Ce travail de thèse consiste au développement de méthodes statistiques pour détecter des com- portements anormaux dans les données fonctionnelles que produit le satellite tout au long de sa vie. Un premier travail a été de comprendre comment mettre en évidence les anomalies grâce à des projections sur des bases de fonctions. En complément de cette revue des projections, nous avons appliqué plusieurs méthodes de détection d’anomalies, telles que la One-Class SVM et le Local Outlier Factor (LOF). En plus de ces deux méthodes, nous avons développé notre propre méthode pour prendre en compte la saisonnalité des courbes que nous considérons. En se basant sur cette étude, nous avons développé une nouvelle procédure pour sélectionner automatiquement les coefficients les plus intéressants pour la détection d’anomalies dans un cadre semi-supervisé. Notre méthode est une procédure de tests multiples où nous appliquons un test à deux échantillons à tous les niveaux de coefficients. Nous nous sommes également intéressés aux covariances des télémesures entre elles pour la détection d’anomalies. Pour cela, nous cherchons à comparer les covariances entre un groupe de télémesures pour deux journées, ou périodes consécutives. Nous avons appliqué trois tests sta- tistiques ayant des angles d’approche différents. Nous avons également développé dans ce but un nouveau test asymptotique. Outre la démonstration de la convergence de notre test, nous démontrons par des exemples que ce test est dans la pratique le plus puissant sur les données dont nous disposons. Dans cette thèse, nous avons abordé plusieurs aspects de la détection d’anomalies dans les données fonctionnelles issues des satellites. Pour chacune des méthodes, nous avons pu détecter toutes les anomalies, améliorant sensiblement le taux de fausses alarmes. / In this PhD, we have developed statistical methods to detect abnormal events in all the functional data produced by the satellite all through its lifecycle. The data we are dealing with come from two main phases in the satellite’s life, telemetries and test data. A first work on this thesis was to understand how to highlight the outliers thanks to projections onto functional bases. On these projections, we have also applied several outlier detection methods, such as the One-Class SVM, the Local Outlier Factor (LOF). In addition to these two methods, we have developed our own outlier detection method, by taking into account the seasonality of the data we consider. Based on this study, we have developed an original procedure to select automatically the most interesting coefficients in a semi-supervised framework for the outlier detection, from a given projection. Our method is a multiple testing procedure where we apply the two sample-test to all the levels of coefficients.We have also chosen to analyze the covariance matrices representing the covariance of the te- lemetries between themselves for the outlier detection in multivariate data. In this purpose, we are comparing the covariance of a cluster of several telemetries deriving from two consecutive days, or consecutive orbit periods. We have applied three statistical tests targeting this same issue with different approaches. We have also developed an original asymptotic test, inspired by both first tests. In addition to the proof of the convergence of this test, we demonstrate thanks to examples that this new test is the most powerful. In this PhD, we have tackled several aspects of the anomaly detection in the functional data deriving from satellites. For each of these methods, we have detected all the major anomalies, improving significantly the false discovery rate.

Esta dissertação faz um estudo da aplicação da técnica estatística denominada Análise de Dados Funcionais (ADF) à geração de descritores usados em reconhecimento de padrões, mais especificamente, no reconhecimento de objetos de interesse em imagens. Estes objetos podem ser representados por vetores de características, também chamados de assinaturas, obtidos por uma técnica chamada de Dimensão Fractal Multiescala (DFM). Ocorre que estes vetores apresentam alta dimensionalidade (número de elementos), fazendo-se assim necessário o uso de uma abordagem que reduza este número de valores, sem que haja uma grande perda da informação transmitida pela assinatura. Neste contexto, diversas técnicas de extração de um reduzido conjunto de descritores da assinatura são apresentadas pela literatura. Entre estas, as mais populares são Fourier e \\emph, ambas relativamente simples de se apresentar e com resultados satisfatórios. A proposta aqui apresentada é de se utilizar ADF em combinação com DFM na geração de descritores de padrões. Os resultados obtidos com o uso desta abordagem na geração de descritores demostraram que a técnica possibilita bons resultados, mesmo em situações em que não é possível o uso de muitos descritores. Os experimentos demostraram que ADF apresenta um bom potencial para aplicação neste tipo de problema, permitindo que o método de classificação alcance bons resultados mesmo com poucos descritores. São sugeridos trabalhos futuros em que ADF possa ser usada, pesquisando-se por métodos ainda mais eficazes. / This work studies the application of a statistical technique named Functional Data Analysis (FDA) for the generation of descriptors. These descriptors can be used for pattern recognition, more specifically, for the recognition of relevant objects in an image. These objects can be represented by features vectors, also known as signatures, obtained by a technique named Multi-scale Fractal Dimension (MFD). These vectors present a high dimensionality (number of elements), causing to be necessary the use of an approach for the reduction of this number of values, but without a large loss of information carried by the signature. In this context, several techniques for the extraction of a reduced set of signature descriptors are studied in the literature. Among these techniques, the most classic are Fourier and wavelets, both with simple presentation and providing satisfactory results. The proposal presented here is the use of FDA combined with MFD for the generation of pattern descriptors. The results obtained by the use of this approach for the generation of descriptors showed that this technique allows the obtention of good results, even in situations in wich is not possible the use of many descriptors. FDA was also applied to the extraction of descriptors of MFD texture signatures. Also in this case, the results were interesting. The experiments showed the FDA presents a good potential for the application to this type of problem, allowing the obtention of good results even by using a few descriptors. It is suggested future works in which FDA can be used, researching for still more efficient methods.

