (DI)
PREVENTIVE MAINTENANCE (PM) GUIDELINE VALUES FOR THE
FOLLOWING PM TREATMENTS

Non-Structural Bituminous Overlay
Surface Milling with a Non-Structural Bituminous Overlay
Single Chip Seal
Multiple Course Micro-Surfacing
Bituminous Crack Seal

118

Non-Structural Bituminous Overlay

 

 
 

Std. Dev I 4.67
'- Mann - 2.9
0 K N - 540.00

40 ‘90 ‘30 q‘0 Q0 ’Qo’eo’t ’3? ’6? ‘52 SPO

 

Frequency

1 Year after PM

Figure B. 1- Histogram of D1 1 Year after PM when the pre-existing condition is less than
the guideline value

 

  

10

SM. Dev-4.63
man-5.6
NI 119.00

   

 

‘

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 10.0
1.0 3.0 5.0 7.0 9.0 11.0 13.0 15.0 17.0

Frequency

 

0

1 Year after PM

Figure B.2 - Histogram of DI 1 Year afier PM when the pre-existing condition is greater
than the guideline value

119

 

140‘

120

 

 

5'

5 sun. Dev - 10.53
g- Mun- 10.6

E ‘N-545.00

qo¢o’0’d“%%“bqf%%%%%

.0 .0 .0 .0 .0 .0 .0 .0 .0 .0 .0
2 Years after PM

Figure 8.3 - Histogram of D1 2 Years after PM when the pre-existing condition is less
than the guideline value

 

100

140

 
   

 

so
60
40
E
g 20 snow-11.16
3 k Mum-9.2
E 0 ‘N-424.00

    

Qo’qo%*b‘b'%%"o%%’%_o

‘0 ‘0 '0 '0 '0 '0 '0 '0
2 Years after PM

Figure 3.4 - Histogram of D1 2 Years after PM when the pre-existing condition is greater
than the guideline value

120

 

120

 

 

5‘

5 Std. Dav - 3.72
§ Mam - 4.8

E N - 404.00

 

qerqeirrrr a
0 0 0 0 0 0.0 {’0 to Q0 040%0§0 to

3 Years anor PM

Figure B.5 - Histogram of D1 3 Years afier PM when the pre-existing condition is less
than the guideline value

 

12

10‘

 

 

 

6‘
5 SM. [hv I 5.86
é um - 0.4
m N-68.00
a0
3 Year: after PM
Figure 8.6 - Histogram of D1 3 Years afier PM when the pre-existing condition is greater
than the guideline value

121

 

120

1W'

 

Std. Dav I21.18
mat-19.0
N-528.w

fl I I
o‘bo‘bo 'o-o

 

  

Frequency

 

0'0 lo-o‘b-o ‘b-o 'o-o %o%a’bo%o $0 @060
4 Year: after PM

Figure 3.7 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value

 

 

8U. m I 19.33
M I 21.7
N - 270.00

 

  

L

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 ”.0 ”.0
5.0 15.0 25.0 35.0 45.0 55.0 05.0 75.0 85.0

4 Years after PM

Figure B.8 - Histogram of D1 4 Years after PM when the pre-existing condition is greater
than the guideline value

122

 

 

 

 

 

 

 

 

5‘ 1o
5 en. Dav - 000
(:7 k M . 8.7
E o } M24200
0-0 ''0 Q0 {’0 ’4‘0 Q20 etc ‘50 {’0 "3.90 '00
5 Years afler PM
Figure B.9 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value
30
20d
10 I
6‘
5 $111. Dov -1e.35
§ Mann - 200
u“. o N-70.00
5.0 15.0 25.0 35.0 45.0 55.0 05.0 75.0
10.0 20.0 30.0 40.0 50.0 00.0 70.0
5 Years after PM
Figure B. 10 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value

123

 

10!

 

 

 

6‘
5 8111. cu - 2015
§ Mam-37.0
a: 0 _ M 111.00
0'0 ’Qo éb-o ”ha ’90 $30 $0 ’b-o $0 $0 ’90 "520
6 Years alter PM
Figure 3.11 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value

 

 

sum-110.50
Mun-131.5
N'1ifl)

 

50.0 1WD 150.0 2WD 250.0 “.0 350.0

6 Years after PM

Figure B.12 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value

124

 

304

204

  

1Ol

 

 

 

 

 

 

 

 

5‘
5 S10. Dav - 9709
§ h Mam - 00.4
E 0 M9500
"’025*?020231222223132‘34323232‘222 .3230
8 Years alter PM
Figure B. 1 3 - Histogram of D1 8 Years afier PM when the pre-existing condition is less
than the guideline value
7 _
6‘
8 S“. W 8 290.30
i. Mam - 201.1
I: N I 0.“)
0.0 2000 400.0 000.0 000.0
8 Year: after PM
Figure B. 14 - Histogram of D1 8 Years after PM when the pre-existing condition is
greater than the guideline value

125

Surface Milling with a Non-Structural Bituminous Overlay

 

14

124

ml

84

 

  

Sid. Dov I 5.95
than I 6.1
N I 47.00

 

 

L

0.0 5.0 10.0 15.0 20.0 25.0 ”.0
2.5 7.5 12.5 17.5 22.5 27.5

Frequency

1 Year after PM

Figure B. 1 5 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value

 

 

snow-4.4a
mun-5.6
N-59.00

 

Frequency

 

1 Year after PM

Figure B. 16 - Histogram of D1 1 Year after PM when the pre-existing condition is greater
than the guideline value

126

 

140

120‘

 

 

6‘

5 snow-7.77
§ MIBJ

I: M49000

 

'0 ‘0 ’0 ’6. ‘% % % ybovofio®ofio§0®0

o o
.o -o -o -o -o
2Years amerPM

Figure B. 17 - Histogram of D1 2 Years after PM when the pre-existing condition is less
than the guideline value

 

 

6‘

5 snow-0.2a
§ war-5.5

i- M7100

 

0.0 4.0 0.0 12.0 10.0 20.0 24.0
2.0 0.0 10.0 14.0 10.0 22.0 20.0

2 Years after PM

Figure B. 1 8 - Histogram of D1 2 Years afier PM when the pre-existing condition is
greater than the guideline value

127

 

10

14

 

 

6‘
5 Std. (by I 5.00
§ Mun - 0.3
1: N . 50.00
0.0 4.0 8.0 12.0 10.0 20.0
2.0 0.0 10.0 14.0 18.0
3 Years alter PM
Figure 3.19 - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value

 

 

 

snow-4.00
man-5.8
M2000

 

Frequency

0.0 2.0 4.0 0.0 0.0 10.0 12.0 14.0 10.0 10.0

3 Years after PM

Figure B.20 - Histogram of D1 3 Years after PM when the pre-existing condition is
greater than the guideline value

128

 

snow-0.73
wean-15.8
‘N-355.00

ea '00 ‘22., ~00 '40 ‘0, “be ’00 420 ‘50 ’420’20

 

 

 

4 Years after PM

Figure B.21 - Histogram of D1 4 Years afier PM when the pre-existing condition is less
than the guideline value

 

 

 

 

6‘
5 Std. Dav a 10.00
§ mm . 10.5
I: N I 22.00
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0
4 Years after PM
Figure 3.22 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value

129

 

Std. Dev = 36.80
Mean = 48.3
N = 500

Frequency

 

20,0 40 o
5 Years altar PM

Figure B.23 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value

 

 

_n

swim-30.00
lbw-30.4
N-11.00

Frequency

0

 

0.0 20.0 40.0 00.0 00.0 100.0 120.0

5 Years alter PM

Figure B.24 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value

130

 

Std. Dev =11.40
man=21.8
N=2,00

Frequency

   

10.0 20.0
7 Years afler PM

Figure 8.25 - Histogram of D1 7 Years after PM when the pre-existing condition is less
than the guideline value

 

 

 

Std. Dav I4.19
Man I6.6
N-7.oo

Frequency

 

7 Years afler PM

Figure B26 - Histogram of D1 7 Years after PM when the pre-existing condition is
greater than the guideline value

131

Single Chip Seal

 

120*

 

 

6‘

5 Std. Dev - 21.37
g- man I191

'u: NISOODO

o )
'0 ’0-0 (b0 ub-o 'o-o Q30 {’0 0-0 4?0 $0 ,4?0 60.0

1 Year after PM

Figure B.27 - Histogram of D1 1 Year afier PM when the pre-existing condition is less
than the guideline value

 

NIH

 

 

 

5‘
S sum-32.83
§ M=ZEA
LT. Naaszm
Qleebeeeeerlzrrrzor
o 0. ..... 0. . . I
0 0 0 0 0 0 0 0 -oo-o‘boub-o’o-o%ooo-oo-ooo-o

1 Year after PM

Figure B.28 - Histogram of D1 1 Year afier PM when the pre-existing condition is greater
than the guideline value

132

 

120‘

100

 

 

 

 

 

6'
5 Std. Dav - 9.18
g. man - 8.1
E M43200
40 to Q0 {5’0 3‘0 ‘50 97.0 ‘30 ‘Q’o ‘bo ‘20
2 Years utter PM
Figure 3.29 - Histogram of D1 2 Years after PM when the pre-existing condition is less
than the guideline value
6‘
5 Std. Dev - 23.25
§ Mun - 16.7
E y N - 613.00
0'0 ’00 ‘bo ‘b-o '0-0 $30 %30 ’b-o 0% aha I‘bOI’Qoi‘bol‘bo
2 Years after PM
Figure B.30 — Histogram of D1 2 Years after PM when the pre-existing condition is
greater than the guideline value

133

 

 

 

Std. Dev - 15.69
than: 14.8
NI 512.00

-o ’0.0 ‘50 ‘50 9.0 {’0 $30 1’0 $0 $0 #20

 

 

Frequency

0

3 Year: after PM

Figure B.3l - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value

 

m!

not

201
snow-33.90

mar-39.2
‘N-aeam

 

 

 

 

Frequency

0 L
0'0 %obo%0%0 ,%o’%o’%o%0’%0‘%§b:%o%o%fio
3 Years after PM

Figure B.32 - Histogram of D1 3 Years after PM when the pre-existing condition is
greater than the guideline value

134

D
’

Std. Dev I849
Mam=7.1
1N=94.00

   
    

 

