USF Libraries
USF Digital Collections

Application of an improved transition probability matrix based crack rating prediction methodology in Forida's highway n...

MISSING IMAGE

Material Information

Title:
Application of an improved transition probability matrix based crack rating prediction methodology in Forida's highway network
Physical Description:
Book
Language:
English
Creator:
Nasseri, Sahand
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Crack rating
Non-linear regression optimization
Pavement condition survey database
Delayed maintenance and rehabilitation
Project/network level decision making
Dissertations, Academic -- Civil & Environmental Engineering -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: With the growing need to maintain roadway systems for provision of safety and comfort for travelers, network level decision-making becomes more vital than ever. In order to keep pace with this fast evolving trend, highway authorities must maintain extremely effective databases to keep track of their highway maintenance needs. Florida Department of Transportation (FDOT), as a leader in transportation innovations in the U.S., maintains Pavement Condition Survey (PCS) database of cracking, rutting, and ride information that are updated annually. Crack rating is an important parameter used by FDOT for making maintenance decisions and budget appropriation. By establishing a crack rating threshold below which traveler comfort is not assured, authorities can screen the pavement sections which are in need of Maintenance and Rehabilitation (M&R). Hence, accurate and reliable prediction of crack thresholds is essential to optimize the rehabilitation budget and manpower.Transition Probability Matrices (TPM) can be utilized to accurately predict the deterioration of crack ratings leading to the threshold. Such TPMs are usually developed by historical data or expert or experienced maintenance engineers' opinion. When historical data are used to develop TPMs, deterioration trends have been used vii indiscriminately, i.e. with no discrimination made between pavements that degrade at different rates. However, a more discriminatory method is used in this thesis to develop TPMs based on classifying pavements first into two groups. They are pavements with relatively high traffic and, pavements with a history of excessive degradation due to delayed rehabilitation. The new approach uses a multiple non-linear regression process to separately optimize TPMs for the two groups selected by prior screening of the database.The developed TPMs are shown to have minimal prediction errors with respect to crack ratings in the database that were not used in the TPM formation. It is concluded that the above two groups are statistically different from each other with respect to the rate of cracking. The observed significant differences in the deterioration trends would provide a valuable tool for the authorities in making critical network-level decisions. The same methodology can be applied in other transportation agencies based on the corresponding databases.
Thesis:
Thesis (M.S.C.E.)--University of South Florida, 2008.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Sahand Nasseri.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 59 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001992330
oclc - 315968447
usfldc doi - E14-SFE0002379
usfldc handle - e14.2379
System ID:
SFS0026697:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001992330
003 fts
005 20090317101125.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 090317s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002379
035
(OCoLC)315968447
040
FHM
c FHM
049
FHMM
090
TA145 (Online)
1 100
Nasseri, Sahand.
0 245
Application of an improved transition probability matrix based crack rating prediction methodology in Forida's highway network
h [electronic resource] /
by Sahand Nasseri.
260
[Tampa, Fla] :
b University of South Florida,
2008.
500
Title from PDF of title page.
Document formatted into pages; contains 59 pages.
502
Thesis (M.S.C.E.)--University of South Florida, 2008.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
520
ABSTRACT: With the growing need to maintain roadway systems for provision of safety and comfort for travelers, network level decision-making becomes more vital than ever. In order to keep pace with this fast evolving trend, highway authorities must maintain extremely effective databases to keep track of their highway maintenance needs. Florida Department of Transportation (FDOT), as a leader in transportation innovations in the U.S., maintains Pavement Condition Survey (PCS) database of cracking, rutting, and ride information that are updated annually. Crack rating is an important parameter used by FDOT for making maintenance decisions and budget appropriation. By establishing a crack rating threshold below which traveler comfort is not assured, authorities can screen the pavement sections which are in need of Maintenance and Rehabilitation (M&R). Hence, accurate and reliable prediction of crack thresholds is essential to optimize the rehabilitation budget and manpower.Transition Probability Matrices (TPM) can be utilized to accurately predict the deterioration of crack ratings leading to the threshold. Such TPMs are usually developed by historical data or expert or experienced maintenance engineers' opinion. When historical data are used to develop TPMs, deterioration trends have been used vii indiscriminately, i.e. with no discrimination made between pavements that degrade at different rates. However, a more discriminatory method is used in this thesis to develop TPMs based on classifying pavements first into two groups. They are pavements with relatively high traffic and, pavements with a history of excessive degradation due to delayed rehabilitation. The new approach uses a multiple non-linear regression process to separately optimize TPMs for the two groups selected by prior screening of the database.The developed TPMs are shown to have minimal prediction errors with respect to crack ratings in the database that were not used in the TPM formation. It is concluded that the above two groups are statistically different from each other with respect to the rate of cracking. The observed significant differences in the deterioration trends would provide a valuable tool for the authorities in making critical network-level decisions. The same methodology can be applied in other transportation agencies based on the corresponding databases.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: Manjriker Gunaratne, Ph.D.
653
Crack rating
Non-linear regression optimization
Pavement condition survey database
Delayed maintenance and rehabilitation
Project/network level decision making
690
Dissertations, Academic
z USF
x Civil & Environmental Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2379



PAGE 1

Application of an Improved Transition Probabi lity Matrix Based Crack Rating Prediction Methodology in Floridas Highway Network by Sahand Nasseri A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Civil Engineering Department of Civil and Environmental Engineering College of Engineering University of South Florida Major Professor: Manjriker Gunaratne, Ph.D. Jian Lu, Ph.D. Kandethody Ramachandran, Ph.D. Date of Approval: February 28, 2008 Keywords: Crack Rating, Non-Linear Regression Optimization, Pavement Condition Survey Database, Delayed Maintenance a nd Rehabilitation, Project/Network Level Decision Making Copyright 2008, Sahand Nasseri

PAGE 2

Dedication To my parents, Siavosh and Mahshid Nasseri, and my brothers, Shahin and Sepehr Nasseri, who have been supp orting me throughout my education with unconditional help and love. They have been my inspiration and support. To all my friends overseas and in U.S, especially at the University of South Florida, for their support, help, and friendship. Finally, I want to dedicate this to my love of life, Shabnam.

PAGE 3

Acknowledgements I would like to express my sincere appreciations to Dr. Gunaratne, who was not only my advisor during my graduate studies but my mentor, support, and above all a friend at the University of South Flor ida. I am profoundly appreciative of his understandings and patience during the course of the research, project, and thesis writing. I am also grateful to the FDOTs fu nded project BD-544 for providing financial support. I would also like to pass on my sin cere thanks to the pr oject manager of the project Mr. Abdenour Nazef fo r his guidance and support. I would also like to extend my sincere thanks to my Masters committee members, Dr. Lu and Dr. Ramachandran, for their help. Also, I would like to extend special thanks to Dr. Yang for his help and guidance in some statistical analysis of this thesis.

PAGE 4

i Table of Contents List of Tables iii List of Figures iv Abstract vi Chapter One. Introduction 1 FDOT Pavement Condition Database (PCS) 1 Pavement Evaluation 3 Pavement Cracking 4 Crack Rating 5 Condition Prediction 7 Condition Prediction Based on Markov Models 8 Updated TPM Development Methods 10 Problem Statement 12 Proposed Research 13 Thesis Organization 13 Chapter Two. Experimental Methodology 14 Data Filtering 14 Impact Grouping 15 TPM Development 16 TPM Optimization 18 Verification of TPM 20 Chapter Three. Analysis of Results 21 Results 21 Network/Project Level Decision-Making 21 Applicability of Grouping 22 Developed TPMs 29 Applicability of Developed TPMs 32 Application of Test Results 39 Chapter Four. Conclusions and Limitations 43 Conclusions 43 Limitations 45 References 46

PAGE 5

ii Appendices 47 Appendix A: History of Florida Pavement Condition Survey (1973 ) 48

PAGE 6

iii List of Tables Table 1. Statewide comparison of groupi ng outcome 16 Table 2. Characteristics of th e sections operating in their 2nd cycle 23 Table 3. Characteristics of th e sections operating in their 3rd cycle 24 Table 4. Difference in mean CR values for the two groups operating in their 2nd cycle 25 Table 5. Difference in mean CR values for the two groups operating in their 3rd cycle 25 Table 6. Comparison of Mean Square Error (MSE) 38 Table 7. Recommended verification group si ze 39 Table 8. Prediction comparison of a stru cturally deficient section in its 2nd cycle 41 Table 9. Prediction comparison of a stru cturally deficient section in its 3rd cycle 41 Table 10. Prediction comparison of an excessively trafficked section in its 2nd cycle 41 Table 11. Prediction comparison of an excessively trafficked section in its 3rd cycle 41 Table 12. Predicted CR values for 2008 42

PAGE 7

iv List of Figures Figure 1. Extract from Floridas PCS database 3 Figure 2. Multi Purpose Survey Vehicle (MPSV) 4 Figure 3. Comparison of degradation between structural deficient and excessive traffic impact (cons truction duty cycle =2) 27 Figure 4. Comparison of degradation between structural deficient and excessive traffic impact (cons truction duty cycle =3) 27 Figure 5. Mean difference between structural deficient and traffic groups (construction duty cycle =2) 28 Figure 6. Mean difference between structural deficient and traffic groups (construction duty cycle =3) 29 Figure 7-A. Comparison with the 5% veri fication set at their 2nd duty cycle (structural integrity deficient sections) 32 Figure 7-B. Comparison with the 10% ve rification set at their 2nd duty cycle (structural integrity deficient sections) 33 Figure 7-C. Comparison with the 20% ve rification set at their 2nd duty cycle (structural integrity deficient sections) 33 Figure 8-A. Comparison with the 5% veri fication set at their 3rd duty cycle (structural integrity deficient sections) 34

PAGE 8

v Figure 8-B. Comparison with the 10% veri fication set at their 3rd duty cycle (structural integrity deficient sections) 34 Figure 8-C. Comparison with the 20% veri fication set at their 3rd duty cycle (structural integrity deficient sections) 35 Figure 9-A. Comparison with the 5% veri fication set at their 2nd duty cycle (excessive traffic sections) 35 Figure 9-B. Comparison with the 10% ve rification set at their 2nd duty cycle (excessive traffic sections) 36 Figure 9-C. Comparison with the 20% ve rification set at their 2nd duty cycle (excessive traffic sections) 36 Figure 10-A. Comparison with the 5% veri fication set at their 3rd duty cycle (excessive traffic sections) 37 Figure 10-B. Comparison with the 10% veri fication set at their 3rd duty cycle (excessive traffic sections) 37 Figure 10-C. Comparison with the 20% veri fication set at their 3rd duty cycle (excessive traffic sections) 38