Lung cancer has been considered one of the leading causes of deaths while cancer re- mains the second most common cause of deaths in the USA. Understanding the behavior of a disease over time could yield important information to make decisions about the disease. Statistical models could provide crucial clues and help to make a decision about the dis- ease, budget allocation, evaluation, and implement prevention. Longitudinal trend analysis of the diseases helps to understand long term effects and nature. Cholesterol level is one of the most contributing risk factors for Coronary Heart Disease. Studying cholesterol statistically helps to know more about its nature and provides crucial information to mitigate its effectiveness in diagnosing its impact to public health. In our study, we have analyzed lung cancer mortality in the USA based on age at death, period at death, and birth cohort to investigate its nature in longitudinal effects. The attempt has been made to estimate mortality rate based on age for different age groups and to find the relative risk of mortality due to period effect and relative risk due to birth cohort for lung cancer in the United States. Our statistical analysis and modeling are based on the data obtained from Surveillance Epidemiology and End Results (SEER) program of the United States. We have also investigated the probabilistic behavior of average cholesterol level based on gender and ethnicity. The study reveals significant differences with respect to the distribution they follow and their basic inferences which could be beneficial to draw conclusions in various ways in addressing related issues. At the same time, the change of cholesterol level over time for an individual might be a good source to study the association of cholesterol level, coronary heart disease and their effects on age. The cholesterol data is obtained from inter-university Consortium for Political and Social Research and National Health and Nutrition Examination Survey (NHANS) of the United States. Understanding the average change in total serum cholesterol level over time as people get older could be vital to explore it. We have studied the longitudinal behavior of the association of sex and time with cholesterol level. It is observed that age, sex, and time have an individual effect and can impact differently upon collective considerations. Their adverse effect in increasing cholesterol level could promote to worsen the cholesterol re- lated issues and hence heart related diseases. We believe our study pivots knowing more about target population of cholesterol level and helps to have the useful inference about cholesterol levels for public health. Finally, we also analyzed the average cholesterol data using a functional data analysis approach to understand its nature and effect on age. Since functional data analysis approach presents more flexibility in modeling, it could provide more insight in studying cholesterol level.

Dans cette thèse, nous étudions des tests du type : (H0) : E [U | X] = 0 p.s. contre (H1) : P {E [U | X] = 0} < 1 où U est le résidu de la modélisation d'une variable Y en fonction de X. Dans ce cadre et pour plusieurs cas particuliers – significativité de variables, régression quantile, données fonctionnelles, modèle single-index –, nous proposons une statistique de test permettant d'obtenir des valeurs critiques issues d'une loi asymptotique pivotale. Dans chaque cas, nous donnons également une méthode de bootstrap appropriée pour les échantillons de petite taille. Nous montrons la consistance envers des alternatives locales – ou à la Pitman – des tests proposés, lorsque ce type d'alternative ne tend pas trop vite vers l'hypothèse nulle. À chaque fois, nous vérifions à partir de simulations sous l'hypothèse nulle et sous une séquence d'hypothèses alternatives que les résultats théoriques sont en accord avec la pratique. / In this thesis, we study test statistics of the form : (H0) : E [U | X] = 0 p.s. contre (H1) : P {E [U | X] = 0} < 1 where U is the residual of some Y modeling with respect to covariates X. In this setup and for several particular cases – significance, quantile regression, functional data, single-index model –, we introduce test statistics that have pivotal asymptotic critical values. For each case, we also give a suitable bootstrap procedure for small samples. We prove the consistency against local – or Pitman – alternatives for the proposed test statistics, when such an alternative does not get close to the null hypothesis too fast. Simulation studies are used to check the effectiveness of the theoretical results in applications.