Frequency

QéYQQIIIII
0000000€020¢0¢0Vba

4 Years afler PM

Figure B.33 - Histogram of D1 4 Years afier PM when the pre-existing condition is less
than the guideline value

 

 

Std. Dev = 9.78
1 Mean = 15.2
N= 151.00

Frequency

h
0-0 '-0 Q0 ’éo ’4‘0 ‘50 910 ‘3’0 ‘z-Z’o 4530 '90
4 Years afler PM

Figure B.34 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value

135

 

 

 

E
3
5
ll.
5 Years alter PM
Figure B.35 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value

 

 

Std. Dev I 62.42
men I 56.4
N-196.00

 

 

 

5 Years alter PM

Figure B.36 - Histogram of D1 5 Years afier PM when the pre-existing condition is
greater than the guideline value

136

 

100

m1

 

401

Std. Dev I 24.67
Mean I 24.9
N I 303.00

 

 

Frequency

0 ..
%%%%%@e%e@%

’0 '0 '0 .0 .0 .0 .0 .0 .0
6 Years alterPM

Figure B.37 - Histogram of DI 6 Years afier PM when the pre-existing condition is less
than the guideline value

 

Std. Dev I 86.46
when I 78.7
N I 45.00

 

 

 

Frequency

0

0.0 50.0 100.0 150.0 2WD 250.0 3ND
25.0 75.0 125.0 175.0 225.0 275.0 325.0

6 Year: after PM

Figure B.38 - Histogram of D1 6 Years afler PM when the pre-existing condition is
greater than the guideline value

137

 

 

 

5‘

5 Std. Dev = 0.06
“3" Mean - 5.1

E o N - 09.00

7 Years alter PM

Figure B.39 - Histogram of D1 7 Years afier PM when the pre-existing condition is less
than the guideline value

 

Std. Dev =4.81
Mean =4.7
N= 22.00

Frequency

 

7 Years alter PM

Figure B.4O - Histogram of D1 7 Years after PM when the pre-existing condition is
greater than the guideline value

138

Multiple Course Micro-Surfacing

 

70

 

 

6‘

5 sea. Dev -e.o1
8- Mun - 123
O

L: M40400

0-0 70 Q0 ’90 ’330 ‘50 ‘3'0 1&0 Q‘o 430 '00
1 Year alter PM

Figure 3.41 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value

 

 

 

 

Frequency

5.0 15.0 25.0 35.0 45.0 55.0 05.0 75.0
10.0 20.0 30.0 40.0 50.0 00.0 70.0

1 Year alter PM

Figure B.42 - Histogram of D1 1 Year after PM when the pre-existing condition is greater
than the guideline value

139

 

 

 

 

 

6‘
E, sum-9.99
3 marl-10.3
0
u: N-ezmo

0 6‘ I

o o qo '00 1'60 ‘60 420 '60 ‘60 11:0 4‘00 111:0

2 Years alter PM

Figure 3.43 - Histogram of D1 2 Years afier PM when the pre-existing condition is less
than the guideline value

 

6‘

5 5m. Dev = 17.09
g Wan = 27.4

E N= 34.00

 

2 Years after PM

Figure B.44 - Histogram of D1 2 Years afier PM when the pre-existing condition is
greater than the guideline value

140

 

100-

 

8111. Dev I 16.97
Mean I 15.0
N I $0.00

 

 

Frequency

0
0-0 "7 o’QWWWbfi‘fi‘h’Q'c‘b‘fi‘fifi‘o’bfiQQt

3 Years alter PM

Figure 3.45 - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value

 

  

Std. Dev I 36.53
than I 73.4
NI45.00

 

 

 

I _

0.0 40.0 ND 120.0 1ND 2010
20.0 00.0 1ND 140.0 1&10 220.0

Frequency

0

3 Year: after PM

Figure 3.46 - Histogram of D1 3 Years afier PM when the pre-existing condition is
greater than the guideline value

141

 

2001

 

  

 

 

 

 

 

 

100 1
6‘
g sun. M - 15.14
g- m I 14.5
11'. 0 r _ ___ _ _N-094.00
0'0 ’90 Q’o “be '90 $30 Q30 Ibo $30 $0 ’910 60.0
4 Years alter PM
Figure B.47 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value
30
6‘
5 Std. Dev - 21.03
3 man - 59.0
LL n- 141.00
0'0 ,o-o Q90 ub-o 1b-o “be 0% )0-0 $30 ‘b-o @J’Qo {to
4 Years alter PM
Figure 3.48 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value

142

 

 

AM
5w

 

 

6'

5 s10. Dev -21.39
§ Mae-1:105

l: N-564.00

q0€b~ovoo%o%o’%%lvo@%%%

'0 0 '0 ‘0 '0 '0 ‘0
5Years aflerPM

Figure B.49 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value

 

Std. Dev = 33.25
Mean = 57.0
N = 18.00

Frequency

00 20.0 40.0 60.0

 

5 Years after PM

Figure B.50 - Histogram of DI 5 Years afier PM when the pre-existing condition is
greater than the guideline value

143

 

 

 

 

 

 

 

 

6‘
5 310. cu - 30.07
g Mam - 31.4
E 01-45700
0'0 ‘bo '0-0 630 Q1’0 “20 {to 49.0 (‘20 {be (2%
6 Years alter PM
Figure B.51 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value
16
6‘
8 $111. Dev - 112.23
g Mm - 107.2
u_ N - 37.00
6 Years after PM
Figure B52 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value

144

Bituminous Crack Seal

 

 

 

 

6

5 Std.Dev=9.37
2") Mum-11.2
i M99000

0 7 0 I I ‘3
ooo%%%%%%%%%%

1 Year after PM

Figure 3.53 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value

 

sm. Dev I 14.85
M I 21.8
NI 159.00

 

 

Frequency

 

0.0 10.0 20.0 30.0 40.0 50.0 ”.0 70.0 ”.0 90.0
5.0 15.0 25.0 35.0 45.0 55.0 05.0 75.0 05.0 ”.0

1 Year alter PM

Figure 3.54 - Histogram of D1 1 Year after PM when the pre-existing condition is greater
than the guideline value

145

 

 

1”

Frequency

0

0-0

 

I

 

snow-11.79
men-13.3
NI2010.00

» a

2 Year: after PM

Figure B.55 - Histogram of D1 2 Years after PM when the pre-existing condition is less

than the guideline value

 

 

 

 

6‘
5 $10. Dev - 21.23
§ Mun - 27.9
E ‘ N - 340.00
0'0 {’0 $30 ‘bo 7% 6‘00 $20 "0.0 Q?o Q30 "boi’qoibol‘bo
2 Years alter PM
Figure 8.56 - Histogram of DI 2 Years after PM when the pre-existing condition is
greater than the guideline value

146

 

 

Stu. Dev - 13.33
that I 18.1
‘N-01o00

0‘0 ’0-0‘2’ "b '0 % $2o’b-o‘bo Q20 [92060.0]? ,‘5

’0 '0 '0 ‘0 o .0

 

 

Frequency

3 Years alter PM

Figure 8.57 - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value

 

SM. my I 19.18
men I31.4
NI78.00

 

 

Frequency

 

0.0 10.0 20.0 30.0 40.0 50.0 ”.0 70.0 ”.0
5.0 15.0 25.0 35.0 45.0 55.0 05.0 75.0

3 Years alter PM

Figure B.58 - Histogram of D1 3 Years after PM when the pre-existing condition is
greater than the guideline value

147

 

 

8111. my I 16.34
than I 16.4
N I 1531.00

0'0 2% (be ‘50 '0-0 *20 Q’o )Qo Q~’o {’0 ’90 60.0

 

 

Frequency

0

4 Years after PM

Figure B.59 - Histogram of D1 4 Years afier PM when the pre-existing condition is less
than the guideline value

 

1”

 

  

 

 

6‘ 201
= $111. Dev . 40.05
g- Ibm I 39.3
I: 0 ‘ N I 30000
to1o'Qo‘O-a‘éfiefiérefizas'bfijamfifi.
4 Years alter PM
Figure B.60 - Histogram of D1 4 Years afier PM when the pre-existing condition is
greater than the guideline value

148

 

1”

 

 

 

 

 

 

 

6‘
5 313. cu - 17.13
g Mann - 20.5
11.‘ Naeo400
0'0 ,0-0 $30 ube 1% Q70 Q30 ’b-o Q70 ‘b-o @J’Qol‘bol‘bo
5 Years alter PM
Figure B.61 - Histogram of D1 5 Years afier PM when the pre-existing condition is less
than the guideline value
20
6‘
5 sun. Dav - 22.70
§ um - 37.9
0“. n-aono
5.0 15.0 25.0 35.0 45.0 55.0 65.0 75.0 85.0
10.0 20.0 ”.0 40.0 50.0 ”.0 70.0 ”.0 ”.0
5 Years after PM
Figure B.62 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value

149

 

2W4

 

 

   

 

 

 

g 8111. Dev - 39.59
g- Mm - 34.0
I: 0 01-07400
0‘0 ‘bo vo-o ‘fio 0% IQ?0 {to “0.041% @0'200‘150
6 Years alter PM
Figure B.63 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value
6‘
5
3
E
0.0 50.0 1”.0 150.0 2”.0 250.0 300.0
25.0 75.0 125.0 175.0 225.0 275.0
6 Years alter PM
Figure B.64 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value

150

 

810. m 8 41.30
Mean I583
141319.”
%%%%%%%%@%%%%

'0 '0 '0 '0 .0 .0 .0 .0 .0 .0 .0 .0

 

 

 

Frequency

7 Years alter PM

Figure B.65 - Histogram of D1 7 Years afier PM when the pre-existing condition is less
than the guideline value

 

14

Std. Dev-27.15
mat-60.3
NI39.00

 

 

 

10.0 30.0 50.0 70.0 ”.0 110.0 130.0 150.0
20.0 40.0 ”.0 ”.0 1”.0 120.0 140.0 1”.0

7 Years alter PM

Figure 8.66 - Histogram of D1 7 Years after PM when the pre-existing condition is
greater than the guideline value

151

 

 

APPENDIX C

HISTOGRAMS USED TO EVALUATE THE RIDE QUALITY INDEX (RQI) PM
GUIDELINE VALUES FOR THE FOLLOWING PREVENTIVE
MAINTENANCE (PM) TREATMENTS BASED ON THE DISTRESS INDEX (DI)