PAGE 9

vi Application of an Improved Transition Probability Matrix Based Crack Rating Prediction Methodology in Floridas Highway Network Sahand Nasseri ABSTRACT With the growing need to maintain ro adway systems for provision of safety and comfort for travelers, network level decisi on-making becomes more vital than ever. In order to keep pace with this fast evolvi ng trend, highway authorities must maintain extremely effective databases to keep track of their highway maintenance needs. Florida Department of Transportation (FDOT), as a l eader in transportation innovations in the U.S., maintains Pavement Condition Survey (P CS) database of crack ing, rutting, and ride information that are updated annually. Crack rating is an important paramete r used by FDOT for making maintenance decisions and budget appropriati on. By establishing a crack rating threshold below which traveler comfort is not assured, authorities ca n screen the pavement sections which are in need of Maintenance and Reha bilitation (M&R). Hence, accur ate and reliable prediction of crack thresholds is essential to op timize the rehabilitation budget and manpower. Transition Probability Matrices (TPM) can be utilized to accurately predict the deterioration of crack ratings leading to the threshold. Such TPMs are usually developed by historical data or expert or experienced maintenance engineers opinion. When historical data are used to develop TPMs, deterioration trends have been used

PAGE 10

vii indiscriminately, i.e. with no discrimination made between pavements that degrade at different rates. However, a more discriminatory method is used in this thesis to develop TPMs based on classifying pavements first in to two groups. They are pavements with relatively high traffic and, pavements with a history of excessive degradation due to delayed rehabilitation. The new approach uses a multiple non-linear regression process to separately optimize TPMs for the two groups selected by prior screening of the database. The developed TPMs are shown to have minimal prediction errors with respect to crack ratings in the database that were not used in the TPM formation. It is concluded that the above two groups are statistically different fr om each other with resp ect to the rate of cracking. The observed significant differences in the deterioration tr ends would provide a valuable tool for the authorities in making critical network-level decisions. The same methodology can be applied in other transpor tation agencies base d on the corresponding databases.

PAGE 11

1 Chapter One Introduction FDOT Pavement Condition Database (PCS) For any highway agency to manage thei r roadway systems successfully, it is necessary to maintain a road way condition inventory with constant annual updates. To fulfill this crucial need, Florida Department of Transportation (FDOT) has produced a Pavement Condition Survey (PCS) database. Th e PCS database was standardized in 1986 to incorporate pavement condition data from all 7 districts in Florida. Before 1986, separate districts would pe rform their individual survey by hiring either their own personnel or contractors. In the first 3 years of data co llection after st andardization, BM&R was the sole contractor performing the pavement condition survey. In 1989, the data collection was assigned to the State Ma terials Office personnel in Gainesville, FL (Appendix A). Since the introduction of the PCS database, FDOT has expended a major effort in the maintenance and improvement of this database. Since the initiation of the database, over 3000 rated m iles (from 15,566 to 18,693) and 2500 sections (from 5,812 to 8,469) have been added to the database. Currently, all the maintena nce and rehabilitation work associated with the FDOT pavement management systems is based on the PCS

PAGE 12

2 database. FDOT has achieved an elite stat us and recognition in the nation for its comprehensive database. At present, the database contains about 9000 sections from all 7 districts of Florida. For each section, an identification num ber is used to distinguish that section. In the PCS database, for each section of the ro adway, there are some fixed characteristics such as roadway ID, roadway direction (left or right), county and di strict allocation, and US or statewide roadway ID number (i .e. SR45, US41). There are also some characteristics that would change if any re habilitation and maintenance is performed on that section, such as the begin mile post and the end mile post, surface asphalt type, asphalt thickness, number of duty cycles, a nd total lane mileage. Finally, there are characteristics that change annually such as age, Equivalent Singl e Axle Load (ESAL), and average daily traffic. The other parts of the database are fo r the input of condition ratings based on annual survey of the roadway. The updated c ondition data include the Cracking Rating (CRK) which is of interest in this paper, Rutting Index (RUT), Ride Quality (RIDE), and Pavement Condition Rating (PCR). An extract from the Florida PCS database is shown in Figure 1. Each section is then divided into subsections based on characteristics of the section so that each sub-section becomes more or less homogenous with respect to roadway geometry, traffic and c ondition features. T ypical characteristics that are included in the database are: geographic location, pa vement type (flexible, rigid), pavement surface type (open graded, dense graded), tra ffic level (A, B, C, D, and E), construction cycle, and extent of deterior ation. Hence, it is obvious that every time any rehabilitation

PAGE 13

or maintenance is performed on a sub-secti on, a new sub-section(s) would emerge and the database is updated since the characteristic s of the rehabilitated sub-section changes invariably. This implies that as time goes on, there would be more sections added to the database while the lengths of sub-sec tions would become smaller. Figure 1. Extract from Floridas PCS database Pavement Evaluation The data for the PCS database is gathered using FDOTs customized vehicle called the Multi-Purpose Survey Vehicle (M PSV) shown in Figure 2. MPSV is equipped with sophisticated on-board instrumentation and associated computer systems. All the relevant data that is coll ected by the MPSV is entered in PCS on an annual basis. 3

PAGE 14

Figure 2. Multi Purpose Survey Vehicle (MPSV) Pavement Cracking Of the different types of distresses, rutting and cracki ng are the two major distress types that are dominant in Floridas flexible pavements. Cracking is also a dominant type of distress in Floridas relatively small pe rcentage of rigid pavements. A crack is a discontinuity in the pavement surface with minimum dimensions of 1 mm (1/25 in) width and 25 mm (1 in) length (AASHT O-PP44-01). There are differe nt types of cracks which may include longitudinal cracks, transverse cracks, block cracks, edge cracks, and alligator cracks for flexible pavements and longitudinal cracks, tr ansverse cracks, and corner cracks for rigid pavement. In general, cracks are divided into 3 levels of severity and intensity (AASHTO-PP44-01). Severity Level 1: Cracks 3 mm (1/8 in) Severity Level 2: Cracks with dimension > 3 mm (1/8 in) and 6 mm (1/4in) Severity Level 3: Cracks with dimensions > 6 mm (1/4 in) 4

PAGE 15

5 In asphalt pavements, cracks develop and propagate with time due to many causes such as age-induced fatigue that results in reduced tensile strength required to overcome wheel induced pavement flexural stresses, a condition which eventually leads to failure under repeated loading; age-i nduced hardening of the binde r causing inadequate tensile strength to meet the stresse s induced by daily temperature cycling; excessive tensile stresses induced by the swelling/shrinkage of roadbed (subgrade) soils when pavements are constructed in expansive soils; imp roper lane-joint and lane-shoulder joint construction causing edge and longitudina l cracks; and low temperature induced hardening of the binder which results in inad equate tensile strength to overcome even normal vehicle-induced strains (low temper ature cracking in asphalt). Of the above, obviously only the first four types are relevant to asphalt pavements in Florida due to its temperate climate. On the other hand, crack s in concrete paveme nts of Florida occur primarily due to temperature induced curling stresses (Kum ara et al, 2003). The major focus of this thesis is on load induced damage and the delayed maintenance and rehabilitation damage caused by poor roadbed (subgrade) of Florida s flexible pavement network. Crack Rating Crack Rating (CR) is a unique distress index of each section which can be used in network level decision making and budget appropr iation since it is a appropriate measure of roadway safety and comfort. A shortcoming of this rating is the subjectivity involved in it. Although the raters are trained for ra ting consistency, human errors are inevitable and are also evident in the database. To address this issue, many softwares and

PAGE 16

6 instrumentations have been developed for auto mation of the crack rating. At present, this is a newly focused study area in pavement management. When the automation is widely established, the practicality a nd applicability of the methodol ogy advanced in this thesis will be more evident since the CR ratings woul d follow an expected pattern as compared to the current random and less predictable pattern. CR is a manually assigned rating to a pave ment section in the range of 0-10 with 10 indicating an excellent pavement condition with respect to cracking, while 0 indicates a heavily deteriorated pavement. CR is assi gned based on a windshield survey which is performed by a trained rater as the Multi Pu rpose Surveying Vehicle (MPSV) traverses a particular section. Then, the extent of each type of crack seen on the road is recorded in the relevant charts. Based on the severity and de nsity of the dominant crack type in inside and outside wheel paths of each section, a de duct value is extracted from the FDOTs Flexible Pavement Manual Survey Handbook (FDOT, 2003) and CR is calculated by subtracting the deduct value from a perf ect 10 CR rating as shown in Equation 1. CR = 10 (CO + CW) (1) Where CO = amount of crack outside wheel path CW = amount of crack inside wheel path The CR rating is then recorded in the a ppropriate column of the PCS database each year. It should be noted that for newly rehabilitated sections, CR would be 10. Therefore, by locating the sudden rise of CR from a low CR value to 10, the starting year of the new duty cycle of that section can be determined. This concept is widely used in

PAGE 17

the analysis of the database especially in de termining and sorting th e individual cycles of a section. Condition Prediction Predicting the future condition of a paveme nt provides pavement engineers with a valuable tool to prioritize the pavement se ctions for M&R activitie s with better accuracy and efficiency. Therefore, reliable performance prediction models are becoming a necessity in todays pavement manageme nt systems (Gendreau et al, 1994). Some researchers have developed analyti cal expressions to predict the future condition of a pavement (Kong et al, 2002). Howe ver, such equations are only applicable in specific locations because there is a multit ude of variables involved with cracking such that one expression cannot incorporate such a vast number of variable s and be universally representative. For instance, Equation 2 has been devel oped for Brevard County of Florida (Kong et al, 2002). ) 1000 (08561.01979.0 a s cym ycc (2) Where ycc = Last year crack rating, ym1c = Year before last year crack rating, s = the slope of rating deterioration, a = annual average daily traffic. It can be seen clearly that the methods available in the lite rature have been generated using data as random variables and th at they lack the relevant technical input. 7