The aim of this thesis is to analyze two major indicators representing the processes of natural changes in population, which are not influenced by age structure -- age specific fertility rate and age specific mortality rate and to introduce functional data analysis, which has recently become increasingly applied in many scientific fields. The functional data analysis is introduced on the Czech demographic data set for more than last sixty years. The thesis briefly describes historical development of mortality and fertility and also describes selected methods of functional data analysis by using statistical program R. Functional data analysis provides a wide variety of graphics tools to explore the data which we called functional and help us to explore all typical and atypical features, identify outliers etc.

[ES] En esta Tesis Doctoral se plantea un nuevo modelado biomecánico del cuello para su aplicación en los campos de la valoración funcional y la ergonomía. Se ha realizado un estudio cinemático para describir los axoides asociados a los ejes instantáneos de rotación (EIR) del movimiento de la cabeza respecto del tórax, así como las variables cinemáticas (posiciones, velocidades y aceleraciones), que se describen como funciones continuas mediante técnicas de Análisis de Datos Funcionales (FDA). Se han analizado movimientos cíclicos continuos. El movimiento más reproducible es el de flexo-extensión. Para el análisis dinámico, se ha elaborado un modelo articular de dinámica inversa, que permite estimar las fuerzas, momentos a nivel de C7 y potencia desarrollada, a partir de la cinemática del movimiento y de las características inerciales del sistema cuello-cabeza, obtenidas a partir de estudios previos y ajustadas mediante un procedimiento de calibración. El modelo ha sido validado experimentalmente y se ha cuantificado su fiabilidad, que resulta suficiente para aplicaciones clínicas. Se han estudiado una muestra de sujetos sanos (n=45) con la finalidad de obtener una base de normalidad, analizar el efecto de características individuales (edad, sexo y características antropométricas), y una muestra de pacientes con dolor inespecífico de cuello (n=24), para analizar las diferencias asociados a la patología. Este es el primer estudio en el que se obtiene una representación continua del EIR y se realiza una descripción de la cinemática y dinámica usando funciones continuas mediante técnicas estadísticas de FDA. Los resultados muestran que los patrones cinemáticos y dinámicos están influidos por el sexo y por la longitud del cuello. Las mujeres presentan más movilidad que los hombres y una mayor longitud del cuello va asociada a movimientos más lentos. El efecto de la edad no es demasiado acusado en la muestra analizada, con edades inferiores a 50 años en la mayoría de casos. Las diferencias cinemáticas entre personas sanas y con dolor de cuello son muy claras, siendo mayores en las variables de posición y de aceleración que en los rangos, por lo que son éstas variables más adecuadas para definir protocolos de valoración biomecánica clínica. Desde el punto de vista dinámico, hay diferencias muy claras en las fuerzas y en la potencia desarrollada, aunque no tanto en el momento articular. Las diferencias en las variables funcionales son mucho más evidentes que en las correspondientes variables numéricas extraídas de éstas (rangos, valores máximos y mínimos). En definitiva, se demuestra que el registro de movimientos continuo y el uso conjunto de información cinemática y dinámica ofrece una visión más completa de los patrones biomecánicos de movimiento del cuello y puede mejorar los actuales sistemas de valoración funcional. / [CA] En aquesta tesi doctoral es planteja un nou modelatge biomecànic de el coll per la seva aplicació en els camps de la valoració funcional i l'ergonomia. S'ha realitzat un estudi cinemàtic per descriure els axoides associats als eixos instantanis de rotació (EIR) de el moviment del cap respecte de tòrax, així com les variables cinemàtiques (posicions, velocitats i acceleracions), que es descriuen com a funcions contínues mitjançant tècniques d'Anàlisi de Dades Funcionals (FDA). S'han analitzat moviments cíclics continus. El moviment més reproduïble és el de flexo-extensió. Per a l'anàlisi dinàmic, s'ha elaborat un model articular de dinàmica inversa, que permet estimar les forces, moments a nivell de C7 i potència desenvolupada, a partir de la cinemàtica de el moviment i de les característiques inercials de sistema coll-cap, obtingudes a partir d'estudis previs i ajustades mitjançant un procediment de calibratge. El model ha estat validat experimentalment i s'ha quantificat la seva fiabilitat, que resulta suficient per a aplicacions clíniques. S'han estudiat una mostra de subjectes sans (n = 45) amb la finalitat d'obtenir una base de normalitat, analitzar l'efecte de característiques individuals (edat, sexe i característiques antropomètriques), i una mostra de pacients amb dolor inespecífic de coll (n = 24), per analitzar les diferències associats a la patologia. Aquest és el primer estudi en el qual s'obté una representació contínua de l'AIR i es realitza una descripció de la cinemàtica i dinàmica usant funcions contínues mitjançant tècniques estadístiques de FDA. Els resultats mostren que els patrons cinemàtics i dinàmics estan influïts pel sexe i per la longitud de coll. Les dones presenten més mobilitat que els homes i una major longitud de coll va associada a moviments més lents. L'efecte de l'edat no és massa acusat en la mostra analitzada, amb edats