Non-Structural Bituminous Overlay
Surface Milling with a Non-Structural Bituminous Overlay
Single Chip Seal
Multiple Course Micro-Surfacing
Bituminous Crack Seal

152

Non-Structural Bituminous Overlay

 

20

10

Frequency

  

L

  
 

 

  

o s? 1_ 63 63 I I I I I
'0 0 0 0 0 0.0 ea to 0.0 Q0 ‘bo §o

1 Year after PM

Std. Dev I 5.03
M I 4.9
N I 232.00

Figure C.1 - Histogram of D1 1 Year afier PM when the pre-existing condition is less

than the guideline value

 

Frequency

 

 

 

0.0 2.5 5.0 73 15.0 15.5 150 17.5 20.0

1 Year afier PM

 

SM. Dev 87.52
mm 85.1
I". 13.00

Figure C.2 - Histogram of D1 1 Year after PM when the pre-existing condition is greater

than the guideline value

153

 

 

 

 

6‘

5 Std. Dav - 6.04
3 Mann = 6.8

E N - 470.00

90907035090 ’0 <3 ’1 6' 9%‘3’ €,%%

H00H00H000'0‘0'0

2 Years after PM

Figure C.3 - Histogram of D1 2 Years alter PM when the pre-existing condition is less
than the guideline value

 

Frequency

 

2 Years afler PM

Figure C.4 - Histogram of D1 2 Years afier PM when the pre-existing condition is greater
than the guideline value

154

 

10

Frequency

0 h
0.0 4.0 8.0 12.0 10.0 20.0 24.0 20.0 32.0 30.0

 

 

Std. Dav I 8.01
khan I 8.8
N I 233.00

2.0 0.0 10.0 14.0 18.0 22.0 20.0 3.0 34.0 30.0

3 Years aner PM

Figure C.5 - Histogram of D1 3 Years after PM when the pre-existing condition is less

than the guideline value

 

Frequency

3 Years afler PM

 

 

snow-9.99
MIaz
NI11.00

Figure C.6 - Histogram of D1 3 Years afier PM when the pre-existing condition is greater

than the guideline value

155

 

 

 

 

    

 

601

501

40

so

20
5‘
5 snow-7.11
3 1° Mesa
5 L '
u. o stuoo

0.0 4.0 8.0 12.0 18.0 20.0 24.0 28.0 32.0 ”.0
2.0 8.0 10.0 14.0 18.0 22.0 28.0 30.0 34.0

4 Years after PM

Figure C.7 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value

 

10

14!

12‘

10!

84

 

 

Stow-19.43
mart-25.2
NI41.00

Frequency

 

0.0 10.0 20.0 30.0 40.0 50.0 ”.0 70.0 ”.0 ”.0

4 Years after PM

Figure C.8 - Histogram of D1 4 Years afier PM when the pre-existing condition is greater
than the guideline value

156

 

 

201

10!

 

 

 

 

 
 

 

 

8‘
5 Std. Dav - 0.71
i “an I 8.1
E o 11- 100.00
0-0 80 70 630 ‘90 ’00 [ea "to QOQO‘POQ’OQO'PO‘PO‘PO
5 Years after PM
Figure C.9 - Histogram of DI 5 Years after PM when the pre-existing condition is less
than the guideline value
a .
5
4
3
2
6‘ 1
5 Std. W I 10.70
§ Mann . 13.0
11“. 0 NI 11.00
5.0 10.0 15.0 20.0 .0 30.0 35.0
5 Years after PM
Figure 010 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value

157

 

     

_s
0

Std. Dev = 7.81
Mean =14.9
N= 213.00

 

s

 

Frequency

 

O
0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0 32.0 36.0
2.0 8.0 10.0 14.0 18.0 22.0 28.0 30.0 34.0

6 Years after PM

Figure C.11 - Histogram of D1 6 Years afier PM when the pre-existing condition is less
than the guideline value

 

Frequency
in

9
o

 

6 Years after PM

Figure C.12 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value

158

Surface Milling with a Non-Structural Bituminous Overlay

 

 

 

Frequency

 

4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0
5.0 7.0 9.0 11.0 13.0 15.0 17.0 18.0

1 Year after PM

Figure C.13 - Histogram of D1 1 Year afier PM when the pre-existing condition is less
than the guideline value

 

 

 

8‘

5 Std. Dev I 3.25
car Mean = 12.5
E N = 26.00

4.0 0.0 so
1 Year afler PM

Figure C. 14 - Histogram of DI I Year after PM when the pre-existing condition is greater
than the guideline value

159

 

10

snow-0.70
mat-8.8
NI 115.1!)

 

 

Frequency

0
0.0 5.0 10.0 15.0 20.0 25.0 ”.0 35.0 40.0 45.0

2.5 7.5 12.5 17.5 22.5 27.5 32.5 37.5 42.5
2 Years afler PM

Figure C. 15 - Histogram of D1 2 Years afier PM when the pre-existing condition is less
than the guideline value

 

 

 

Frequency

 

 

2 Years after PM

Figure C.16 - Histogram of D1 2 Years afier PM when the pre-existing condition is
greater than the guideline value

160

 

 

 

Frequency

0.0 4.0 8.0 12.0 16.0 20.0 24.0 28.0
2.0 6.0 10.0 14.0 18.0 22.0 28.0 30.0

3 Years after PM

Figure C.17 - Histogram of D1 3 Years alter PM when the pre-existing condition is less
than the guideline value

 

Frequency

 

0.0 5.0 10.0 15.0 20.0 25.0 .
2.5 7.5 12.5 17.5 22.5 27.5 32.5

3 Years after PM

Figure C.18 - Histogram of D1 3 Years after PM when the pre—existing condition is
greater than the guideline value

161

 

Std. Dev I 7.81
than = 12.5
N = 73.00

 

 

Frequency

 

0.0 5.0 10.0 15.0 20.0 25.0 ”.0 35.0 40.0
2.5 7.5 12.5 17.5 22.5 27.5 32.5 37.5

4 Years alter PM

Figure C.19 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value

 

 

  

6'

5 Std. Dev =9.55
é- Mean =101

‘LC 0 N= 27.00

 

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0
4Years aflerPM

Figure C.20 - Histogram of DI 4 Years after PM when the pre-existing condition is
greater than the guideline value

162

 

snow-0.05
man-29.9

 

 

Frequency

 

5 Years alter PM

Figure C.21 - Histogram of D1 5 Years after PM when the pre-existing condition is less
than the guideline value

 

 

 

 

6‘
5 Std. Dev - 2.32
§ um - 429
E 51-3200
5 Years alter PM
Figure C.22 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value

163

 

3.5

 

 

Std. Dev = 5.43
When = 18.9
N: 8.00

Frequency

 

6 Years after PM

Figure C.23 - Histogram of D1 6 Years after PM when the pre-existing condition is less
than the guideline value

 

 

 

5‘

5 Std. Dev I 4.28
3- Mun - 20.0
E N - 7.00

 

15.0 17.5 20.0 22.5 25.0 27.5

6 Years after PM

Figure C.24 - Histogram of DI 6 Years after PM when the pre-existing condition is
greater than the guideline value

164

Single Chip Seal

 

120‘

Std. my I20.71
M8193
NI021.00

,\
0‘0 ’00 'b-o Q-’o '0-0 ‘20 630 0'0 $20 $0 I(be I,0-0

 

 

1 Year after PM

Figure C.25 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value

 

 

 

5‘

5 Std. Dev I 22.41
§ Man - 24.2

1‘: N - 220.00

qorqoeeeeeeeeaorbog

-o -o -o -o -o -o -o -o .0
1Year after PM

Figure C.26 - Histogram of D1 1 Year after PM when the pre-existing condition is greater
than the guideline value

165

 

120‘

1”

 

 

4o
5‘
5 2° 810.001: -0.00
§ Mann - 10.3
t: 0 NI 040.00

0.040105023050500

-o -o -o -o -o -o -o -o -o -o
2 Years after PM

Figure C.27 - Histogram of D1 2 Years afier PM when the pre-existing condition is less
than the guideline value

 

 

 

 

 

13‘

5 sum-11.07
§ Mun-11.5

E . M10000

2 Years after PM

Figure C.28 - Histogram of D1 2 Years after PM when the pre-existing condition is
greater than the guideline value

166

 

 

 

6‘

5 sun. Dev - 20.37
g. Mm - 37.0

E ____ N - 571.00

 

 

0'0 ‘Yovqofi3o‘bo’Q 4% ,%’% @%%€%

.0 .0 .0 .0 4o .0 .0 no
3 Year: alter PM

Figure C.29 - Histogram of D1 3 Years after PM when the pre-existing condition is less
than the guideline value

 

 

10

 

 

 

6‘
5 310. 04v - 21.31
§ um - 32.7
E NI 100.00
0'0 ,0-0 "be ubo 1'0-0 ob-o $2a ’b-o 4?0 $0 ’90 60.0
3 Years after PM
Figure C.30 - Histogram of D1 3 Years after PM when the pre-existing condition is
greater than the guideline value

167

 

120‘

1”

 

 

40
6‘
5 20 $10. Dev I 10.21
§ Mam - 9.0
a: 0 NI 373.00

0'0 6:0 ,QoQo‘toQ‘o‘bofiobo%0%0%0%0%'0’b0)¢0%0

4 Years alter PM

Figure C.31 - Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value

 

 

 

 

6

5 $10. Dev - 13.93
g mu:- 17.2

I: N'214.”

0. 7 I‘ I
0 0-0 éb-o ‘bo 'O-o (be 6i0 0-0 $30 1’0 420

4 Years after PM

Figure C.32 - Histogram of D1 4 Years afier PM when the pre-existing condition is
greater than the guideline value

168

 

 

 

200!

1”

 

 

Frequency

0

Qa%%<1;’db%%%%%'¢b‘o'%

Std. Dev I 53.00
mat-30.4
MIND.”