PAGE 18

Although researchers have incorporated complex concepts such as neural network (Yang et al, 2005) and nonlinear regression models (Ortiz et al, 2006) into the pavement condition prediction, there is still much more re search that have to be performed. Hence, an enhanced method of integration of engineering knowledge into the development of a more accurate prediction model is presented in this paper. Condition Prediction Based on Markov Models According to the Markov Chain based method of crack condition-prediction, which has been described in detail by Butt (1991), a pavement condition measuring scale can be divided into discrete intervals called condition states . In the case of the crack rating the scale can be divided into 10 condition states each 1 unit wide. In order to accurately predict the likely futu re behavior of pavements which are currently at a given condition state, in terms of probabilities, the transition probability matrix can be used (Shahin et al, 2003). In general, a Transition Probability Matrix (TPM) is used when the condition of a facility is transi ting from one state (i) to the ne xt lower state (j) in a single step as shown in Equation 3. } {1iXjXPpn n ij (3) Where the transition probability matrix [P] consists of the one-step transition probabilities, pij. The most basic, yet time consuming, method to determine probability of the TPM elements is to solely use the historic data. In order to find the transition probability matrix [P] Pii is defined as the probability of a pavement section remaining in the same condition state in the following year and Pij is the probability that the pavement condition state degrades from i to j. as stated before, it is assumed that i and j 8

PAGE 19

cannot differ by more than one [1] state. Usi ng historic data, one can find the number of sections that remained in the same condition state (i) in each year (Nii) and also the number of sections that degraded into the lower condition state (Nij). Then Equation 4 can be used to find the probability Pij. i ij ijN N P (4) Where: Ni is the number of sections that started the year in condition state i A shortcoming of this method is that the proportions are likely to vary from year to year thereby acquiring an average to be us ed to ensure accuracy Also, the application of this method can be problematic in many ag encies due to the insu fficiency of reliable historic data. Since a simple averaging proce ss might not be significantly accurate to be used in high-level analysis, in this thesis, a more sophisticat ed and reliable mathematical method is applied. The Markov chain is said to be time homogeneous if the tr ansition probabilities from one state to another (pij) are independent of the time. The m-step transition probability is the probability of transitioning from state i to state j in m steps as shown in Equation 5. } {)(iXjXPPn mn m ij (5) Therefore, by applying the Markov Chain rule, the state vector at time m [P(m)] can also be found in terms of the transition probability matrix [P] and the initial state vector, P(0). mPPmP )0()( (6) 9

PAGE 20

10 By applying the above formulation to the pavement crack rating data recorded in the PCS database, the future crack condition of a pavement section can be predicted. If this process is applied to all the sections pr esent in the roadway network database of the state, network level rehab ilitation decisions can be ma de effectively based on the established tolerance levels. This development would certainly enhance the planning process of a Pavement Management System (PMS). Updated TPM Development Methods One of the most common and time efficient methods to develop TPMs is by observation of deterioration tre nds. Identification of specific hi storic data and analysis of trends will ultimately lead to the developm ent of TPMs. Although this method is the most convenient approach of developing TPMs, it requires historic data In the absence of historic data an alternative em pirical method that can be used to estimate the TPM is to use expert opinion from a panel of experien ced engineers. However, in recent years, research has been done to develop more scientific methods to obtain TPMs. Among them is a method that involves the use of a recu rrent or a dynamic Markov chain for modeling the pavement crack performance with time in which the transition probabilities are determined based on a logistic model (Ya ng et al., 2005). In this method, a dynamic Markov chain process is presumed to work for the pavement condition survey database available for the entire roadway network of Florida. The limitations of such a methodology roots back to the limitations of Markovian models in general. In research reported in Ortiz-Garcia ( 2006) three alternative methods are proposed to improve the efficiency of developing TP Ms. The first method assumes that the raw

PAGE 21

data (i.e. CR) used in the regression analysis of the deterministic model are readily available. If the condition of a site j at timet is denoted by cjt, the objective function Z can be given by: tj jttyc Z2)( min (7) Where: y(t) is the average pavement condition at time t The objective function aims, therefore, at minimizing the sum of the squared differences between each of the data points and the average conditi on calculated from the distribution of a condition, at. The second method also uses the raw da ta, but after a regression equation has been obtained to describe the progression. If y(t) denotes the regression equation, the objective function, Z, employed to obtain the transiti on probabilities is as follows: ttyty Z2)()(min (8) The objective function aims, therefore, at minimizing the difference between the average of at and the ordinates of the regression e quation. This minimizes the distance between the regression curve and the transition matrix fitted curve. In the third method the raw data are aggregated into bands of condition and presented in the form of distributions. Using the same nomenclature as above, if at(i) denotes the ith element of the TPM predicted distribution at time t, and at(i) is the ith element of the original data distributions at time t, the objective function Z takes the form: ti t tiaia Z2 ')()( min (9) 11

PAGE 22

12 It must be noted that using Equation 5, at can be obtained as at = a0 Pt. The objective function aims, therefore, at minimizing the differenc e between the distributions of condition obtained from the ra w data and the distributions predicted by the transition probabilities. It may be observed from the definitions of the three different objective functions, as in Equations 7-9, that the itera ted values in the optimization process are the element probabilities, pij, of the transition matrix. In the nonlinear optimization algorithm used by Ortiz-Garcia (2006) a search is made for the optimum pij from initial pij values. It assesses the gradient of the objective function on the curr ent region and changes the pij along the path of greatest gradient. The s earch continues until the objective function cannot be minimized further. After analysis, the author determined that the third method would yield the most optimized and practical tr ansition matrix to be used in the present methodology. Problem Statement One shortcoming of the current practi ce is that TPMs are developed based on observation of trends in historic al data or by using expert op inion. Additionally, all of the studies and researches have been carried out with no different iation among different deterioration trends based on the respective cause s of deterioration. Su ch shortcomings of application of TPMs on the network database may have forc ed Florida Department of Transportation (FDOT) to disregard the prediction method in their decision-making and utilize simple crack thresholds in their rehabilitation decisions.

PAGE 23

13Proposed Research Development of an improved and more practical TPM could enhance the current prediction process significantly. Therefore, optimizing the historical data-based TPM by using mathematical techniques to improve th e accuracy of prediction models is one objective of this thesis. Ca tegorizing the database into two independent groups of excessively trafficked sections and struct urally deficient sections due to postponed rehabilitation and the development of speci fic TPMs for each group is another objective of this thesis. Thesis Organization This thesis is divided into four chap ters. The first chapter is the introduction. Chapter Two consists of a detailed methodology and procedures that are used to obtain the results. Chapter Three is the results and the appropriate analysis. Finally, the conclusions and limitations are disc ussed in the fourth chapter.

PAGE 24

14 Chapter Two Experimental Methodology Data Filtering There are geometric and pavement condition data on approximately 9000 pavement sections in the Flor idas 2007 PCS database. To fac ilitate the handling of such a vast amount of data for the analytical need s of this project, the entire database was divided into 7 parts which correspond to the se ven (7) administrative districts of Florida. This process provides manageable sub-database s which are easier to handle. In addition, the subdivision has the advantage that if the geographical effects were to be considered, the database would already be divide d into desired geographical boundaries. Since the Crack Rating (CR) is a subj ective rating by its very nature and a substantial degree of human error is involved in it, data must be first filtered to eliminate abnormalities. The filtering process will ensure th at the sections that have unusual trends are eliminated and will not be allowed to affect the results. Unusual trend can be defined as a sudden CR drop (more than 2 states pe r year) or a sudden CR increase due to erroneous rating recorded with no obvious sign of rehabilitation or cycle change. What would remain in the database is a series CR records in declining order within each construction cycle for each section.

PAGE 25

15Impact Grouping After the filtering process was complete another sub-division was needed to further clarify the filtered sections into sm aller and more specifically oriented batches. Based on previous research (Yang, et al, 2002 ) construction duty cycle have been seen to have a critical impact on crack ing and deterioration of the pavement. Therefore, the duty cycle can be identified as a major categorizing criterion. In this respect, the most recent cycle of a section would dete rmine the group it belongs to (i.e. cycle 1, 2, 3). After completion of this process, it was observed that most sections in Floridas PCS database were in their 2nd or 3rd duty cycle. Hence, operating in ei ther cycle 2 or 3 was chosen to be one criterion for categorization. Next step was to identify other significant attributes that lead cracking to approach CR based threshold conditions. For th e purpose of this thesis, two such effects were chosen, (1) heavy traffic impact and (2) loss of structural integrity due to delayed maintenance and rehabilitation. In order to understand the effect of heavy traffic on the deterioration of a pavement, sections that ar e currently operating under traffic levels of C or worse were chosen for the traffic impact study. On the other hand, sections with low pre-rehabilitation CR values (equal or less than FDOTs 6.4 threshold value) were grouped for the low structural integrity impact study. Amongst the s ections in the heavy traffic impact set, the sections that had lo w CR values before rehabilitation (for the considered construction cycle) and a traffic volume that was close to the boundary of traffic levels C and B, were excluded and adde d to the structural integrity impact group. Similarly, the sections that had pre-rehabilitation CR value close to the th reshold value and relatively high traffic volumes were rem oved from the structural integrity set and

PAGE 26

16 transferred to the traffic impact study group. Th en the relevant data of all the sections were transferred to a database where pre-re habilitation CR value, Average Daily Traffic (ADT), Equivalent Single Axle Load (ESA L), and the crack ratings for the desired construction cycle were recorded. After th e completion of this meticulous sorting procedure, the database was ready for statistical analysis. Table 1. Statewide comparison of grouping outcome Traffic Low Structural Integrity Districts Ave. ESAL Ave. AADT Ave. CR Ave. CR @ pre M&R year Ave. ESAL Ave. AADT Ave. CR Ave. CR @ pre M&R year 1 11,921,631 16.4 7.1 6.5 7,317,625 13.5 7.7 4.5 2 10,635,814 14.1 6.7 7.0 3,878,968 12.4 6.8 4.9 3 7,597,771 8.2 8.3 6.6 4,378,110 10.2 6.9 4.0 4 14,975,981 10.2 7.7 8.1 4,545,181 10.1 6.6 5.5 5 12,301,748 10.9 7.8 7.5 8,164,607 9.8 7.2 4.9 6 9,989,384 8.2 8.3 7.7 4,712,325 5.3 7.4 5.9 7 18,380,200 12.6 8 6.9 5,511,969 8.4 7.0 4.7 Total 12,257,504 12 8 7 5,501,255 10 7 5 TPM Development When the grouping was completed as mentioned above, each group contained two (2) subdivisions (cycle 2 and 3) for each of the seven (7) districts of Florida. For the purpose of generating the Transi tion Probability Matrices (TPM ) for the entire state of Florida, all the data belongi ng to each of the subdivision s were placed in different databases based on the initial grouping and duty cycle disc repancies. Then, the TPM generation process was performed separately to produce a specific TPMs representative