inferiors a 50 anys en la majoria de casos. Les diferències cinemàtiques entre persones sanes i amb mal de coll són molt clares, sent majors en les variables de posició i d'acceleració que en els rangs, de manera que són aquestes variables més adequades per definir protocols de valoració biomecànica clínica. Des del punt de vista dinàmic, hi ha diferències molt clares en les forces i en la potència desenvolupada, encara que no tant en el moment articular. Les diferències en les variables funcionals són molt més evidents que en les corresponents variables numèriques extretes d'aquestes (rangs, valors màxims i mínims). En definitiva, es demostra que el registre de moviments continu i l'ús conjunt d'informació cinemàtica i dinàmica ofereix una visió més completa dels patrons biomecànics de moviment de coll i pot millorar els actuals sistemes de valoració funcional. / [EN] In this Doctoral Thesis, a new biomechanical modeling of the neck is proposed for its application in the fields of functional assessment and ergonomics. A kinematic study has been carried out to describe the axoids associated with the instantaneous axes of rotation (EIR) of the movement of the head with respect to the thorax, as well as the kinematic variables (positions, velocities and accelerations), which are described as continuous functions using techniques of Functional Data Analysis (FDA). Continuous cyclical movements have been analyzed. The most reproducible movement is flexion-extension. For the dynamic analysis, a joint model of inverse dynamics has been developed, which allows estimating the forces, moments at the C7 level and developed power, from the kinematics of the movement and the inertial characteristics of the neck-head system, obtained from starting from previous studies and adjusted by means of a calibration procedure. The model has been experimentally validated and its reliability has been quantified, which is sufficient for clinical applications. A sample of healthy subjects (n = 45) were studied in order to obtain a normality base, analyze the effect of individual characteristics (age, sex and anthropometric characteristics), and a sample of patients with nonspecific neck pain (n = 24), to analyze the differences associated with the pathology. This is the first study in which a continuous representation of the EIR is obtained and a description of the kinematics and dynamics is made using continuous functions using FDA statistical techniques. The results show that the kinematic and dynamic patterns are influenced by gender and neck length. Women are more mobile than men and longer neck length is associated with slower movements. The effect of age is not too pronounced in the analyzed sample, with ages less than 50 years in most cases. The kinematic differences between healthy people and those with neck pain are very clear, being greater in the position and acceleration variables than in the ranges, making these variables more suitable for defining clinical biomechanical assessment protocols. From the dynamic point of view, there are very clear differences in the forces and the power developed, although not so much in the articulation moment. The differences in the functional variables are much more evident than in the corresponding numerical variables extracted from them (ranges, maximum and minimum values). In short, it has been shown that continuous movement recording and the joint use of kinematic and dynamic information offers a more complete view of the biomechanical patterns of neck movement and can improve current functional assessment systems. / Este trabajo ha sido desarrollado en el Instituto Universitario de Ingeniería Mecánica y Biomecánica de la Universitat Politècnica de València, y ha sido financiado con el proyecto de investigación Junior PIJ-15-08 de la Escuela Politécnica Nacional de Quito-Ecuador. / Venegas Toro, WR. (2021). Modelado biomecánico del cuello basado en la imagen cinemática de la función articular para su aplicación en tecnologías para la salud y el bienestar del ser humano [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/174522 / TESIS

La classification et le regroupement des données fonctionnelles longitudinales ont fait beaucoup de progrès dans les dernières années. Plusieurs méthodes ont été proposées et ont démontré des résultats prometteurs. Pour ce mémoire, on a comparé le comportement des algorithmes de partitionnement sur un ensemble de données décrivant les trajectoires de voitures dans une intersection de Montréal. La motivation est qu’il est coûteux et long de faire la classification manuellement et on démontre dans cet ouvrage qu’il est possible d’obtenir des prédictions adéquates avec les différents algorithmes. Parmi les méthodes utilisées, la méthode distclust utilise l’approche des K-moyennes avec une notion de distance entre les courbes fonctionnelles. On utilise aussi une classification par mélange de densité gaussienne, mclust. Ces deux approches n’étant pas conçues uniquement pour le problème de classification fonctionnelle, on a donc également appliqué des méthodes fonctionnelles spécifiques au problème : fitfclust, funmbclust, funclust et funHDDC. On démontre que les résultats du partitionnement et de la prédiction obtenus par ces approches sont comparables à ceux obtenus par ceux basés sur la distance. Les méthodes fonctionnelles sont préférables, car elles permettent d’utiliser des critères de sélection objectifs comme le AIC et le BIC. On peut donc éviter d’utiliser une partition préétablie pour valider la qualité des