.0 .0 .0 .0 .0 .0 .0 no no no

5 Years alter PM

Figure C.33 - Histogram of D1 5 Years after PM when the pre-existing condition is less

than the guideline value

 

101

 

 

Frequency

0

snow I45.13
mat-59.5
NI 121.00

 

0.0 40.0 80.0 120.0 1”.0 2”.0 240.0 2”.0

20.0 ”.0 1”.0 140.0 1”.0 220.0 280.0

5 Years alter PM

Figure C.34 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value

169

 

 

 

 

6‘
5 Std. M I 02.75
‘3’ Mann . 01.9
E N I 324.”
Q%%%%IIIII e
0 '0 '0 '0 '0 %0‘%0%0%é%0 .fiobo .0 .0 .0 .0 fake

6 Years alter PM

Figure C.35 - Histogram of D1 6 Years afier PM when the pre-existing condition is less
than the guideline value

 

u-

m

sumuflm
men-41.3
N827.”

 

 

Frequency

 

 

0.0 50.0 ' 100.0 ' 150.0 ' 200.0 ' 250.0 300.0
25.0 75.0 125.0 175.0 225.0 275.0

6 Years alter PM

Figure C.36 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value

170

Multiple Course Micro-Surfacing

 

Frequency

 

 

Std. Dev I 7.91
than 812.0
N- 413.00

a. Y or r r e
o o 060 Qo‘to YOQO‘PO‘Q‘O'QO

1 Year after PM

Figure C.37 - Histogram of D1 1 Year after PM when the pre-existing condition is less

than the guideline value

 

Frequency

1 Year afler PM

Std. m- 11.03

 

 

M3103
I‘d-58.00

Figure C38 - Histogram of Di 1 Year after PM when the pre-existing condition is greater

than the guideline value

171

 

 

20

 

 

 

 

 

 

6‘
8 sun. Dav - 9.02
§ Mm - 10.0
E o M41300
90 70 Q0 {.30 ’630 1’0 97.0 ‘30 ‘P0 ‘50 ‘20 '20
2 Years afler PM
Figure C.39 - Histogram of D1 2 Years after PM when the pre-existing condition is less
than the guideline value
30
3‘
8 SH. 0011 . 22.80
§- Mun - 30.2
E N-182.m
:20 ’0-0 {to Q30 '00 ‘bo Q30 )o-a $30 Qa @00qu0
2 Years after PM
Figure C.40 - Histogram of D1 2 Years after PM when the pre-existing condition is
greater than the guideline value

172

 

100

 

 

 

Frequency

0 7 0 o I I I I 7 ‘3
'oeo '0 ‘0 '0 qo‘9070qoqo‘b-33‘3075b33-‘o‘b-o‘éfifia‘ka

3 Year: after PM

Figure C.41 - Histogram of D1 3 Years after PM when the
than the guideline value

pre-existing condition is less

 

Frequency

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0
2.5 7.5 12.5 17.5 22.5 27.5 32.5 37.5 42.5

3 Years after PM

 

Std. Dev I 8.71
than I 8.9
N I 52.00

 

Figure C.42 - Histogram of D1 3 Years afier PM when the pre-existing condition is
greater than the guideline value

173

 

100

001

 

   

 

 

 

 

 

5‘
5 Std. Dav - 9.91
3 Mann - 10.2
g 0 K __ N 3 401.00
0-0 4‘0 ’00 <50 #20 '31.}, “ha 4:0 120 1:0 "5.0 €10 “b0 “fro
4 Year: after PM
Figure C.43 — Histogram of D1 4 Years after PM when the pre-existing condition is less
than the guideline value
40
30 I
20
10
6‘
5 sea. Dev - «.54
§ Mun - 35.9
E 0 _ n- 159.00
0.0 ‘b0 b0 Q20 $20 @0%0%0%0Q?0%0§209%20%0
4 Year: after PM
Figure C.44 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value

174

 

140‘

 

120
100
90
90
40
6’
5 20 mow-22.04
g. man-17.3
E o ‘N-aesm

 

0'0 %o%o%oq?o,%%’%@%%%

'0 '0 '0 '0 '0 '0 '0
5 Year: after PM

Figure C.45 - Histogram of D1 5 Years afier PM when the pre-existing condition is less
than the guideline value

 

12

 

 

 

6‘

5 so. an - 14.04
§ M I 19.5

m N - «.00

0.0 10.0 20.0 30.0 40.0 50.0 ”.0
5.0 15.0 25.0 35.0 45.0 55.0 05.0

5 Year: after PM

Figure C.46 - Histogram of D1 5 Years afier PM when the pre-existing condition is
greater than the guideline value

175

 

20

Frequency

0

 

 

810.091! I 35.39
men I 31.7
NI 243.00

0'0 %o%o%o%o,%%,'0%%%¢b

'0 '0 '0 '0 '0 '0 '0

6 Years after PM

Figure C.47 - Histogram of D1 6 Years after PM when the pre-existing condition is less

than the guideline value

 

10!

 

Frequency

0

 

sum-53.99
men-00.9

 

111-11600

0.0 40.0 80.0 120.0 100.0 200.0 240.0 2&0

20.0 00.0 1WD 140.0 1ND 220.0 200.0

6 Years after PM

Figure C.48 - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value

176

Bituminous Crack Seal

 

190

1401

120

 

 

6'

5 SM. Dev I 10.27
3» men I 11.0

o

I: N- 873.00

 

0 7 0 I I <3 1.?

1 Year after PM

Figure C.49 - Histogram of D1 1 Year after PM when the pre-existing condition is less
than the guideline value

 

Std. Dev I 10.08
men =15.0
NI188.00

 

 

 

Frequency

0'0 6:0 ’00 {5:0 {to 96:0 “b0 “3:0 ”0 a0 ‘fio $0

1 Year after PM

Figure C.50 - Histogram of D1 1 Year after PM when the pre-existing condition is greater
than the guideline value

177

 

190

1401

120'

 

   

Std. Dev I134
k Wat-8.1
NI761.00

0 7 5 I I
'0 '0 ‘0 ‘30 0.0 ‘bo fig %0 fig *0 b0

  

L

 

 

  

Frequency
8

2 Years after PM

Figure C.51 - Histogram of D1 2 Years afier PM when the pre-existing condition is less
than the guideline value

 

 

 

 

2.5 7.5 12.5 17.5 22.5 27.5 32.5
5.0 10.0 15.0 20.0 25.0 30.0 35.0

2 Years after PM

Figure C.52 - Histogram of D1 2 Years afier PM when the pre-existing condition is
greater than the guideline value

178

 

 

 

 

6‘

5 sad. Dev - 14.00

g Mun - 15.7

L: Nneaam

0‘0 ’0-0 ‘bo ‘bo 7% $.30 620 $10 $30 Q’o ’ng ”0.0
3 Years alter PM
Figure C.53 - Histogram of DI 3 Years after PM when the pre-existing condition is less
than the guideline value

 

 

 

 

6'
§ 810. Dav - 9.52
8- M I 20.6
I: n- 159.00

0-0 ¢o "0.0 <50 720 0:0 7’90 4:0 ‘00 11:0 120 110

3 Years after PM
Figure C.54 - Histogram of D1 3 Years after PM when the pre-existing condition is
greater than the guideline value

179

 

 

 

0

 

Frequency

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 ”.0 ”.0
5.0 15.0 25.0 35.0 45.0 55.0 65.0 75.0 05.0

4 Year: after PM

Figure C.55 - Histogram of D1 4 Years afier PM when the
than the guideline value

SM. Dev I 17.69
than I 18.9
NI 796.00

pre-existing condition is less

 

10

 

Frequency

4 Years after PM

 

Std. Dev I 10.84
men I 15.4
NI35.00

 

Figure C.56 - Histogram of D1 4 Years after PM when the pre-existing condition is
greater than the guideline value

180

 

140

 

Frequency

 

 

%ae%9%%e%%@%gggg

0 ‘0 '0 '0 '0 '0 '0 '0 '0 .0 .0 .0 .0 .0 .0

5 Years after PM

9111. Dev - 19.37
Mm - 23.7
‘N-751.00

Figure 057 - Histogram of D1 5 Years afier PM when the pre-existing condition is less

than the guideline value

 

 

 

    

 

 

40

301

201

101
6'
§ sum-19.15
g khan-26.0
E 0 ~ - ' - ' _ NI155.00

5YeareaflerPM

Figure C58 - Histogram of D1 5 Years after PM when the pre-existing condition is
greater than the guideline value

181

 

 

Sid. W I 12.01
M I 10.2
NI 335.00

0¢g’0%%*’0%%%%%%%

’0 .0 .0 .0 .0 .0 .0 .0 .0 .0 .0 .0

 

   

Frequency

 

6 Year: after PM

Figure C.59 - Histogram of D1 6 Years afier PM when the pre-existing condition is less
than the guideline value

 

10

4d

‘ SB. Dev I 18.72
than I 29.8

NI 10.00

 

 

  

Frequency

0

10.0 20.0 30.0 40.0 50.0 90.0 70.0 ”.0

6 Years after PM

Figure C.6O - Histogram of D1 6 Years after PM when the pre-existing condition is
greater than the guideline value

182

 

201

 

  

 
  

‘1

 

h

   

 

810. my I 31.46
men I 49.9
N I 199.00

q I I I I I <3
0 Q70 70.0 abo 0% (5.0 ‘20 '20 $20 fioqbogbo 'Qo

7 Years after PM

Figure C.61 - Histogram of D1 7 Years afier PM when the pre-existing condition is less

than the guideline value

 

10

 

Frequency

 

10.0 20.0 30.0 40.0 50.0 00.0 70.0
5.0 15.0 25.0 35.0 45.0 55.0 05.0 75.0

7 Years after PM

Figure C.62 - Histogram of D1 7 Years after PM when the pre-existing condition is
greater than the guideline value

183

 

Std. Dev I 13.98
than I41.8
NI78.00

 

APPENDIX D

TWO SAMPLE T-TEST RESULTS FROM MINI-TAB USED TO EVALUATE
THE DISTRESS INDEX (DI) PREVENTIVE MAINTENANCE (PM) GUIDELINE
VALUES FOR THE FOLLOWING PM TREATMENTS

Non-Structural Bituminous Overlay
Surface Milling with a Non-Structural Bituminous Overlay
Single Chip Seal
Multiple Course Micro-Surfacing
Bituminous Crack Seal