PAGE 27

of each sub-division. The outcome was four different TPMs which correspond to the specific criteria used to de velop them (i.e. two groups with two cycle each). In order to develop the TPM for each s ubdivision, a percentage of the sections were randomly taken out of the specific batch, by using a random number generating function built into Microsoft Excel, and placed in a different database. This small group is then used to test the accu racy of the TPMs developed based on the larger group. To determine the percentage of sec tions to be used for testing, three different percentages, 5%, 10%, and 20%, were tried out. For each of the remaining larger groups (95%, 90%, and 80%) a mean CR value was calculated for each year. To be consistent with other engineering ratings and indices assigned to pavements and the resulting TPMs, the TPM for this thesis has been set to have10 states of length 1, in the CR scale of 0-10, as can be seen in Equation 10. Equation 10 is an expansion of Equation 3 in which pi corresponds to pii and qi corresponds to pij. State p1 0 0 0 0 0 0 0 0 0 q1 p2 0 0 0 0 0 0 0 0 0 q2 p3 0 0 0 0 0 0 0 0 0 q3 p4 0 0 0 0 0 0 0 0 0 q4 p5 0 0 0 0 0 0 0 0 0 q5 p6 0 0 0 0 0 0 0 0 0 q6 p7 0 0 0 0 0 0 0 0 0 q7 p8 0 0 0 0 0 0 0 0 0 q8 p9 0 0 0 0 0 0 0 0 0 q9 1 1 2 3 4 5 6 7 8 9 10 (10) 17

PAGE 28

The most convenient method to obtain p1 through p9 is by observation of the trend of the mean CR values (p10 is always 1 since the pavement cannot deteriorate any further from the 10th state and the equivalent of q10 = 0). This trend generally produces an Scurve indicating that CR must be stable at high ratings (CR>8). Then CR degradation must be sharp for intermediate rating values (5 < CR < 8) and finally follow a more gradual degradation trend for lower ratings ( CR < 5) since deteriora tion rate slows down after CR surpasses a threshold st ate. According to this esta blished trend, a preliminary overall TPM was developed to encompass all fo ur groups solely based on observation of the general deterioration trend in the mean CR values in the PCS database. Then by using a multiple nonlinear regression function built in Microsoft Excel an optimum TPM was obtained for each group. The pro cess to obtain these optimized TPMs is described in the following section. Explanatory TPM Optimization The first step of optimization is to use the preliminary overall TPM and predict future CR values by post-multiplying the TPM by the current perfect CR vector shown in Equation 11. 1 0 0 0 0 [C] = Perfect Initial Condition CR Vector = 0 (11) 0 0 0 0 18

PAGE 29

19 Since the Markov chain rule is applied, the condition vector of each year can be post-multiplied by the TPM to obtain the conditi on vector of the following year only. The length of the analysis was set to 15 years of age since most pavement sections are rehabilitated before reaching this age and sect ions older than 15 year s of age are found in the database only occasionally. In the next step, the expected value of the following years CR can be determined by multiplying the previously obtained CR vector by the state average CR vector in Equation 12. [A] = State Average CR Vector = [9.5 8.5 7.5 6.5 5.5 4.5 3.5 2.5 1.5 0.5] (12) To better illustrate the mentioned matr ix operation, the E quations 13 and 14 are used to determine the condition of the pavement section after one year and after m years respectively. [CR]1(10x1) = [P](10x10).[C](10x1) (13) [CR]m(10x1) = [P]m (10x10).[C](10x1) (14) Where [CR]1 and [CR]m are the crack rating vectors after one and m years of rehabilitation respectively, and [P ] is the developed TPM. If the future crack rating of a pavement se ction after m years is to be predicted, the following equation can be used to calculate the expected crack rating: CR predicted= [A](1x10).[CR]m(10x1) (15) Now, there are two sets of CR ratings for the analysis period of 15 years: one that is calculated by using the preliminary TPM and Equation 15 and the other is the mean CR value of the specific sections (CRdatabase) which can be obtained from the database. To

PAGE 30

optimize the TPM for each group, the calculate d CR value from Equation 15 is set equal to the mean CR value. CRpredicted = CRdatabase for i=1,2,,15 (16) The multiple nonlinear optimization functi on then iterates the TPM elements to implement the equality with such defined constrains as the p and q values are between 0 and 1 in Equation 10, and other elements of the TPM are zero. This equating process should be performed for each year so that when it is completed, the manipulated TPM would be optimized. Then the Mean Square Error (MSE) was calculated to check the difference between the average CR values and the TPM predicted values in each year. n CR CR MSEn i database predicted 1 2) ( (17) Verification of TPM Next step is test the deve loped TPM on the small set of validation sections that was set aside originally. To do so, the mean CR values of the validation group are calculated for each year (CRsmallaverage) and compared to the TPM prediction. Again the MSE is calculated to obser ve the differences. 15 ) (15 1 2 i ge smallavera predictedCR CR MSE (18) 20

PAGE 31

Chapter Three Analysis of Results Results Network/Project Level Decision-Making In order to perform a systematic pave ment management process, the following essential steps can be executed at the network and project le vels respectively. Inventory preparation and maintenan ce 21 Pavement condition survey Condition assessment Network Level Condition prediction Condition analysis Work planning Project Level The importance of the database is clearly seen in pavement management especially at the network level. A systematic approach to pavement management would start with network level projects and lead to more in-depth project level tasks. This will ensure optimum budget prioritization and effici ent labor deployment. On the other hand, an ad hoc approach to pavement manageme nt could lead to accumulation of unfunded major M&R requirements (Shahin, 2003).

PAGE 32

22 The FDOT PCS database is designed to contain pavement condition survey data from the entire roadway network of Florid a. Any analysis and decision-making based on this data becomes an input for network leve l rehabilitation decision-making. For instance, finding a threshold for differentiating wellperforming sections from deteriorated sections, based on the crack ratings availabl e in the database, is considered a major network level project that will lead to screen ing of pavement secti ons for rehabilitation. When sections in the database are screened and the critical ones are set aside for more specific analysis and rehabilitation, further consideration of them is a project level activity. To exemplify this point consider a sect ion determined to be at the crack threshold level and hence is earmarked for more detailed analysis (i.e. manual survey), it is considered to be a project level task. The significance of the PCS database on network or project level activities and decision-making is now evident. Therefore, the ensuing section is dedicated to the analysis of results obtained based on the application of the improved TPM development methodology on the PCS database described in Chapter Two. First, results of each step of the study in the methodology section is presen ted and analyzed in sequence. Finally, application of the overall methodology is presented. Applicability of Grouping After the generation of TPMs, a major part of the analysis performed in this thesis was to verify the accuracy and applicability of the grouping process explained in the Experimental Methodology (Chapter Two). The two major groups are the excessively trafficked group and structural deficient group. To illustrate that the two groups are

PAGE 33

23 distinct and have different ch aracteristics, specific statistical methods were used. Since a large number of sections exists in each cat egory (sometimes up to 600 sections), based on the Central Limit Theorem (CLT), normal distribution approximation was used to represent the distribu tion of the CR values at each age. Because of this approximation, the normal distribution table and other characteristics of the normal distribution can be applied to the data. Table 1 and 2 show the information on the all the filtered sections that are currently operating in their second and third construc tion duty cycles, respectively. Table 2. Characteristics of the sections operating in their 2nd cycle Age 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 x1 9.97 9.93 9.79 9.57 9.28 8.91 8.44 7.83 7.33 6.74 6.23 5.83 5.6 5.29 5.25 x2 9.98 9.81 9.7 9.57 9.35 9.16 8.84 8.5 8.04 7.63 7.25 6.99 6.63 6.23 5.99 n1 631 631 631 630 629 622 614 594 567 520 470 386 278 188 118 n2 232 232 232 232 231 230 227 224 220 210 195 159 139 124 105 1 0.26 0.33 0.6 0.85 1.07 1.22 1.38 1.56 1.69 1.86 2.04 2.1 2.21 2.17 2.2 2 0.15 0.54 0.88 0.95 1.1 1.24 1.32 1.53 1.72 1.85 1.95 1.83 1.95 2.07 2.11 Where x1 and x2 are the sample mean CR values of structural integrity deficient and excessive traffic groups respectively n1 and n2 are number of sections in structur al integrity deficient and excessive traffic groups respectively 1 and 2 are standard deviations of structur al integrity deficient and excessive traffic groups respectively

PAGE 34

Table 3. Characteristics of the sections operating in their 3rd cycle Age 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 x1 9.97 9.87 9.58 9.29 8.95 8.54 7.92 7.29 6.64 6.17 5.6 5.69 5.42 5.33 5.27 x2 9.99 9.92 9.84 9.7 9.46 9.05 8.69 8.28 7.75 7.1 6.55 6.53 6.48 6.31 6.11 n1 310 310 310 310 306 300 295 285 271 235 207 143 107 77 57 n2 189 189 189 187 189 189 188 185 180 165 155 116 101 85 59 1 0.16 0.43 0.92 1.14 1.34 1.5 1.61 1.88 1.94 2.08 2.27 2.19 2.38 2.56 2.57 2 0.07 0.3 0.48 0.71 0.9 1.22 1.3 1.55 1.68 2.03 2.18 1.96 1.75 1.88 1.99 Depending on the desired confidence level, the following expression can be used to calculate an interval in which the mean differences, 1-2, would fall at each age. Equation 19 would yield a lower and an upper limit for the 1-2 interval. 24 (19) 21 2 21 / 21)( nn ZXX22Where Z /2 is the two tail normal variate corres ponding to the confidence interval of (1) In the resultant 1-2 interval, 1 and 2 are the population mean CR values of the structural integrity deficiency and ex cessively trafficked groups respectively To be consistent with other engineer ing confidence interval applications, a confidence interval of 95% was chosen for this analysis; thus, =0.025 and Z /2 =1.96. The lower limit (L) and upper limit (U) of this 95% confidence interval at each age are presented in Table 3 and 4.