algorithmes, et ainsi laisser les données parler d’elles-mêmes. Finalement, on obtient des estimations détaillées de la structure fonctionnelle des courbes, comme sur l’impact de la réduction de données avec une analyse en composantes principales fonctionnelles multivariées. / The study of the clustering of functional data has made a lot of progress in the last couple of years. Multiple methods have been proposed and the respective analysis has shown their eÿciency with some benchmark studies. The objective of this Master’s thesis is to compare those clustering algorithms with datasets from traÿc at an intersection of Montreal. The idea behind this is that the manual classification of these data sets is time-consuming. We show that it is possible to obtain adequate clustering and prediction results with several algorithms. One of the methods that we discussed is distclust : a distance-based algorithm that uses a K-means approach. We will also use a Gaussian mixture density clustering method known as mclust. Although those two techniques are quite e˙ective, they are multi-purpose clustering methods, therefore not tailored to the functional case. With that in mind, we apply four functional clustering methods : fitfclust, funmbclust, funclust, and funHDDC. Our results show that there is no loss in the quality of the clustering between the afore-mentioned functional methods and the multi-purpose ones. We prefer to use the functional ones because they provide a detailed estimation of the functional structure of the trajectory curves. One notable detail is the impact of a dimension reduction done with multivari-ate functional principal components analysis. Furthermore, we can use objective selection criteria such as the AIC and the BIC, and avoid using cluster quality indices that use a pre-existing classification of the data.

An vielen verschiedenen Stellen der angewandten Statistik sind die zu untersuchenden Objekte abhängig von stetigen Parametern. Typische Beispiele in Finanzmarktapplikationen sind implizierte Volatilitäten, risikoneutrale Dichten oder Zinskurven. Aufgrund der Marktkonventionen sowie weiteren technisch bedingten Gründen sind diese Objekte nur an diskreten Punkten, wie zum Beispiel an Ausübungspreise und Maturitäten, für die ein Geschäft in einem bestimmten Zeitraum abgeschlossen wurde, beobachtbar. Ein funktionaler Datensatz ist dann vorhanden, wenn diese Funktionen für verschiedene Zeitpunkte (z.B. Tage) oder verschiedene zugrundeliegende Aktiva gesammelt werden. Das erste Thema, das in dieser Dissertation betrachtet wird, behandelt die nichtparametrischen Methoden der Schätzung dieser Objekte (wie z.B. implizierte Volatilitäten) aus den beobachteten Daten. Neben den bekannten Glättungsmethoden wird eine Prozedur für die Glättung der implizierten Volatilitäten vorgeschlagen, die auf einer Kombination von nichtparametrischer Glättung und den Ergebnissen der arbitragefreien Theorie basiert. Der zweite Teil der Dissertation ist der funktionalen Datenanalyse (FDA), speziell im Zusammenhang mit den Problemen, der empirischen Finanzmarktanalyse gewidmet. Der theoretische Teil der Arbeit konzentriert sich auf die funktionale Hauptkomponentenanalyse -- das funktionale Ebenbild der bekannten Dimensionsreduktionstechnik. Ein umfangreicher überblick der existierenden Methoden wird gegeben, eine Schätzmethode, die von der Lösung des dualen Problems motiviert ist und die Zwei-Stichproben-Inferenz basierend auf der funktionalen Hauptkomponentenanalyse werden behandelt. Die FDA-Techniken sind auf die Analyse der implizierten Volatilitäten- und Zinskurvendynamik angewandt worden. Darüber hinaus, wird die Implementation der FDA-Techniken zusammen mit einer FDA-Bibliothek für die statistische Software Xplore behandelt. / In many different fields of applied statistics an object of interest is depending on some continuous parameter. Typical examples in finance are implied volatility functions, yield curves or risk-neutral densities. Due to the different market conventions and further technical reasons, these objects are observable only on a discrete grid, e.g. for a grid of strikes and maturities for which the trade has been settled at a given time-point. By collecting these functions for several time points (e.g. days) or for different underlyings, a bunch (sample) of functions is obtained - a functional data set. The first topic considered in this thesis concerns the strategies of recovering the functional objects (e.g. implied volatilities function) from the observed data based on the nonparametric smoothing methods. Besides the standard smoothing methods, a procedure based on a combination of nonparametric smoothing and the no-arbitrage-theory results is proposed for implied volatility smoothing. The second part of the thesis is devoted to the functional data analysis (FDA) and its connection to the problems present in the empirical analysis of the financial markets. The theoretical part of the thesis focuses on the functional principal components analysis -- functional counterpart of the well known multivariate dimension-reduction-technique. A comprehensive overview of the existing methods is given, an estimation method based on the dual problem as well as the two-sample inference based on the functional principal component analysis are discussed. The FDA techniques are applied to the analysis of the implied volatility and yield curve dynamics. In addition, the implementation of the FDA techniques together with a FDA library for the statistical environment XploRe are presented.