184

Non-Structural Bituminous Overlay
Two Sample T-Test and Confidence Interval

Two sample T for 1 Year_B vs 1 Year_A
N Mean StDev SE Mean

1Year_B 540 2.89 4.87 0.21

1 Year_A 119 5.91 4.63 0.42

95% CI for mu 1 Year_B - mu 1 Year_A: ( -3.95, -2.08)
T-Test mu 1 Year_B = mu 1 Year_A (vs not =): T= -6.38 P=0.0000 DF= 180

Two Sample T-Test and Confidence Interval

Two sample T for 2 Year_B vs 2 Year_A
N Mean StDev SE Mean

2 Year_B 545 10.8 10.5 0.45

2 Year_A 424 9.2 11.2 0.54

95% CI for mu 2 Year_B - mu 2 Year_A: ( 0.23, 3.00)
T-Test mu 2 Year_B = mu 2 Year_A (vs not =): T= 2.29 P=0.022 DF= 882

Two Sample T-Test and Confidence Interval
Two sample T for 3 Year_B vs 3 Year_A

N Mean StDev SE Mean
3 Year_B 494 4.79 3.72 0.17
3 Year_A 68 6.38 5.86 0.71

95% CI for mu 3 Year_B - mu 3 Year_A: (-3.04, -O.13)
T-Test mu 3 Year_B = mu 3 Year_A (vs not =): T= -2.18 P=0.033 DF= 74

Two Sample T-Test and Confidence Interval
Two sample T for 4 Year_B vs 4 Year_A

N Mean StDev SE Mean
4 Year_B 526 19.0 21.2 0.92
4 Year_A 276 21.7 19.3 1.2

95% CI for mu 4 Year_B - mu 4 Year_A: (-5.64, 0.2)
T-Test mu 4 Year_B = mu 4 Year_A (vs not =): T= -1.83 P=0.067 DF= 604

Two Sample T-Test and Confidence Interval

Two sample T for 5 Year_B vs 5 Year_A
N Mean StDev SE Mean

185

5Year_B 242 8.66 6.98 0.45
5Year_A 70 20.6 16.4 2.0

95% CI for mu 5 Year_B - mu 5 Year_A: ( -15.93, -7.9)
T-Test mu 5 Year_B = mu 5 Year_A (vs not =): T= -5.95 P=0.0000 DF= 76

Two Sample T-Test and Confidence Interval

Two sample T for 6 Year_B vs 6 Year_A
N Mean StDev SE Mean

6 Year_B 111 37.9 23.2 2.2

6Year_A 11 131 111 33

95% CI for mu 6 Year_B - mu 6 Year_A: (-168.0, -19)
T-Test mu 6 Year_B = mu 6 Year_A (vs not =): T= -2.80 P=0.019 DF= 10

Two Sample T—Test and Confidence Interval
Two sample T for 8 Year_B vs 8 Year_A

N Mean StDev SE Mean
8 Year_B 95 86.4 97.7 10
8 Year_A 8 201 290 103

95% CI for mu 8 Year_B - mu 8 Year_A: ( -359, 129)
T-Test mu 8 Year_B = mu 8 Year_A (vs not =): T= -1.11 P=O.30 DF= 7

186

Surface Milling with a Non-Structural Bituminous Overlay
Two Sample T-Test and Confidence Interval

Two sample T for 1 Year_B vs 1 Year_A
N Mean StDev SE Mean

1Year_B 47 6.13 5.95 0.87

1Year_A 59 5.59 4.48 0.58

95% CI for mu 1 Year_B - mu 1 Year_A: ( -1.54, 2.62)
T-Test mu 1 Year_B = mu 1 Year_A (vs not =): T= 0.52 P=0.60 DF= 83

Two Sample T-Test and Confidence Interval

Two sample T for 2 Year_B vs 2 Year_A
N Mean StDev SE Mean

2 Year_B 490 8.10 7.77 0.35

2 Year_A 71 5.50 6.26 0.74

95% CI for mu 2 Year_B - mu 2 Year_A: ( 0.97, 4.23)
T-Test mu 2 Year_B = mu 2 Year_A (vs not =): T= 3.16 P=0.0020 DF= 103

Two Sample T-Test and Confidence Interval
Two sample T for 3 Year_B vs 3 Year_A

N Mean StDev SE Mean
3 Year_B 56 6.27 5.80 0.78
3 Year_A 26 5.81 4.00 0.78

95% CI for mu 3 Year_B - mu 3 Year_A: ( -1.75, 2.66)
T-Test mu 3 Year_B = mu 3 Year_A (vs not =): T= 0.41 P=0.68 DF= 68

Two Sample T-Test and Confidence Interval
Two sample T for 4 Year_B vs 4 Year_A

N Mean StDev SE Mean
4 Year_B 355 15.82 9.73 0.52
4 Year_A 22 16.5 10.9 2.3

95% CI for mu 4 Year_B - mu 4 Year_A: ( -5.59, 4.2)
T-Test mu 4 Year_B = mu 4 Year_A (vs not =): T= -0.28 P=0.78 DF= 23

Two Sample T-Test and Confidence Interval

Two sample T for 5 Year_B vs 5 Year_A
N Mean StDev SE Mean

187

SYear_B 5 48.3 36.8 16
5Year_A11 30.4 36.1 11

95% CI for mu 5 Year_B - mu 5 Year_A: (-29, 65)
T-Test mu 5 Year_B = mu 5 Year_A (vs not =): T= 0.91 P=0.39 DF= 7

Two Sample T-Test and Confidence Interval
Two sample T for 7 Year_B vs 7 Year_A
N Mean StDev SE Mean
7 Year_B 2 21.8 11.4 8.1
7 Year_A 7 8.62 4.19 1.6

95% CI for mu 7 Year_B - mu 7 Year_A: ( -91.2, 117.6)
T-Test mu 7 Year_B = mu 7 Year_A (vs not =): T= 1.61 P=0.35 DF= l

188

Single Chip Seal
Two Sample T-Test and Confidence Interval

Two sample T for 1 Year_B vs 1 Year_A
N Mean StDev SE Mean

1 Year_B 599 19.1 21.9 0.89

1 Year_A 352 28.4 32.8 1.7

95% CI for mu 1 Year_B - mu 1 Year_A: ( -13.18, -5.5)
T-Test mu 1 Year_B = mu 1 Year_A (vs not =): T= -4.74 P=0.0000 DF= 536

Two Sample T-Test and Confidence Interval

Two sample T for 2 Year_B vs 2 Year_A
N Mean StDev SE Mean

2 Year_B 432 8.72 9.16 0.44

2 Year_A 618 16.7 23.3 0.94

95% CI for mu 2 Year_B - mu 2 Year_A: ( -10.03, -5.97)
T-Test mu 2 Year_B = mu 2 Year_A (vs not =): T= -7.74 P=0.0000 DF= 860

Two Sample T—Test and Confidence Interval
Two sample T for 3 Year_B vs 3 Year_A

N Mean StDev SE Mean
3 Year_B 512 14.8 15.7 0.69
3 Year_A 368 39.2 33.9 1.8

95% CI for mu 3 Year_B - mu 3 Year_A: ( -28.18, -20.7)
T-Test mu 3 Year_B = mu 3 Year_A (vs not =): T= -12.88 P=0.0000 DF= 480

Two Sample T-Test and Confidence Interval
Two sample T for 4 Year_B vs 4 Year_A

N Mean StDev SE Mean
4 Year_B 94 7.06 6.49 0.67
4 Year_A 151 15.25 9.78 0.80

95% CI for mu 4 Year_B - mu 4 Year_A: (~10.24, -6.14)
T-Test mu 4 Year_B = mu 4 Year_A (vs not =): T= -7.88 P=0.0000 DF= 241

Two Sample T-Test and Confidence Interval

Two sample T for 5 Year_B vs 5 Year_A
N Mean StDev SE Mean

189

5Year_B 558 19.8 42.3 1.8
5Year_A196 56.4 62.4 4.5

95% CI for mu 5 Year_B - mu 5 Year_A: ( -46.1, -27.1)
T-Test mu 5 Year_B = mu 5 Year_A (vs not =): T: -7.62 P=0.0000 DF= 260

Two Sample T-Test and Confidence Interval

Two sample T for 6 Year_B vs 6 Year_A
N Mean StDev SE Mean

6 Year_B 303 24.9 24.9 1.4

6 Year_A 45 78.7 86.5 13

95% CI for mu 6 Year_B - mu 6 Year_A: ( -80.0, -28)
T-Test mu 6 Year_B = mu 6 Year_A (vs not =): T= -4.15 P=0.0001 DF= 45

Two Sample T-Test and Confidence Interval
Two sample T for 7 Year_B vs 7 Year_A

N Mean StDev SE Mean
7 Year_B 69 5.09 8.86 1.1
7 Year_A 22 4.66 4.81 1.0

95% CI for mu 7 Year_B - mu 7 Year_A: ( -2.5, 3.4)
T-Test mu 7 Year_B = mu 7 Year_A (vs not =): T= 0.30 P=0.77 1DF = 66

190

Multiple Course Micro-Surfacing
Two Sample T-Test and Confidence Interval

Two sample T for l Year_B vs 1 Year_A
N Mean StDev SE Mean

1Year_B 464 12.32 8.01 0.37

1Year_A 26 37.4 13.6 2.7

95% CI for mu 1 Year_B - mu 1 Year_A: ( -30.60, -19.5)
T-Test mu 1 Year_B = mu 1 Year_A (vs not =): T= -9.30 P=0.0000 DF= 25

Two Sample T-Test and Confidence Interval

Two sample T for 2 Year_B vs 2 Year_A
N Mean StDev SE Mean

2 Year_B 621 10.30 8.93 0.36

2 Year_A 34 27.4 17.1 2.9

95% CI for mu 2 Year_B - mu 2 Year_A: ( -23.06, -11.0)
T-Test mu 2 Year_B = mu 2 Year_A (vs not =): T= -5.77 P=0.0000 DF= 33

Two Sample T-Test and Confidence Interval
Two sample T for 3 Year_B vs 3 Year_A

N Mean StDev SE Mean
3 Year_B 690 15.0 16.0 0.61
3 Year_A 45 73.4 36.5 5.4

95% CI for mu 3 Year_B - mu 3 Year_A: ( -69.34, -47.3)
T-Test mu 3 Year_B = mu 3 Year_A (vs not =): T: -10.64 P=0.0000 DF= 45