PAGE 35

Table 4. Difference in mean CR values for the two groups operating in their 2nd cycle Age Limit 1 2 3 4 5 6 7 8 9 10 25 U 0.021 0.189 0.210 0.138 0.085 -0.066 -0.196 -0.429 -0.441 -0.593 L -0.035 0.040 -0.035 -0.140 -0.245 -0.440 -0.602 -0.901 -0.973 -1.187 Table 4. (Continued) Age Limit 11 12 13 14 15 U -0.694 -0.810 -0.617 -0.461 -0.172 L -1.353 -1.516 -1.448 -1.418 -1.303 Table 5. Difference in mean CR values for the two groups operating in their 3rd cycle Age Limit 1 2 3 4 5 6 7 8 9 10 U 0.000 0.020 -0.132 -0.249 -0.320 -0 .267 -0.516 -0.680 -0.775 -0.526 L -0.041 -0.109 -0.378 -0.574 -0.714 -0.752 -1.038 -1.305 -1.450 -1.342 Table 5. (Continued) Age Limit 11 12 13 14 15 U -0.482 -0.332 -0.498 -0.279 0.000 L -1.406 -1.345 -1.629 -1.674 -1.678

PAGE 36

26 If the sign of the upper and lower limits are the same (positive or negative), it means that one of the means is dominant at th at age. In the results shown in Table 3, in the first five years, depending on the intensity of each impact source (extremely high traffic loading or structural inadequacies ) either one can be the dominant cause of deterioration. However, after the age of 6, both signs become negative. This means that after 6 year of age, at a 95% confidence, the pavement sections that are operating in their 2nd cycle tend to have lower mean CR value if they belong to the st ructurally deficient group as compared to the excessive traffic group. The above observa tion holds true for the sections of the 3rd cycle after the age of 3. However, for practical applications, not all the differences would be considered significant. Therefore, a threshold should be set as the minimum required difference in the CR readi ngs for that difference to be significant. After further studies, a difference of one (1) in the CR is determined to be significant. This means that if the average CR values of two different groups differ by more than 1 unit, that difference can be considered signifi cant. In the case of Floridas PCS database, for sections operating in their 2nd construction duty cycle, afte r the age of eight (8) years, the structural deficient pavement sections behave differently from the traffic loading impacted sections. The same conclusion holds true for sections operating at their 3rd duty cycles after the age of seven (7). The significan ce of this finding is th at it shows that the sections that have delayed M&R deteriorate faster than the secti ons that have higher traffic loadings. It is a critical manage rial decision making criterion which will be explained in more detail in the app lication section (Figures 3 and 4).

PAGE 37

0 1 2 3 4 5 6 7 8 9 10 11 012345678910111213141516AgeCR Delayed M&R Traffic Figure 3. Comparison of degradation between structural deficient and excessive traffic impact (construction duty cycle =2) 0 1 2 3 4 5 6 7 8 9 10 11 012345678910111213141516AgeCR Delayed M&R Traffic Figure 4. Comparison of degradation between structural deficient and excessive traffic impact (construction duty cycle =3) 27

PAGE 38

Another interesting finding of the mean comparison is that as the age increases, so does the interval length betw een the upper and lower limits This phenomenon can be due to the randomness involved with the data and the fact that as the sections age, more randomness is introduced to the data points (Figures 5 and 6). Additionally, as the interval length increases, it becomes harder a nd more challenging to pr edict the ratings in the future. This is why a more scientific and mathematically involved procedure is needed to develop the TPM rather than pure observation. 95% Confidence Interval-1.8 -1.6 -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 012345678910111213141516 AgeMean Difference Upper Limit Lower Limit Mean Difference Significant Mean Difference Treshold Figure 5. Mean difference between structural deficient and traffic groups (construction duty cycle =2) 28

PAGE 39

95% Confidence Interval-1.8 -1.6 -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 012345678910111213141516 AgeMean Difference Upper Limit Lower Limit Mean Difference Significant Mean Difference Treshold Figure 6. Mean difference between structural deficient and traffic groups (construction duty cycle =3) Developed TPMs Now that it is proven there is a differe nce between the deterioration rates of sections depending on the cause of deterior ation, different TPMs can be developed to represent each category. The matrices repr esented in Equations 20 through 23 are the TPMs developed based on the methodology explaine d in prior sections. Equations 20 and 21 are TPMs for structurally defici ent sections operating at their 2nd and 3rd cycle respectively, and Equations 22 and 23 are TP Ms for excessively trafficked sections operating at their 2nd and 3rd cycle respectively. 29

PAGE 40

0.89188 0 0 0 0 0 0 0 0 0 0.09947 0.68699 0 0 0 0 0 0 0 0 0 0.28838 0.59811 0 0 0 0 0 0 0 0 0 0.39825 0.44981 0 0 0 0 0 0 0 0 0 0.54977 0.40018 0 0 0 0 0 0 0 0 0 0.60010 0.35021 0 0 0 0 0 0 0 0 0 0.65014 0.30014 0 0 0 0 0 0 0 0 0 0.70009 0.25007 0 0 0 0 0 0 0 0 0 0.75004 0.15002 0 0 0 0 0 0 0 0 0 0.85001 1 0.89371 0 0 0 0 0 0 0 0 0 0.09796 0.69009 0 0 0 0 0 0 0 0 0 0.29086 0.59963 0 0 0 0 0 0 0 0 0 0.39942 0.45054 0 0 0 0 0 0 0 0 0 0.55033 0.40060 0 0 0 0 0 0 0 0 0 0.60041 0.35043 0 0 0 0 0 0 0 0 0 0.65029 0.30024 0 0 0 0 0 0 0 0 0 0.70015 0.25010 0 0 0 0 0 0 0 0 0 0.75006 0.15003 0 0 0 0 0 0 0 0 0 0.85002 1 (21) (20) 30

PAGE 41

0.90143 0 0 0 0 0 0 0 0 0 0.10496 0.69269 0 0 0 0 0 0 0 0 0 0.29350 0.59911 0 0 0 0 0 0 0 0 0 0.39918 0.44992 0 0 0 0 0 0 0 0 0 0.54991 0.40009 0 0 0 0 0 310 0 0 0 0.60005 0.35010 0 0 0 0 0 0 0 0 0 0.65006 0.30006 0 0 0 0 0 0 0 0 0 0.70004 0.25003 0 0 0 0 0 0 0 0 0 0.75002 0.15001 0 0 0 0 0 0 0 0 0 0.85001 1 0.90399 0 0 0 0 0 0 0 0 0 0.10708 0.69050 0 0 0 0 0 0 0 0 0 0.29128 0.59953 0 0 0 0 0 0 0 0 0 0.39937 0.45042 0 0 0 0 0 0 0 0 0 0.55026 0.40051 0 0 0 0 0 0 0 0 0 0.60035 0.35036 0 0 0 0 0 0 0 0 0 0.65024 0.30020 0 0 0 0 0 0 0 0 0 0.70013 0.25009 0 0 0 0 0 0 0 0 0 0.75005 0.15003 0 0 0 0 0 0 0 0 0 0.85001 1 (22) (23) In all cases, it can be seen that the CR va lues decrease sharply after they pass the beginning condition and flatten out once they reach the degradation threshold where the pavement cannot deteriorate significantly any further. This behavior agrees with the Sshape curve that is used to represent the degradation of pavement condition.

PAGE 42

Applicability of Developed TPMs Before applying the developed TPMs on the data, some statistical analysis must be performed to verify the accuracy and appli cability of the TPMs. In order to do so, the TPM predicted CR values are plotted against the average CR values of the small group of verification sections (Figures 7 through 10). Then, the mean square error is used to verify the accuracy of the developed TPMs. As it can be seen in the following figures, the errors were insignificant and negligible in term of crack rating; therefore, it can be concluded that the developed TPMs are repr esentative of the actua l trends that exist in the database. 0 1 2 3 4 5 6 7 8 9 10 11 01234567891011121314151617 AgeCR 5% average TPM Predicted Figure 7-A. Comparison with the 5% verification set at their 2nd duty cycle (structural integrity deficient sections) 32

PAGE 43

0 1 2 3 4 5 6 7 8 9 10 11 01234567891011121314151617 AgeCR 10% average TPM Predicted Figure 7-B. Comparison with the 10% verification set at their 2nd duty cycle (structural integrity deficient sections) 0 1 2 3 4 5 6 7 8 9 10 11 01234567891011121314151617 AgeCR 20% average TPM Predicted Figure 7-C. Comparison with the 20% verification set at their 2nd duty cycle (structural integrity deficient sections) 33

PAGE 44

0 1 2 3 4 5 6 7 8 9 10 11 012345678910111213141516 AgeCR 5% Average TPM Predicted Figure 8-A. Comparison with the 5% verification set at their 3rd duty cycle (structural integrity deficient sections) 0 1 2 3 4 5 6 7 8 9 10 11 012345678910111213141516 AgeCR 10% Average TPM Predicted Figure 8-B. Comparison with the 10% verification set at their 3rd duty cycle (structural integrity deficient sections) 34

PAGE 45

0 1 2 3 4 5 6 7 8 9 10 11 012345678910111213141516 AgeCR 20% Average TPM Predicted Figure 8-C. Comparison with the 20% verification set at their 3rd duty cycle (structural integrity deficient sections) 0 1 2 3 4 5 6 7 8 9 10 11 01234567891011121314151617 AgeCR 5% average TPM Predicted Figure 9-A. Comparison with the 5% verification set at their 2nd duty cycle (excessive traffic sections) 35