The content of this dissertation is divided into two main topics: 1) nonlinear profile monitoring and 2) an improved approximate distribution for the T² statistic based on the successive differences covariance matrix estimator. Part 1: Nonlinear Profile Monitoring In an increasing number of cases the quality of a product or process cannot adequately be represented by the distribution of a univariate quality variable or the multivariate distribution of a vector of quality variables. Rather, a series of measurements are taken across some continuum, such as time or space, to create a profile. The profile determines the product quality at that sampling period. We propose Phase I methods to analyze profiles in a baseline dataset where the profiles can be modeled through either a parametric nonlinear regression function or a nonparametric regression function. We illustrate our methods using data from Walker and Wright (2002) and from dose-response data from DuPont Crop Protection. Part 2: Approximate Distribution of T² Although the T² statistic based on the successive differences estimator has been shown to be effective in detecting a shift in the mean vector (Sullivan and Woodall (1996) and Vargas (2003)), the exact distribution of this statistic is unknown. An accurate upper control limit (UCL) for the T² chart based on this statistic depends on knowing its distribution. Two approximate distributions have been proposed in the literature. We demonstrate the inadequacy of these two approximations and derive useful properties of this statistic. We give an improved approximate distribution and recommendations for its use. / Ph. D.

In-service pavements require maintenance and rehabilitation (MandR) interventions to keep them in compliance with structural and functional standards. With the increased focus on the sustainability of our roadway systems, it has become important to document the cost and environmental impacts of different MandR strategies over the life cycle of the pavement to facilitate project selection decisions in the future. Asphalt pavement recycling, while cost-effective and environmentally friendly compared to other traditional MandR treatments, still faces some widespread implementation push-back, leading to policy enactments by the FHWA aimed at encouraging the use of recycling in road projects. Many agencies and contractors have cited the lack of project selection criteria, and uncertainty about long-term performance of these recycling alternatives as reasons impeding rapid implementation of these treatments in road projects. One of the gray areas of the FHWA's 2015 Recycled Material Policy in project selection was, until recently, the lack of guidelines or tools for the assessment of the environmental suitability of candidate MandR treatments. Today, it is almost impossible to evaluate the environmental suitability of various recycling-based end-of-service-life treatments because available databases do not have relevant information on the details of unit processes, construction equipment and activities, and use-stage roughness data. Development of future MandR plans throughout the service life of pavements rehabilitated with recycling-based treatments is somewhat limited as deterioration is not fully understood. Also, available modeling tools no not address all LCA phases, or in cases where they do, key life cycle phases including the MandR, and use phases are not well covered due to the lack of quantification highlighted earlier. To address the highlighted concerns, this dissertation developed a user-friendly comprehensive LCA tool that was further validated with a case study to quantify the service life (when the pavement has reached a critical threshold performance value) and potential environmental benefits of pavement recycling projects executed by the Virginia Department of Transportation over the past decade. The tool, pySuPave, includes an excel spreadsheet user-inputs interface, and database of economic flows for unit processes used in the production of pavement materials and subsequent construction of the pavement system, considering transportation of materials and construction machinery to plants and construction site. A python-based program was used to perform matrix-based computations to generate the environmental burdens from the available public LCA Ecoinvent database. A substantive part of the dissertation was dedicated to evaluating the performance of in-service pavements rehabilitated with cold recycling and full-depth reclamation treatments, focusing on developing pavement performance prediction models (PPPM) that goes on to improve modelling of the MandR and use stages in the pavement LCA and ultimately bridges the knowledge gap on how these treatments perform in the long term. This part of the dissertation was presented in two chapters; trends in pavement recycling and performance data collection, and development of PPPMs for recycled asphalt pavements. The first provides an update and examines the current state of pavement recycling techniques, highlighting trends in the various recycling methods, examining what is and is not working from the agency perspective, and assessing the progress made in the last decade through a web-based survey. The survey results did not indicate significant changes in the