Two Sample T-Test and Confidence Interval
Two sample T for 4 Year_B vs 4 Year_A

N Mean StDev SE Mean
4 Year_B 684 14.5 15.1 0.58
4 Year_A 141 58.0 21.0 1.8

95% CI for mu 4 Year_B - mu 4 Year_A: ( -47.21, -39.9)
T-Test mu 4 Year_B = mu 4 Year_A (vs not =): T= -23.37 P=0.0000 DF= 171

Two Sample T-Test and Confidence Interval

Two sample T for 5 Year_B vs 5 Year_A
N Mean StDev SE Mean

191

5Year_B 584 16.5 21.4 0.89
5Year_A 18 57.0 33.2 7.8

95% CI for mu 5 Year_B - mu 5 Year_A: ( -57.13, -23.9)
T-Test mu 5 Year_B = mu 5 Year_A (vs not =): T= -5.l3 P=0.0001 DF= 17

Two Sample T-Test and Confidence Interval
Two sample T for 6 Year_B vs 6 Year_A

N Mean StDev SE Mean
6 Year_B 457 31.4 30.1 1.4
6 Year_A 37 107 112 18

95% CI for mu 6 Year_B - mu 6 Year_A: ( -113.4, -38)
T-Test mu 6 Year_B = mu 6 Year_A (vs not =): T: -4.10 P=0.0002 DF= 36

192

Bituminous Crack Seal
Two Sample T-Test and Confidence Interval

Two sample T for l Year_B vs 1 Year_A
N Mean StDev SE Mean

1 Year_B 990 11.24 9.37 0.30

1 Year_A 159 21.8 14.8 1.2

95% CI for mu 1 Year_B - mu 1 Year_A: ( -13.00, -8.2)
T-Test mu 1 Year_B = mu 1 Year_A (vs not =): T= -8.73 P=0.0000 DF= 178

Two Sample T-Test and Confidence Interval

Two sample T for 2 Year_B vs 2 Year_A

N Mean StDev SE Mean
2 Year_B 2010 13.3 11.8 0.26
2 Year_A 346 27.9 21.2 1.1

95% CI for mu 2 Year_B - mu 2 Year_A: ( -16.85, -12.2)
T-Test mu 2 Year_B = mu 2 Year_A (vs not =): T= -12.42 P=0.0000 DF= 382

Two Sample T-Test and Confidence Interval
Two sample T for 3 Year_B vs 3 Year_A

N Mean StDev SE Mean
3 Year_B 610 18.1 13.3 0.54
3 Year_A 78 31.4 19.2 2.2

95% CI for mu 3 Year_B - mu 3 Year_A: ( -17.76, -8.9)
T-Test mu 3 Year_B = mu 3 Year_A (vs not =): T= -5.95 P=0.0000 DF= 86

Two Sample T-Test and Confidence Interval
Two sample T for 4 Year_B vs 4 Year_A

N Mean StDev SE Mean
4 Year_B 1531 16.4 16.3 0.42
4 Year_A 308 39.3 40.7 2.3

95% CI for mu 4 Year_B - mu 4 Year_A: ( -27.52, -18.3)
T-Test mu 4 Year_B = mu 4 Year_A (vs not =): T= -9.73 P=0.0000 DF= 327

Two Sample T-Test and Confidence Interval

Two sample T for 5 Year_B vs 5 Year_A
N Mean StDev SE Mean

193

5Year_B 604 26.5 17.1 0.70
5Year_A 80 37.9 22.7 2.5

95% CI for mu 5 Year_B - mu 5 Year_A: ( -l6.62, -6.2)
T-Test mu 5 Year_B = mu 5 Year_A (vs not =): T= -4.33 P=0.0000 DF= 91

Two Sample T-Test and Confidence Interval
Two sample T for 6 Year_B vs 6 Year_A

N Mean StDev SE Mean
6 Year_B 874 34.6 39.6 1.3
6 Year_A 29 86.6 96.2 18

95% CI for mu 6 Year_B - mu 6 Year_A: ( -88.7, -15)
T-Test mu 6 Year_B = mu 6 Year_A (vs not =): T= -2.90 P=0.0072 DF= 28

Two Sample T-Test and Confidence Interval
Two sample T for 7 Year_B vs 7 Year_A

N Mean StDev SE Mean
7 Year_B 319 58.3 41.3 2.3
7 Year_A 39 60.3 27.2 4.3

95% CI for mu 7 Year_B - mu 7 Year_A: ( -11.9, 7.8)

194

APPENDIX E

TWO SAMPLE T-TEST RESULTS FROM MINI-TAB USED TO EVALUATE
THE RIDE QUALITY INDEX (RQD PREVENTIVE MAINTENANCE (PM)
GUIDELINE VALUES FOR THE FOLLOWING PM TREATMENTS BASED ON
THE DISTRESS INDEX

Non-Structural Bituminous Overlay
Surface Milling with a Non-Structural Bituminous Overlay
Single Chip Seal
Multiple Course Micro-Surfacing
Bituminous Crack Seal

195

Non-Structural Bituminous Overlay
Two Sample T-Test and Confidence Interval

Two sample T for l Year_B vs 1 Year_A
N Mean StDev SE Mean

1Year_B 232 4.92 5.03 0.33

1 Year_A 13 5.08 7.52 2.1

95% CI for mu 1 Year_B - mu 1 Year_A: ( -4.76, 4.4)
T-Test mu 1 Year_B = mu 1 Year_A (vs not =): T= -0.08 P=0.94 DF= 12

Two Sample T-Test and Confidence Interval

Two sample T for 2 Year_B vs 2 Year_A
N Mean StDev SE Mean

2 Year_B 470 6.80 6.04 0.28

2 Year_A 46 8.16 5.34 0.79

95% CI for mu 2 Year_B - mu 2 Year_A: (-3.03, 0.32)
T-Test mu 2 Year_B = mu 2 Year_A (vs not =): T= -1.62 P=0.11 DF= 56

Two Sample T-Test and Confidence Interval
Two sample T for 3 Year_B vs 3 Year_A

N Mean StDev SE Mean
3 Year_B 233 8.79 8.67 0.57
3 Year_A 11 8.17 9.99 3.0

95% CI for mu 3 Year_B - mu 3 Year_A: (-6.20, 7.5)
T-Test mu 3 Year_B = mu 3 Year_A (vs not =): T= 0.20 P=0.84 DF= 10

Two Sample T—Test and Confidence Interval
Two sample T for 4 Year_B vs 4 Year_A

N Mean StDev SE Mean
4 Year_B 244 6.33 7.11 0.46
4 Year_A 41 25.2 19.4 3.0

95% CI for mu 4 Year_B - mu 4 Year_A: ( -25.07, -12.7)
T-Test mu 4 Year_B = mu 4 Year_A (vs not =): T= -6.15 P=0.0000 DF= 41

Two Sample T-Test and Confidence Interval

Two sample T for 5 Year_B vs 5 Year_A
N Mean StDev SE Mean

196

5Year_B186 8.14 6.71 0.49
5Year_A 11 13.0 10.7 3.2

95% CI for mu 5 Year_B - mu 5 Year_A: ( -12.14, 2.4)
T-Test mu 5 Year_B = mu 5 Year_A (vs not =): T= -1.49 P=0.l7 DF= 10

Two Sample T-Test and Confidence Interval
Two sample T for 6 Year_B vs 6 Year_A

N Mean StDev SE Mean
6 Year_B 213 14.92 7.81 0.53
6 Year_A 7 13.95 9.13 3.4

95% CI for mu 6 Year_B - mu 6 Year_A: ( -7.57, 9.5)
T-Test mu 6 Year_B = mu 6 Year_A (vs not =): T= 0.28 P=0.79 DF= 6

197

Surfacing Milling with a Non-Structural Bituminous Overlay
Two Sample T-Test and Confidence Interval

Two sample T for l Year_B vs 1 Year_A
N Mean StDev SE Mean

1Year_B 46 12.21 3.50 0.52

1 Year_A 26 12.47 3.25 0.64

95% CI for mu 1 Year_B - mu 1 Year_A: ( -1.90, 1.38)
T-Test mu 1 Year_B = mu 1 Year_A (vs not =): T= -0.31 P=0.75 DF= 55

Two Sample T-Test and Confidence Interval

Two sample T for 2 Year_B vs 2 Year_A
N Mean StDev SE Mean

2 Year_B 115 8.77 8.79 0.82

2 Year_A 34 9.68 9.43 1.6

95% CI for mu 2 Year_B - mu 2 Year_A: ( 4.56, 2.7)
T-Test mu 2 Year_B = mu 2 Year_A (vs not =): T= -0.51 P=0.6l DF= 51

Two Sample T-Test and Confidence Interval
Two sample T for 3 Year_B vs 3 Year_A

N Mean StDev SE Mean
3 Year_B 79 13.06 7.33 0.82
3 Year_A 39 14.48 9.39 1.5

95% CI for mu 3 Year_B - mu 3 Year_A: ( -4.85, 2.0)
T-Test mu 3 Year_B = mu 3 Year_A (vs not =): T= -0.83 P=0.41 DF= 61

Two Sample T-Test and Confidence Interval
Two sample T for 4 Year_B vs 4 Year_A

N Mean StDev SE Mean
4 Year_B 73 12.49 7.81 0.91
4 Year_A 27 10.11 9.55 1.8

95% CI for mu 4 Year_B - mu 4 Year_A: ( -1.77, 6.5)
T-Test mu 4 Year_B = mu 4 Year_A (vs not =): T= 1.16 P=0.25 DF= 39

Two Sample T-Test and Confidence Interval

Two sample T for 5 Year_B vs 5 Year_A
N Mean StDev SE Mean

198

5Year_B 46 29.89 9.95 1.5
5Year_A 32 42.9 22.3 3.9

95% CI for mu 5 Year_B - mu 5 Year_A: ( -21.6, -4.5)
T-Test mu 5 Year_B = mu 5 Year_A (vs not =): T= -3.10 P=0.0036 DF= 39

Two Sample T-Test and Confidence Interval
Two sample T for 6 Year_B vs 6 Year_A
N Mean StDev SE Mean
6 Year_B 8 18.89 5.43 1.9
6 Year_A 7 20.00 4.28 1.6