PAGE 46

0 1 2 3 4 5 6 7 8 9 10 11 01234567891011121314151617 AgeCR 10% average TPM Predicted Figure 9-B. Comparison with the 10% verification set at their 2nd duty cycle (excessive traffic sections) 0 1 2 3 4 5 6 7 8 9 10 11 01234567891011121314151617 AgeCR 20% average TPM Predicted Figure 9-C. Comparison with the 20% verification set at their 2nd duty cycle (excessive traffic sections) 36

PAGE 47

0 1 2 3 4 5 6 7 8 9 10 11 012345678910111213141516 AgeCR 5% Average TPM Predicted Figure 10-A. Comparison with the 5% verification set at their 3rd duty cycle (excessive traffic sections) 0 1 2 3 4 5 6 7 8 9 10 11 012345678910111213141516 AgeCR 10% Average TPM Predicted Figure 10-B. Comparison with the 10% verification set at their 3rd duty cycle (excessive traffic sections) 37

PAGE 48

0 1 2 3 4 5 6 7 8 9 10 11 012345678910111213141516 AgeCR 20% Average TPM Predicted Figure 10-C. Comparison with the 20% verification set at their 3rd duty cycle (excessive traffic sections) However, for each case an optimum veri fication group size should be determined to be applied in future studies. A mere co mparison as presented in Table 6 cannot be deterministic because as it can be seen in the relative figures, some verification plots have undesired abnormalities. The abnormalities usually happen if the test batch is small (5% or 10%). This might be due to their small sa mple size, but once the sample size increases, the plots smoothen out. Table 6. Comparison of Mean Square Error (MSE) Structural Integrity Deficient Excessive Traffic 5% 10% 20% 5% 10% 20% Cycle =2 0.267 0.314 0.272 0.143 0.088 0.082 Cycle =3 0.627 0.214 0.156 0.854 1.314 0.384 38

PAGE 49

39 Based on the review of the findings in Table 6 and Figures 7 through 10, the verification group size in Ta ble 7 is recommended for application. The above recommendation was based on the logic that if the comparison study with a smaller test group would yield the same results as that with a larger group, the smaller test sample in fact would be adequate for verification, hence introducing less ra ndomness to the study. Table 7. Recommended verification group size Structural Integrity Deficient Excessive Traffic Cycle =2 10% 10% Cycle =3 20% 20% Application of Test Results It can be seen from Figures 5 and 6 that the pavement sections currently performing in their 3rd construction duty cycles have a fa ster deterioration rate (in terms of cracking) compared to the pavement sections currently performing in their 2nd construction duty cycles. However, after reachi ng a threshold state (around a crack rating of 6) all the pavement secti ons, regardless of the cycle they are operating in, deteriorate with the same gradual rate. If further rese arch can be performed on ensuing construction duty cycles (i.e. 4th and 5th cycle) and the similar results ho ld true (i.e. pavement sections performing in higher cycles deteriorate fast er), then this observation can significantly impact the rehabilitation decisions. As th e duty cycle of a pavement increases, its vulnerability to deterior ation also increases. Additionally, statistical analysis verified the applicability of the grouping and the development of individual TPMs. The study of two groups of structural integrity

PAGE 50

40 deficient and excessively trafficked pavement s showed that not all the deterioration and crack propagation is due to traffic loadi ng. Sometimes a pavement section with low traffic can reach a low crack rating state in a short time because of the poor sub-surface condition of that pavement at the time of overlay. The dete riorated condition of the subsurface can induce bottom-up cracking since th ere are high stress areas at the distress locations underneath. The bottom-up cracking can propagate and inte nsify quicker than the top-bottom cracks since it feeds from tw o sources, loading on top and stresses at bottom. Next step is to demonstrate how this analysis can improve the engineering decision-making practice. The results show that the decision-making should not always focus on the highly trafficked sections (level s C, D, and E), but al so consider the low trafficked sections that have experienced low crack ratings before rehabilitation. These sections have the potential to deteriorate faster at times; therefore, making them high priority candidates for maintenance and rehabilitation. Knowing the cause of the lo w rating of a given pavement section is helpful when prioritizing the projects for rehabilitati on. Based on the rating of a section and the probable cause (structural deficiency or exces sive traffic), a pavement manager can use the prediction models based on the relevant transition probability matrices to determine and prioritize the severity of the secti ons in any desired time span. This would considerably help in budget optimization at the network leve l decision-making. There are two important comparisons that can be made to check the applicability and accuracy of the developed TPMs with the actual data. FDOT has been using a linear regression computation to predict the CR fo r 5 years after the last CR entry of the database (currently at 2007). Four sections were chosen for comp arison purposes each

PAGE 51

41 performing in a different cycle and belonging to different group as represented in Table 8 through 11. Table 8. Prediction comparison of a structurally deficient section in its 2nd cycle Section Roadway Roadway Begin End CR 2007 Characteristics Direction ID Milepost Milepost C 01040000 0 0.887 9 TPM Predicted FDOT Predicted Condition State Difference CR 2012 7.24 8.30 1 Table 9. Prediction comparison of a structurally deficient section in its 3rd cycle Section Roadway Roadway Begin End CR 2007 Characteristics Direction ID Milepost Milepost C 01040000 0.887 1.47 8.5 TPM Predicted FDOT Predicted Condition State Difference CR 2012 6.27 7.16 1 Table 10. Prediction comparison of an ex cessively trafficked section in its 2nd cycle Section Roadway Roadway Begin End CR 2007 Characteristics Direction ID Milepost Milepost L 70050000 9.956 15.158 9 TPM Predicted FDOT Predicted Condition State Difference CR 2012 7.58 8.28 1 Table 11. Prediction comparison of an ex cessively trafficked section in its 3rd cycle Section Roadway Roadway Begin End CR 2007 Characteristics Direction ID Milepost Milepost R 36030000 0.652 2.606 6.5 TPM Predicted FDOT Predicted Condition State Difference CR 2012 3.48 5.68 2

PAGE 52

42 As it can be seen in Tables 8 through 11, the difference between the predictions is significant for pavement management purposes As explained before, the condition state difference of one (1) is significant in paveme nt management to differentiate between the future condition of a sections. Moreover, in Table 11, this difference is two states. This can be due to the fact that the CR at this se ction is at a steep degradation stage of the Scurve deterioration trend. Therefore, depending on how the prediction method is developed, the forecasted CR can vary signi ficantly from one prediction method to another. Although a conclusion cannot be drawn at this time, once the crack rating survey results are available, the accuracy of both methods can be checked. Another important comparison can be made in the near future when the CR data for the year 2008 is recorded in the databa se. By comparing the actual data with the predicted values from the developed TPMs, the predictability of the developed TPMs would be verified. The 2008 predicted CR valu es for the same four sections selected above are shown in Table 12. Table 12. Predicted CR values for 2008 Section ID 01040000 01040000 70050000 36030000 CR 2007 9 8.5 9 6.5 CR 2008 8.7 8 8.8 6

PAGE 53

43 Chapter Four Conclusions and Limitations Conclusions As explained in the Chapter Two (Expe rimental Methodology), two distinct pavement groups have been identified in this work: structural integrity deficient group and excessively trafficked group. Also, for each group two different TPMs are developed based on the current construction duty cycle of the pavement sections (cycles 2 and 3). In this study, the two criteria used to group the data, the deterioration cause and current duty cycle, were proven to be stat istically significant. By compar ing the deterioration rates the two study groups of excessive trafficked sect ions and structural integrity deficient sections, it was proven statisti cally that the latter group, the pavement sections that have lower CR at the time of rehabilitation a nd low traffic volume (ESAL < 3,000,000), tend to deteriorate faster than the pavement sections that have a higher CR value at the time of rehabilitation and high traffic volume (ESA L > 3,000,000). This can be attributed to the degraded strength and support of the underlyi ng pavement layers. The significance and applicability of this result is evident in pavement management decision-making where projects are prioritized for maintenance a nd rehabilitation. Based on the above findings, the sections that have low traffic but low cr ack ratings at the time of rehabilitation must

PAGE 54

44 also be considered as top priorities. To illust rate this point, two sections operating close to their crack threshold can be considered: one with excessive traffic and another with exposure to severe low crack leve ls in earlier cycles. By using the results from this thesis, it can be concluded that the latter section will deteriorate faster and its delayed rehabilitation might be much more costly and in extreme cases even impractical. This conclusion is somewhat contrary to the usual tendency of agencies to prioritize the rehabilitation of an excessively trafficked section. Also, by using the relevant TPM, the remaining life span of a section can be estimated more accurately until reaches its threshold. This would help the agencies to appropriate the budget based on the order in which the sections would reach thei r crack threshold level. In general, it was observed that the pave ment sections currently performing in their 3rd construction duty cycles have a faster deterioration rate (in te rms of cracking) as compared to the pavement sections currently performing in their 2nd construction duty cycles. The reason for this phenomenon can be that after each rehabilitation and as the pavement ages, the pavement materials get fatigued which ultimately lead to a faster deterioration rate. This fact is also crucial when prioritizing the projects in that pavements in their 3rd duty cycle should be prioritized over th e pavement sections currently in their 2nd duty cycle. Although the developed TPMs for each group seem to be approximately equal to each other, once they are applied to a pa vements life span, the differences would accumulate and the predicted difference in behavior in the two groups would be evident. Also, the confidence interval for the predicted crack rating grows with the age. This phenomenon can be explained by considering the randomness involved in the rating and,

PAGE 55

45 the variation of the pavement condition with ag e. This suggests that predicting the future condition of pavement sections in the 3rd cycle is a more challenging task compared to that of pavement sections in the 2nd cycle. Overall, it can be concluded that gro uping the pavement sections based on the degradation cause and developing relevant in dividual TPMs is a more accurate mean of predicting pavement behavior. In this manner, instead of applying the same TPM to predict the future condition of all sections of a database, more specific and appropriate TPMs can be developed for e nhanced condition prediction. Limitations Access to an up-to-date database is the key for successful grouping and prediction. If the available data are limited to certain areas or specific time periods, it would not result in accurate prediction. As an example, in order to extract and filter the sections suitable for the structurally defi cient group from the PCS database, it is necessary to know the crack ra ting of the year the rehabilitation was performed. Same need holds true for the excessively trafficked group of pavements with the availability of all of the necessary tr affic information. Statistical analysis is based on acceptance of normality of the data set. This assumption is justified by the large number of samples used in this study under the applicability of the Central Limit Theorem to them. If a small sample is available for analysis, the normality should be checked or other appropriate approximation must be used.