adoption of the asphalt pavement recycling concept in the last decade. However, recycling techniques, such as hot in-place recycling, are being used less and more agencies seem to be adopting lower temperature techniques such as cold in-place recycling, cold central plant recycling and full depth reclamation. Improvements in mix design methods were noticeable, as more agencies have adopted contemporary methods, such as the Superpave design. Among states, very few agencies collected performance data for completed asphalt pavement recycling projects. The second chapter on performance focused on developing individual and family-type PPPMs from the data collected from the states of Virginia and Colorado, respectively. While regression modeling forms the backbone of the approach used, the chapter also presents an approach to developing family-type models using functional data analysis to find groups of projects with similar deterioration trends. In the case of Colorado, cold in-place recycling (CIR) projects completed with an initial IRI between 71 and 91 in/mi are most likely to deteriorate at an average group rate of 1.37 in/mi/year. Similarly, full depth reclamation (FDR) projects will most likely deteriorate following an average group rate of 1.40 in/mi/yr, with an initial IRI between 52 and 70 in/mi. These projects will stay in service well over 30 years if a threshold IRI of 140 in/mi were used a failure criterion. For the individual roughness models developed for VDOT, the initial IRI values and the rate of change for the treatments analyzed were found to range between 48 and 85 in/mi and between 0.70 and 5.20 in/mi/year, respectively, depending on the recycling method and type of stabilization treatment. Finally, a context-based life cycle assessment case study was conducted to benchmark and compare the environmental impacts associated with rehabilitating a low-volume road with various recycled-based and equivalent conventional methods. Several impact indicators were assessed but only the global warming (GW) score and the single score index that combines all the environmental impact indicators into a single number using normalization and weighting factors were reported in this study for the sake of brevity. Four restorative maintenance projects including two CIR (4-in. HMA over a 5-in. CIR with foamed asphalt and emulsion stabilization), one cold central plant recycling (CCPR): 4-in. HMA over a 5-in. foamed asphalt CCPR (CCPR FA), and one non-recycling structural overlay (8-in. HMA over an existing pavement) were evaluated. In addition, the following reconstruction projects were assessed; two FDR (4-in. HMA over a 12-in. FDR with foamed asphalt with 1% cement additive, and a 4-in. HMA over 10.5-in. cement stabilized FDR), and a non-recycling reconstruction project (a new reconstruction project with 8-in. HMA over a 16-in. aggregate base and subbase). The functional unit was a two lane-mile length, 12 feet wide project with a traffic volume of 1000 vehicles (3% trucks) and the analysis was conducted for 50 years. The GW score and a few other impact indicators showed an increase in the observed results where cement is used as a main stabilizer or as an additive. Between the asphalt stabilized projects, the difference in impact scores is only seen when cement is used as an additive as highlighted in the case of foamed asphalt applications. Even for the low-volume road under study, the use stage contributes the largest share to global warming and is—among several factors—attributed to the initial surface roughness of completed projects. Thus, for state DOTs looking to reduce the environmental footprints for road infrastructure projects and achieve federal legislative goals, building smoother roads and taking steps to keep the annual deterioration rate low would be an important measure, in addition to pavement recycling. Comparing the projects based on the overall single score derived from weighting factors from the National Institute of Standards and Technology (NIST) ranks the projects as follows (listed in order decreasing impacts per rehabilitation category); restorative maintenance projects: T. OVERLAY (non-recycling structural overlay—8 in. HMA over an existing pavement) - 1.06 pts, CCPR FA (4 in. HMA over a 5 in. cold central plant recycling with foamed asphalt) - 1.02 pts, CIR FA (4 in. HMA over a 5 in. cold in-place recycling with foamed asphalt) - 1.00 pts, CIR AE (4 in. HMA over a 5 in. cold in-place recycling with emulsion)- 0.86 pts; reconstruction projects: RECONS (a new reconstruction project—8 in. HMA over a 16 in. aggregate base and subbase) -1.42 pts, FDR FA+C (4 in. HMA over a 12 in. FDR with foamed asphalt with 1% cement additive) - 1.15 pts, FDR C (4 in. HMA over 10.5 in. cement stabilized FDR) - 1.02 pts. / Doctor of Philosophy / Due to harsh environmental conditions and continual damage from moving traffic, highway pavements or roadways deteriorate and grow weak over time. Throughout their life in service, different maintenance and rehabilitation (MandR) activities are performed with the intention of slowing down the deterioration to always keep the highway at a certain level of service to road users. For a long time, these MandR activities have included the use of virgin materials in techniques ranging from minor treatment