95% CI for mu 6 Year_B - mu 6 Year_A: ( -6.6, 4.4)
T-Test mu 6 Year_B = mu 6 Year_A (vs not =): T= -0.44 P=0.66 DF= 12

199

Single Chip Seal
Two Sample T-Test and Confidence Interval

Two sample T for 1 Year_B vs 1 Year_A
N Mean StDev SE Mean

1Year_B 621 19.9 20.7 0.83

1 Year_A 228 24.2 22.4 1.5

95% CI for mu 1 Year_B - mu 1 Year_A: ( -7.60, -0.9)
T-Test mu 1 Year_B = mu 1 Year_A (vs not =): T= -2.50 P=0.013 DF= 378

Two Sample T-Test and Confidence Interval

Two sample T for 2 Year_B vs 2 Year_A
N Mean StDev SE Mean

2 Year_B 646 10.31 9.86 0.39

2 Year_A 160 11.5 12.0 0.95

95% CI for mu 2 Year_B - mu 2 Year_A: ( -3.24, 0.79)
T-Test mu 2 Year_B = mu 2 Year_A (vs not =): T= -l .20 P=0.23 DF= 215

Two Sample T-Test and Confidence Interval
Two sample T for 3 Year_B vs 3 Year_A

N Mean StDev SE Mean
3 Year_B 571 37.9 29.4 1.2
3 Year_A 108 32.7 21.3 2.1

95% CI for mu 3 Year_B - mu 3 Year_A: ( 0.5, 9.9)
T-Test mu 3 Year_B = mu 3 Year_A (vs not =): T= 2.19 P=0.030 DF= 193

Two Sample T-Test and Confidence Interval
Two sample T for 4 Year_B vs 4 Year_A

N Mean StDev SE Mean
4 Year_B 373 9.6 10.2 0.53
4 Year_A 214 17.2 13.9 0.95

95% CI for mu 4 Year_B - mu 4 Year_A: (-9.72, -5.44)
T-Test mu 4 Year_B = mu 4 Year_A (vs not =): T= -6.96 P=0.0000 DF= 345

Two Sample T-Test and Confidence Interval

Two sample T for 5 Year_B vs 5 Year_A
N Mean StDev SE Mean

200

5Year_B 600 30.4 53.8 2.2
5Year_A121 59.5 45.1 4.1

95% CI for mu 5 Year_B - mu 5 Year_A: ( -38.2, -19.9)
T-Test mu 5 Year_B = mu 5 Year_A (vs not =): T= -6.24 P=0.0000 DF= 195

Two Sample T—Test and Confidence Interval
Two sample T for 6 Year_B vs 6 Year_A

N Mean StDev SE Mean
6 Year_B 324 61.9 62.7 3.5
6 Year_A 27 41.3 69.2 13

95% CI for mu 6 Year_B - mu 6 Year_A: ( -7.5, 49)
T-Test mu 6 Year_B = mu 6 Year_A (vs not =): T= 1.50 P=0.14 DF= 29

201

Multiple Course Micro-Surfacing
Two Sample T-Test and Confidence Interval

Two sample T for 1 Year_B vs 1 Year_A
N Mean StDev SE Mean

1Year_B 413 11.97 7.91 0.39

1Year_A 58 10.8 11.0 1.4

95% CI for mu 1 Year_B - mu 1 Year_A: ( -1.81, 4.2)
T-Test mu 1 Year_B = mu 1 Year_A (vs not =): T= 0.79 P=0.43 DF= 65

Two Sample T-Test and Confidence Interval

Two sample T for 2 Year_B vs 2 Year_A
N Mean StDev SE Mean

2 Year_B 413 10.01 9.82 0.48

2 Year_A 182 30.2 22.8 1.7

95% CI for mu 2 Year_B - mu 2 Year_A: ( -23.63, -16.7)
T-Test mu 2 Year_B = mu 2 Year_A (vs not =): T= -11.48 P=0.0000 DF= 211

Two Sample T-Test and Confidence Interval
Two sample T for 3 Year_B vs 3 Year_A

N Mean StDev SE Mean
3 Year_B 405 8.47 7.63 0.38
3 Year_A 52 8.86 8.71 1.2

95% CI for mu 3 Year_B - mu 3 Year_A: ( -2.93, 2.1)
T-Test mu 3 Year_B = mu 3 Year_A (vs not =): T= -0.31 P=0.76 DF= 61

Two Sample T-Test and Confidence Interval
Two sample T for 4 Year_B vs 4 Year_A

N Mean StDev SE Mean
4 Year_B 401 10.17 9.91 0.49
4 Year_A 159 35.6 44.5 3.5

95% CI for mu 4 Year_B - mu 4 Year_A: (-32.50, -18.4)
T-Test mu 4 Year_B = mu 4 Year_A (vs not =): T= -7.14 P=0.0000 DF= 164

Two Sample T-Test and Confidence Interval

Two sample T for 5 Year_B vs 5 Year_A
N Mean StDev SE Mean

202

5Year_B 355 17.3 22.6 1.2
5Year_A 44 19.5 14.0 2.1

95% CI for mu 5 Year_B - mu 5 Year_A: ( -7.1, 2.6)
T-Test mu 5 Year_B = mu 5 Year_A (vs not =): T= -0.92 P=0.36 DF= 74

Two Sample T-Test and Confidence Interval
Two sample T for 6 Year_B vs 6 Year_A

N Mean StDev SE Mean
6 Year_B 243 31.7 35.4 2.3
6 Year_A 116 60.9 53.9 5.0

95% CI for mu 6 Year_B - mu 6 Year_A: ( —40.0, -l8.3)
T-Test mu 6 Year_B = mu 6 Year_A (vs not =): T= -5.31 P=0.0000 DF= 163

203

Bituminous Crack Seal
Two Sample T-Test and Confidence Interval

Two sample T for l Year_B vs 1 Year_A
N Mean StDev SE Mean

1Year_B 873 11.0 10.3 0.35

1 Year_A 186 15.0 10.1 0.74

95% CI for mu 1 Year_B - mu 1 Year_A: ( -5.59, -2.37)
T-Test mu 1 Year_B = mu 1 Year_A (vs not =): T= -4.87 P=0.0000 DF= 272

Two Sample T-Test and Confidence Interval

Two sample T for 2 Year_B vs 2 Year_A
N Mean StDev SE Mean

2 Year_B 761 8.09 7.34 0.27

2 Year_A 36 13.14 7.31 1.2

95% CI for mu 2 Year_B - mu 2 Year_A: ( -7.57, -2.5)
T-Test mu 2 Year_B = mu 2 Year_A (vs not =): T= -4.05 P=0.0002 DF= 38

Two Sample T-Test and Confidence Interval
Two sample T for 3 Year_B vs 3 Year_A

N Mean StDev SE Mean
3 Year_B 688 15.7 14.0 0.53
3 Year_A 159 20.58 8.52 0.68

95% CI for mu 3 Year_B - mu 3 Year_A: ( -6.61, -3.22)
T-Test mu 3 Year_B = mu 3 Year_A (vs not =): T= -5.71 P=0.0000 DF= 383

Two Sample T-Test and Confidence Interval
Two sample T for 4 Year_B vs 4 Year_A

N Mean StDev SE Mean
4 Year_B 796 18.9 17.7 0.63
4 Year_A 35 15.4 10.6 1.8

95% CI for mu 4 Year_B - mu 4 Year_A: ( -0.32, 7.4)
T-Test mu 4 Year_B = mu 4 Year_A (vs not =): T= 1.85 P=0.071 DF= 42

Two Sample T-Test and Confidence Interval

Two sample T for 5 Year_B vs 5 Year_A
N Mean StDev SE Mean

204

5Year_B 751 23.7 18.4 0.67
5Year_A155 26.0 13.2 1.1

95% CI for mu 5 Year_B - mu 5 Year_A: ( -4.84, 0.1)
T-Test mu 5 Year_B = mu 5 Year_A (vs not =): T= -1.90 P=0.059 DF= 293

Two Sample T-Test and Confidence Interval
Two sample T for 6 Year_B vs 6 Year_A

N Mean StDev SE Mean
6 Year_B 335 16.2 12.0 0.66
6 Year_A 16 29.6 16.7 4.2

95% CI for mu 6 Year_B - mu 6 Year_A: ( -22.43, -4.4)
T-Test mu 6 Year_B = mu 6 Year_A (vs not =): T= -3.17 P=0.0063 DF= 15

Two Sample T—Test and Confidence Interval
Two sample T for 7 Year_B vs 7 Year_A

N Mean StDev SE Mean
7 Year_B 189 49.9 31.5 2.3
7 Year_A 76 41.6 14.0 1.6

95% CI for mu 7 Year_B - mu 7 Year_A: (2.7, 13.8)

205

BIBLIOGRAPHY

206

10.

LIST OF REFERENCES

. Al-Mansour, Abdullah I., Kuczek, Thomas, and Sinha, Kumares C. “Effects of

Routine Maintenance on Flexible Pavement Condition”. Journal of
Transportation Engineering, Vol 120 No. l. 1994.

Batac, Gilbert and Ray, Michael. “French Strategy for Preventive Road
Maintenance: Why and How?” Transportation Research Record 1183, TRB,
National Research Council, Washington D.C., pp. 22-34.

Brown, E.R. “Preventive Maintenance of Asphalt Concrete Pavements,”
Transportation Research Record 1205, TRB, National Research Council,
Washington D.C., pp. 6-10.

Brown, Ray E., Kandhal, Prithvi S., Kennedy, Thomas W., Lee, Dah-Yinn, and
Roberts, Freddy L. Hot Mix Asphalt Materials, Mixture Design an_c_1
Construction. Maryland: NAPA Education Foundation, 1996.

. Butt, Abbas A., Carpenter, Samuel H., Feighan, Kieran J. and Shahin, Mohamed

Y. “Pavement Performance Prediction Model Using Markov Process”,
Transportation Research Record 1123, TRB, National Research Council,
Washington D.C., pp. 12-19.

Cation, Kathryn A, Lytton, Robert L., Scullion, Thomas and Shahin, Mohamed Y.
“Development of a Preventive Maintenance Algorithm for Use in Pavement
Management Systems,” Transportation Research Record 1123, TRB, National
Research Council, Washington D.C., pp. 1-11.