PAGE 56

46 References American Association of State Highway and Transportation Officials, (2001). Standard Practice for Quantifying Cracks in Asphalt Pavement Surface (No. PP 44-01), AASHTO, April Publication No. HM-20-COM Kong, F. (2002), Shiou-San Kuo, Ph.D., P.E., Hesham S. Mahgoub, Ph.D., Lorrie L. Hoffman, Ph.D.. Transportation Research Board, Development of Flexible Pavement Performance Prediction Model Based on Pavement Data Kumara, M.W. (2003). Dr.M.Gunaratne, Dr.J.Lu, Bruce Dietrich, Journal of Materials in Civil Engineering, Methodology for Random Surface-Ini tiated Crack Growth Prediction in Asphalt Pavements . Ortiz-Garcia, J. (2006), Seosamh B. Costello, Martin S. Snaith, Journal of Transportation Engineering, ASCE, Derivation of Transition Probability Matrices for Pavement Deterioration Modeling. Shahin, M.Y., Pavement Management for Airports, Roads, and Parking Lots.New York: Springer. 2005. Yang, J. J. (2006), Dr.M.Gunaratne, Dr.J.Lu, Br uce Dietrich, Journal of Transportation Engineering, ASCE. Use of Recurrent Markov C hains for Modeling the Crack Performance of Flexible Pavements .

PAGE 57

47 Appendices

PAGE 58

48 Appendix A: History of Florid a Pavement Condition Survey (1973 ) Revised 09/24/2006 1973 Complete Flexible Pavement Survey performed by the Districts. 1974 Complete Flexible Pavement Survey performed by the Districts. 1975 Complete Flexible Pavement Survey performed by the Districts. 1976 Complete Flexible and Rigid Pavement Surveys performed by the Districts 1. Rigid pavement survey was newly added. 1977 Complete Flexible and Rigid Pavement Surveys performed by the Districts. 1978 Complete Flexible and Rigid Pavement Surveys performed by the Districts. 1979 Complete Flexible and Rigid Pave ment Survey by the Districts. 1980 No survey was performed due to change over in equipment Mays Ride Meters originally mounted in survey vehicles were to mounted on Standard Trailers. 1981 Complete Flexible and Rigid Pavement Surveys performed by the Districts. 1) Flexible had ride values above 100 no upper limit. 2) First survey using PCR's and trailers. 3) Number of Lanes was added to the surv ey data collection table.

PAGE 59

49 Appendix A: (Continued) 1982 Complete Flexible and Rigid Pavement Surveys performed by the Districts. 1) Pavement Type 7 & 8 were added to th e Flexible PCS Dist ricts 3 & 5 did not use these codes. 2) Started calculated Ride between 1 & 5 if section was too short to test to prevent basic ratings of 0. 1983 Complete Flexible and Rigid Pavement Surveys performed by the Districts. 1) Pavement Type 7 & 8 were used by all Districts. 2) Procedure for calculating Ride was included in the manual. 3) Ro adway 4 code was added for twolane roads to give the direction surveyed. 1984 Complete Flexible and Rigid Pavement Surveys performed by the Districts. 1) Defect on sections with a basic ra ting below 60 remaining section adjusted from 1983 survey. 2) Ride was not evaluate d; Ride ratings were adjusted from the 1983 survey data. 1985 Flexible and Rigid Pavement Survey s included Ride only. 1) BM&R tested Districts 2, 3 and 5. 2) BM&R assisted with Districts 4 a nd 6. 3) District 1 conducted the District survey. 4) District 3 rated I-10 ri gid for defects. 5) BM&R collected Ride values on the rigid pave ment of I-10. 6) Defect ratings were adjusted from the 1984 survey data. 1986 Complete Flexible and Rigid Pavement Surveys performed by BM&R personnel. 1) 3 ruts per mile. 2) ADT was eliminate d. 3) Adjusted ratings were eliminated. 4) District 3 personnel rate d own rigid pavements. 5) Survey was started in the

PAGE 60

50 Appendix A: (Continued) second week of September 1985. 6) Survey was completed in the first week of September 1986. 7) BM&R personnel rated one section of rigid pavement per county in District 3 (Intersta te) as a verification of th e rigid survey. 8) Type 6 code was added to survey to reflect No Ride. Ride value will match defect. 9) Added Crack Type to Flexible Survey: A = Alligator, B = Block, or C = Combination. 10) Flexible Miles Rated 15,468.834, Rigid Miles Rated 96.923 Total Miles Rated 15,565.757. Flexible Miles Represented 32,937.004. Rigid Miles Represented 277.744. Total Mile s Represented 33,214.748. 11) Flexible sections rated 5,765. Rigid sections rated 47. Total sections rated 5,812. 1987 Complete Flexible and Rigid Pavement Surveys performed by BM&R personnel. 1) Changes made to computer programs flexible edit, flexible compare, and flexible difference. 2) Survey was star ted in the third week of September 1986, and was completed in the last week of June 1987. 3) Verification of rigid pavement survey in the District 3 was pe rformed on seven sections of Interstate 10. 4) Flexible Miles Rated 16,333.001. Ri gid Miles Rated 937.385. Total Miles Rated 17,270.386. 5) Flexible Mile s Represented 33,010.922. Rigid Miles Represented 2,078.848. Total Miles Repres ented 35,089.770. 6) Flexible sections rated 6,196. Rigid sections rated 398 Total sections rated 6,594. 1988 Complete Flexible and Rigid Pavement Surveys performed by BM&R personnel. 1) Survey was started in the third we ek of August 1987, and was completed in 1st

PAGE 61

51 Appendix A: (Continued) week of May 1988. 2) Flexible Mi les Rated 16,423.565. Rigid Miles Rated 939.608. Total Miles Rated 17,363.173. 3) Flexible Miles Represented 33,334.466. Rigid Miles Represented 2,064.341. Total Miles Represented 35,398.807. 4) Flexible sections rated 6,347. Rigid sections rated 401. Total sections rated 6,748. 1989 Complete Flexible and Rigid Pavement Surveys performed by the State Materials Office. 1) Type 5 (new construction) and Type 7 (new overlay) codes were added. 2) L (light), M(Moderate), an d S(severe) codes were added in the Comments field to indicate th e severity of up to 25% crac king. 3) Survey started 2nd week in June 1988, and was completed in 1st week of May 1989. 4) Flexible Miles Rated 16,715.302. Rigi d Miles Rated 926.118. Total Miles Rated 17,641.420. 5) Flexible Mile s Represented 33,875.971. Rigid Miles Represented 2,052.093. Total Miles Repres ented 35,928.064. 6) Flexible sections rated 6,476. Rigid sections rated 399 Total Sections Rated 6,875. 1990 Complete Flexible and Rigid Pavement Su rveys by the State Materials Office. 1) Survey was started on 6/12/89 and wa s completed on 05/02/1990. 2) Trailers were painted and reconditioned causing delay in survey schedule. 3) Added lanes to Type 9 (structures and/or exceptio n) and Type 8 (under construction). 4) Flexible Miles Rated 17,087.904. Rigi d Miles Rated 922.423. Total Miles Rated 18,010.327. 5) Flexible Miles Represented 34,684.121. Rigid Miles

PAGE 62

52 Appendix A: (Continued) Represented 2,060.555. Total Miles Repres ented 36,744.676. 6) Flexible sections rated 6,571. Rigid sections rated 407. Total Sections Rated 6,978. 1991 Complete Flexible and Rigid Pavement Surveys by State Materials Office. 1) Survey was started on 6/11/90, and was completed on 05/02/91. 2) Programming change to allow a menu driven data entry for Flexible Pavement Survey. 3) A modified cracking method wa s added to Flexible Pavement Survey for evaluation. 4) Added Type 0 to identify an exception, not state maintained, or a duplicate roadway section evaluated unde r another county section number that should be exceptions. 5) All verificati on reports completed on May 09, 1991. 6) Survey on a 0 to 10 scale was introduced fo r Flexible and Rigid. 7) Flexible Miles Rated 16,431.367. Rigid Miles Rated 912.414. Total Miles Rated 17,343.781. 8) Flexible Miles Represented 34,915.445. Rigid Miles Represented 2,009.968. Total Miles Represented 36,925.413. 9) Flex ible sections rated 6,456. Rigid sections rated 397. Tota l Sections Rated 6,853. 1992 Complete Flexible and Rigid Paveme nt Surveys by State Materials Office. 1) Survey was started on 8/05/91, and was completed on 5/04/1992. 2) Ultrasonic Profilers replaced Mays Ride Meters Ride Rating (RR ) = 99.7576 + (-0.1569 X IRI) used until 1999 survey. 3) Rut Depth measured manually and with Ultrasonic Profilers for comparison. 4) 0 to 10 scale implemented for Rut, Ride, and Defect scale as new rating system selected by Pavement Management Committee. 5) Rut scale changed to add 1 1/8" and 1" for 10 scale. 6) IRI reported for outside

PAGE 63

53 Appendix A: (Continued) wheel path only with no filtering .7) IR I converted to PSIsv (10 scale) through correlation to CHLOE Profilometer. Correla tion combined all units at all speeds (30, 40 & 50 MPH) and for both wheel paths. 8) Number of lane s added to Type 9 code (State Maintained exception such as bridges, etc.). 9) Responsibility for HPMS sections added to Survey Personnel. 10) Rut depth (Ultrasonic) in 0.001 mile increments for interstate flexible system was added to mainframe database. 11) Ultrasonic Rut Depth was used for Rut rating (Flexible Pavement Survey). If Type 6 (No Ride) then Manual Rut De pth was used. 12) Cracking scale was adjusted from procedures manual to J= 2.5 if confined to wheelpath (CW), and J=1.0 if outside of wheel path (CO). Adjustments made per Mr. Ken Morefield. 13) Flexible Miles Rated 16,504.153. Rigi d Miles Rated 889.772. Total Miles Rated 17,392.183. 14) Flexible Miles Represented 35,402.349. Rigid Miles Represented 2,020.421. Total Miles Re presented 37,422.770. 15) Flexible sections rated 6,726. Rigid rated sect ion 394. Total sections rated 7,118. 1993 Completed Flexible and Rigid Pave ment Surveys by State Materials Office personnel. 1) Survey started 7/06 /92, and was completed on 4/22/1993. 2) Ultrasonic Rut Depth (Actual Values) were recorded in CC 44-47 in Team File and CC 60-63 in permanent file. 3) New inst ruction manuals flexible and rigid for the Pavement Condition Survey published April, 1993. 4) Released survey 5/28/1993. 5) Flexible Miles Rated 16,662.666.Rigid Miles Rated 861.677. Total Miles Rated 17,523.953. 6) Flexible Mile s Represented 35,765.134. Rigid Miles