applications such as fog seals, chip seals, thin overlays through more heavy treatments such as mill and fills, thicker overlays all the way to total reconstruction. Other MandR alternatives include pavement recycling which reuses materials from the existing distressed roadways either in-place or at a nearby mobile plant have gained popularity among several state highway agencies over the last decade. The advantages of using the recycling alternatives compared to non-recycling options are many and have been known to include cost savings, less construction time, and low environmental footprint. Many highway agencies, however, have expressed the lack of information on project selection criteria and the uncertainty about long-term performance of these recycling alternatives as reasons impeding rapid and widespread implementation in road projects. Agencies need selection criteria to help them identify the right treatments to apply to the right road at the right time. In a bid to encourage the use of pavement recycling treatments, the Federal Highway Administration (FHWA) enacted the Recycled Materials Policy in 2006 (revised 2015) but the policy did not fully address certain aspects of project selection. Directives on assessing the environmental suitability of recycling projects, for instance, was not given. There are no tools with modern databases incorporating the various unit processes for pavement recycling to aid agencies carry out this environmental assessment. To address the highlighted concerns, we developed a user-friendly comprehensive environmental assessment tool called pySuPave as part of this dissertation. We later validated the tool with a case study to quantify the potential environmental benefits of pavement recycling projects executed by the Virginia Department of Transportation over the past decade. Next, we conducted a survey of the departments of transportation (DOT) around the United States and Canada to collect performance data from agencies with active in-place recycling programs. Approximately 18% of the DOTs surveyed were able to provide performance data. Data received from Colorado and Virginia were subsequently used to developed models to predict deterioration in recycled pavements. In the case of Colorado, CIR projects completed with an initial roughness (IRI) between 71 and 91 in/mi are most likely to deteriorate at a rate of 1.37 in/mi/year. Similarly, FDR projects will most likely deteriorate following an average group rate of 1.40 in/mi/yr, with an initial IRI between 52 and 70 in/mi. These projects will stay in service well over 30 years if a threshold IRI of 140 in/mi were used a failure criterion. For the individual roughness models developed for VDOT, the initial IRI values and the rate of change for the treatments analyzed were found to range between 48 and 85 in/mi and between 0.70 and 5.20 in/mi/year, respectively, depending on the recycling method and type of stabilization treatment Finally, we conducted an environmental assessment case study to benchmark and compare the environmental burdens i.e., global warming (GW) and other impacts associated with rehabilitating a low-volume road with various recycled-based and equivalent non-recycling methods. Four restorative maintenance projects including two CIR (4-in. HMA over a 5-in. CIR with foamed asphalt and emulsion stabilization), one CCPR (4-in. HMA over a 5-in. foamed asphalt CCPR [CCPR FA]), and one non-recycling structural overlay (8-in. HMA over an existing pavement) were evaluate. In addition, the following reconstruction projects were assessed; two FDR (4-in. HMA over a 12-in. FDR with foamed asphalt with 1% cement additive, and a 4-in. HMA over 10.5-in. cement stabilized FDR), and a non-recycling reconstruction project (a new reconstruction project with 8-in. HMA over a 16-in. aggregate base and subbase). The functional unit was a two lane-mile length, 12 feet wide project with a traffic volume of 1000 vehicles (3% trucks) and the analysis was conducted for 50 years. The study results showed that the recycling-based projects had lower overall environmental burdens compared to their equivalent non-recycling alternatives. The GW score and a few other environmental impact indicators were higher when cement is used as a main stabilizer or as an additive in the recycling projects. Between the asphalt stabilized recycling projects, the difference in impact scores is only seen when cement is used as an additive as emphasized in the case of foamed asphalt applications. Even for the low-volume roads under study, the use stage (when the project is open to road-users) in the pavement life cycle contributes the largest share to global warming and is—among several factors—attributed to the initial surface roughness of completed projects. Thus, for state DOTs looking to reduce the environmental footprints for road infrastructure projects and achieve federal legislative goals, building smoother roads and taking steps to keep the annual deterioration rate low would be an important measure, in addition to pavement recycling. The results from this research support the hypothesis that pavement recycling can reduce global warming and other environmental burdens compared to non-recycling methods. Therefore, agencies should encourage more pavement recycling programs.