Chan, W.T., Fwa, T.F., Hoque, K.Z. and Nixon, E. “Benefits of Active Planning
for Preventive Pavement Maintenance at Network Level”, T RB I 999 Annual
Meeting CD-ROM, TRB, National Research Council, Washington DC.

Chong, GO. and Phang, W.A. “Improved Preventive Maintenance: Sealing
Cracks in Flexible Pavements in Cold Regions,” Transportation Research Record
1205, TRB, National Research Council, Washington D.C., pp. 12-19.

Daleiden, Jerome F., Eltahan, Ahmed A., and Simpson, Amy L. “Effectiveness of
Maintenance Treatments of Flexible Pavements,” Transportation Research
Record 1680, TRB, National Research Council, Washington D.C., pp. 18-25.

Daleiden, Jerry, Hicks, Gary R., and Moulthrop, James S. “Selecting a Preventive
Maintenance Treatment for Flexible Pavements,” T RB 1999 Annual Meeting CD-
ROM, TRB, National Research Council, Washington DC.

207

 

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22

Hicks, Gary R., Dunn, Kimberly, and Moulthrop, James S. “Framework for
Selecting Effective Preventive Maintenance Treatments for Flexible Pavements,”
Transportation Research Record 1597, TRB, National Research Council,
Washington D.C., pp. 1-10.

National Highway Institute. Techniques for Pavement Rehabilitation Reference
Manual 6‘“ Ed. Arlington, VA: National Highway Institute, 1998.

Galehouse, Larry. “Innovative Concepts for Preventive Maintenance,”
Transportation Research Record 1627, TRB, National Research Council,
Washington D.C., pp. 1-6.

Geoffroy, D.N. Evaluation and Benefits of Preventive Maintenance Strategies.
NCHRP Synthesis of Highway Practice 153.

George, K.P. and Rajagopal, A.S. “Pavement Maintenance Effectiveness,”
Transportation Research Record 1276, TRB, National Research Council,
Washington D.C., pp. 62-68.

George, K.P., Shekharan, Raja A. and Vepa, T.S. “Prediction of Pavement
Remaining Life,” Transportation Research Record 1524, TRB, National Research
Council, Washington D.C., pp. 137-144.

Haas, Ralph, Li, Ningyuan, and Xie, Wei-Chau. “Investigation of Relationship
Between Deterministic and Probabilistic Prediction Models in Pavement
Management” Transportation Research Record 1592, TRB, National Research
Council, Washington D.C., pp. 70-79.

Hand, Adam, Lani, Stephen, and Sebaaly, Peter E. “Performance Models for
Flexible Pavement Maintenance Treatments,” Transportation Research Record
1508, TRB, National Research Council, Washington D.C., pp. 9-21.

Huang, Yang H. Pavement Analysis and Desigg. New Jersey: Prentice Hall,
1993.

Hicks, Gary R. and Moulthrop, James S. “Pavement Maintenance: Preparing for
the 21St Century”. Proceedings of the Ninth Maintenance Management
Conference.

Mamlouk, Michael and Zaniewski, John. “Pavement Preventive Maintenance:
The Key to Quality Highways,” TRB 1999 Annual Meeting CD-ROM, TRB,
National Research Council, Washington DC.

. Michigan Department of Transportation (MDOT). Capital Preventive

Maintenance Program Guidelines. Lansing, MI, March, 1998.

208

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

Metcalf, John B and Romanoschi, Stefan A. “Evaluation of Probability
Distribution Function for the Life of Pavement Structures” Transportation
Research Record 1730, TRB, National Research Council, Washington D.C., pp.
91-98.

Moulthrop, James S and Smith, Roger E. State of the Art and State of the
Practice in Pavement Maintenance. A3 C05: Committee on Pavement
Maintenance.

Shahin, Mohamed Y., Sharaf, Essam A., and Sinha, Kumares C. “Analysis of the
Effect of Deferring Pavement Maintenance,” Transportation Research Record
1205, TRB, National Research Council, Washington D.C., pp. 29-35.

Ayyub, Bilal M., and McCuen, Richard H. Probability, Statistica, & Reliabilitv
for Engineers. New York: CRC Press. 1997.

McCabe, George P. and Moore, David S. Introduction to the Practice of
Statistics. New York: W.H. Freeman and Company. 1999.

Baladi, Gilbert Y. The Role of Pavement Management System in Tracking the
PerforrLance of Preventive Maintenance Actions Case Studies. Michigan State
University. 2001.

Lee, Doseung. Development of Roughness Thresholds for the Preventive
Maintenance of Pavements Based on Dynamic Loading Considerations and
Damage Analysis. Masters Thesis. Michigan State University. 2001.

Reigle, Jenifer A. Development of an Integrated Project-Level Pavement
Management Model Using Risk Analysis. PhD Thesis. West Virginia University.
2000.

Hicks, Gary R, Seeds, Stephen B, and Peshkin, David G. Selecting a Preventive
flinteaance Treatment for Flexible Pavement; FHWA-IF-00-027. Foundation
for Pavement Preservation. 2000.

Smith, Roger B and Lytton, Robert L. “Evaluation of Effectiveness of Pavement
Preventive Maintenance”. Proceedings of SHRP and Traffic Safety on Two
Continents. 1989.

Virtala, P. “Modeling the Effects of Maintenance and Rehabilitation Actions on
Road Conditions at Network Level”. 4’” International Conference on Managing
Pavements. 1998.

Reddy, B.B., Kumar, Sivath N, and Veeraragavan, A. “Development of

Maintenance Intervention Criteria for Inservice Flexible Pavements”. 4’”
International Conference on Managing Pavements. 1998.

209

35.

36.

37.

38.

39.

40.

41.

42.

43

44.

45.

Smith, R. B. and Cerecina B. “Development of Maintenance Strategies for a
National Highway System”. 4'“ International Conference on Managing
Pavements. 1998.

Sharaf, Essam A. “A Procedure to Determine Critical Preventive Maintenance
Timing”. 4'” International Conference on Managing Pavements. 1998.

Mamlouk, Michael S. and Zaniewski, John P. “Optimizing Pavement
Preservation: An Urgent Demand for Every Highway Agency,” The International
Journal of Pavement Engineering. Volume 2. Number 2. 2001.

Morian, Dennis, Gibson, Scott, and Epps, Jon. “Study of Maintenance
Treatments for Asphalt Pavements”. 8’” International Conference on Asphalt
Pavements. 1997.

Li, Ningyuan, Xie, Wei-Chau, and Haas Ralph. “Reliability-based Processing of
Markov Chains for Modeling Pavement Network Deterioration,” TRB 1996
Annual Meeting CD-ROM, TRB, National Research Council, Washington DC.

Pedigo, Daryl R., Hudson, Ronald W., and Roberts, Freddy L. “Pavement
Performance Modeling for Pavement Management,” Transportation Research
Record 814, TRB, National Research Council, Washington D.C., pp. 8-21.

Gharaibeth, Nasir, LaTorre, Francesca, Darter, Michael, and Lippert, David. “
Use of Pavement Management System Data for Pavement Engineering
Applications,” Transportation Research Record 1592, TRB, National Research
Council, Washington D.C., pp. 35-45.

Sebaaly, Peter E., Hand, Adam, Epps, John and Bosch, Chuck. “Nevada’s
Approach to Pavement Management,” Transportation Research Record 1524,
TRB, National Research Council, Washington D.C., pp. 109-117.

. Colucci-Rios, Benjamin and Sinha, Kumares C. “Optimal Pavement

Management Approach Using Roughness Measurements,” Transportation
Research Record 1048, TRB, National Research Council, Washington D.C., pp.
14-22.

Shober, Steve F. and Friedrichs, David A. “Pavement Preservation Strategy,”

Transportation Research Record 1643, TRB, National Research Council,
Washington D.C., pp. 44-53.

Galehouse, Larry. A Pavement Preventive Maintenance Proggam. Moving
Roadway Maintenance into the 21St Century. APWA Satellite Videoconference.
1999.

210

46. Walls HI, J. and MR. Smith, Life-Cycle Cost mVsis in Payement Design —
Interim Technical Bulletin, Report No. FHWA-SA-98—079, Federal Highway
Administration, Washington DC. 1998.

47. Correa, Carlos E., Hall, Kathleen T., and Simpson, Amy L. “Performance of
Flexible Pavement Maintenance Treatments in the LTPP SPS-3 Experiment,”
TRB 2003 Annual Meeting CD-ROM, TRB, National Research Council,
Washington DC.

48. Chen, Dar-Hao, Lin, Deng-Pong, and Luo, Huan-Lin. “Effectiveness of
Preventive Maintenance Treatments Using SPS-3 Data,” TRB 2003 Annual
Meeting CD-ROM, TRB, National Research Council, Washington DC.

49. National Strategic Highway Research Program. LTTP Data Analysis:
Effectiveness of Maintenance and Rehabilitation Options, National Research
Council, June 2002.

50. B.T. Bellner & Associates. Effectiveness of the Capital Preventive mintenance
Proggam, November 2001.

51. Epps, J .A., Gibson, SD, and Morian, D.A. Maintaining Flexible Pavements;
The LongTerm Pavement Performance Experiment SPS-3 5 Year Data Analysis,
FHWA-RD-97-102, Federal Highway Administration, Washington DC. 1998.

52. Freeman, Thomas, Pendleton, Olga, and Smith, Roger. Pavement Maintenance
Effectiveness, SHRP-H-358, National Research Council, Washington DC. 1993.

53. Pendleton, Olga. “Reliability in Pavement Design: Issues, Concepts, and
Significance,” Transportation Research Record 1449, TRB, National Research
Council, Washington D.C., pp. 8-12.

54. AASHTO Guide for Design of Pavement Structures. AASHTO, Washington,
D.C., 1986.

55. @ Risk User ’s Manual, 1997, Palisade Corporation, Newfield, NY.

56. Kuo, Shiou-San, Mahgoub, Hesham, Hoffman, Lorrie. “Development of Flexible
Pavement Performance Prediction Model Based on Pavement Data”. T RB 2003
Annual Meeting CD-ROM, TRB, National Research Council, Washington DC.

57. Baladi, G.Y., and Snyder, M.B. Highway Pavementa, Volumes 1 and 2. The
National Highway Institute, Washington D.C., September 1989.

211