PAGE 64

54 Appendix A: (Continued) Represented 1,959.640. Total Miles Repres ented 37,724.774. 7) Flexible sections rated 6,934. Rigid sections rated 389 Total sections rated 7,323. 1994 Completed Flexible Pavement by State Materials Office personnel 1) Survey started 6/07/93. 2) Instructions from Mr. Ken Morefield via Mr. L.L. Smith was to complete flexible survey by April 01, 1994. The rigid pavement will not be accomplished in 1994 in order to co mplete survey by April 01, 1994. 3) Completed survey field-work on Februa ry 3, 1994. 4) Released survey on February 21, 1994. 5) Flexible Miles Rated 16,766.683. Rigid Miles Rated 861.287. Total Miles Rated 17,627.970. 6) Flexible Miles Represented 36,065.275. Rigid Miles Represented 1,959.640. Total Miles Represented 38,024.915. 7) Flexible sections rated 7,026. Rigid rated section 387. Total sections rated 7,413. 1995 Completed Flexible and Rigid Pavement Survey by State Materials Office personnel. 1) Survey started 3/21/94. 2) Light moderate and se vere raveling added to survey as separate identity. 3) Patching added to survey as separate identity. 4) Type 2 added to survey to reflect pa vement improvements without complete overlay (Intersections overlays). 5) Sy stem coded under US number was changed to match system codes. 6) Complete d survey field-work January 26, 1995. 7) Survey released on March 30, 1995. 8) HPMS FHWA added primary and interstate system in one direction Appendix J. 9) Produced PCS and HPMS Facts. 10) Flexible Miles Rate d 16,879.704. Rigid Miles Rated 746.673. Total

PAGE 65

55 Appendix A: (Continued) Miles Rated 17,626.377. 11) Flexible Mile s Represented 36,390.738. Rigid Miles Represented 1,738.909. Total Miles Re presented 38,129.647. 12) Flexible sections rated 7,078. Rigid rated sect ion 347. Total sections rated 7,425. 1996 Completed Flexible and Rigid Pavement Survey by State Materials Office personnel. 1) Survey started 3/27/95. 2) Survey field-work completed 1/17/96. 3) Survey released 3/05/96. 4) Flexib le Miles Rated 17,027.506. Rigid Miles Rated 718.910. Total Miles Rated 17,746.416. 5) Flexible Miles Represented 37,018.830. Rigid Miles Represented 1,694.010. Total Miles Represented 38,712.840. 6) Flexible sections rated 7,209. Rigid rated section 337. Total sections rated 7,546. 1997 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) Survey started 3/22/96, and was comp leted on 1/16/97. 2) Survey released 3/05/97. 3) Flexible Miles Rated 17,121.634. Rigid Miles Rated 692.277. Total Miles Rated 17,813.911. 4) Flexible Mile s Represented 37,307.869. Rigid Miles Represented 1,603.559. Total Miles Repres ented 38,911.428. 5) Flexible sections rated 7,429. Rigid rated section 329. Total sections rated 7,758. 1998 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) Survey started 3/17/97, and was comple ted on 1/13/98. 2) Survey released 4/01/98. 3) Flexible Miles Rated 17,201.156. Rigid Miles Rated 681.677. Total Miles Rated 17,882.833. 4) Flexible Mile s Represented 37,572.317. Rigid Miles

PAGE 66

56 Appendix A: (Continued) Represented 1,592.399. Total Miles Repres ented 39,164.716. 5) Flexible sections rated 7,524. Rigid rated section 330. Total sections rated 7,854. 1999 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) Survey was started on 03/30/98, and was completed on 01/12/99. 2) Survey was released on 3/22/99. 3) Flexible M iles Rated 17,314.411. Rigid Miles Rated 622.325. Total Miles Rated 17,976.736. 4) Flexible Miles Represented 37,925.623. Rigid Miles Represented 1,566.420. Total Miles Represented 39,492.043. 5) Flexible sections rated 7,652 Ri gid rated section 322 Total sections rated 7,974. 6) Converted to laser profilers. 7) Used Ride Number (RN) times 20 for ride rating. Ride number was based on rate 4 filtered to 300 foot wavelength from the outside wheel path. 8) Started using laser profiler for ride acceptance Rate 2 Ride Number (RN) filtered to 300 foot. 9) Warranty specification implemented this year. 2000 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) Survey was started on 03/22/99, and was completed on 1/12/2000. 2) Survey released on 3/24/2000. 3) Flexible Miles Rated 17,486.318. Rigid Miles Rated 605.559. Total Miles Rated 18,091.877. 4) Flexible Miles Represented 38,535.787. Rigid Miles Represented 1,476.148. Total Miles Represented 40,011.935. 5) Flexible sections rated 7,770. Rigid rated section 307. Total sections rated 8,077. 6) Tested Forest Ro ads per Federal High way Administration

PAGE 67

57 Appendix A: (Continued) request. Total miles rated 530.190. Total number of roads 74. 7) Tested HPMS offsystem sections for first time To tal miles rated 357.4. Total sections rated 262. 2001 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) Removed Code for Type and leading zero from State Road number and U.S. Road Number field. 2) Allowed laser meas ured rut depths to be used on Type 6 (no ride) in lieu of manual measuremen ts. 3) Survey started 03/27/2000, and was completed on 01/10/2001. 4) Survey rele ased on 3/12/2001. 5) Flexible Miles Rated 17,624.341 Rigid Miles Rated 546.806 Total Miles Rated 18,170.190. 6) Flexible Miles Represented 38,831.473. Rigid Miles Represented 1,331.175. Total Miles Represented 40,162.648. 7) Flexib le sections rated 7,782. Rigid rated section 302. Total sec tions rated 8,084. 2002 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) Survey started on 04/02/2001, and was completed on 01/17/2002. 2) Added Ride Number to Rut Depth in 0.01 interv als. 3) Added R to indicate profiler reruns under verification codes. 4) Survey released 03/15/2002. 5) Flexible Miles Rated 17,898.876. Rigid Miles Rated 397.640. Total Miles Rated 18,296.516. 6) Flexible Miles Represented 39,428.791. Rigid Miles Represented 1,034.599. Total Miles Represented 40,463.390. 7) Fl exible sections rated 7,777. Rigid sections rated 275. Tota l sections rated 8,052.

PAGE 68

58 Appendix A: (Continued) 2003 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) Added Code for Raters to CC 85 & CC 86 of the Flexible AR EA file. 2) Added Code for Raters to CC 52 & CC 53 of the Flexible PERMANENT file. 3) Survey was started on 03/25/2002, and was comp leted 01/08/2003. 4) Survey released 3/27/03. 5) Flexible Miles Rated 17,916.53. Rigid Miles Rated 369.94. Total Miles Rated 18,286.47. 6) Flexible Mile s Represented 39,800.39. Rigid Miles Represented 978.44. Total Miles Represen ted 40,778.82. 7) Flexible sections rated 7,871. Rigid sections rated 267. Tota l sections rated 8,138. 9) Added rater codes to the area data set in CC 85 & 86. Not included in permanent data set 10) Added to the handbook that all lanes could be considered for overall crack rating (reflective of overall condition). 2004 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) For the 2004 Survey, the profile data is collected using a sampling rate of 6 inch compared to a 12 inch sample interval in previous survey years. 2) Survey started 03/24/2003, and was completed on 01/14/04. 3) Survey released 03/23/04 4) Flexible Miles Rated 18071.48. Rigid Miles Rated 368.24. Total Miles Rated 18439.72. 5) Flexible Miles Represente d 40039.01. Rigid Miles Represented 976.94. Total Miles Represented 41015.50. 6) Fl exible sections rated 7,884. Rigid sections rated 269. Tota l sections rated 8,153. 2005 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) PCS Started 03/29/04, and was comple ted on 12/15/04. 2) Flexible Miles Rated

PAGE 69

59 Appendix A: (Continued) 18061.64. Rigid Miles Rated 363.08. Total Miles Rated 18424.71. 3) Flexible Miles Represented 40380.77. Rigid Mile s Represented 975.7. Total Miles Represented 41356.48. 4) Flexible sections rated 7966. Rigid sections rated 261. Total sections rated 8227. 2006 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) Completed SIS survey June 15, 2005 Miles Rated 185.431. 2) Completed SCRAP/SCOP survey July 31, 2005 Miles Rated 882.672. 3) PCS Started 03/14/05, and was completed on 12/ 14/ 05. 4) Flexible Miles Rated 18251.53. Rigid Miles Rated 364.39. Total Mile s Rated 18615.91. 5) Flexible Miles Represented 40788.13. Rigid Miles Repres ented 993.21. Total Miles Represented 41781, 45. 6) Flexible sections rated 8013. Rigid sections rated 271. Total sections rated 8284. 2007 Completed Flexible and Rigid Pavement Surveys by State Materials Office. 1) All four survey vehicles are us ing Windows XP operating systems. 2) Completed SIS survey 02/28/07 Miles Rated 204.919. 3) Completed SCRAP/SCOP (08/14/2006) Miles Rated 1103.66. 4) PC S Field Work Started 03/20/06, and was completed on 12/19/2006 5) Survey released 03/21/07. 6) Flexible Miles Rated 18328.929. Rigid Mi les Rated 363.891. Total Miles Rated 18692.820. 7) Flexible Miles Represente d 41191.490. Rigid Miles Represented 88.434. Total Miles Represented 42179.924. 8) Flexible sections rated 8199. Rigid sections rated 270. Total sections rated 8469.