Modelling uncertainty of flood quantile estimation at ungauged sites by Bayesian networks

Prediction at ungauged sites is essential for water resources planning and management. Ungauged sites have no observations about the magnitude of floods, but some site and basin characteristics are known. Regression models relate physiographic and climatic basin characteristics to flood quantiles, which can be estimated from observed data at gauged sites. However, these models assume linear relationships between variables Prediction intervals are estimated by the variance of the residuals in the estimated model. Furthermore, the effect of the uncertainties in the explanatory variables on the dependent variable cannot be assessed. This paper presents a methodology to propagate the uncertainties that arise in the process of predicting flood quantiles at ungauged basins by a regression model. In addition, Bayesian networks were explored as a feasible tool for predicting flood quantiles at ungauged sites. Bayesian networks benefit from taking into account uncertainties thanks to their probabilistic nature. They are able to capture non-linear relationships between variables and they give a probability distribution of discharges as result. The methodology was applied to a case study in the Tagus basin in Spain.

Propuesta metodológica para el análisis y protección de la calidad del agua subterránea de acuerdo con los requerimientos de la Directiva Marco y la Directiva derivada de Aguas Subterráneas

La protección de las aguas subterráneas es una prioridad de la política medioambiental de la UE. Por ello ha establecido un marco de prevención y control de la contaminación, que incluye provisiones para evaluar el estado químico de las aguas y reducir la presencia de contaminantes en ellas. Las herramientas fundamentales para el desarrollo de dichas políticas son la Directiva Marco del Agua y la Directiva Hija de Aguas Subterráneas. Según ellas, las aguas se consideran en buen estado químico si: • la concentración medida o prevista de nitratos no supera los 50 mg/l y la de ingredientes activos de plaguicidas, de sus metabolitos y de los productos de reacción no supera el 0,1 μg/l (0,5 μg/l para el total de los plaguicidas medidos) • la concentración de determinadas sustancias de riesgo es inferior al valor umbral fijado por los Estados miembros; se trata, como mínimo, del amonio, arsénico, cadmio, cloruro, plomo, mercurio, sulfatos, tricloroetileno y tetracloroetileno • la concentración de cualquier otro contaminante se ajusta a la definición de buen estado químico enunciada en el anexo V de la Directiva marco sobre la política de aguas • en caso de superarse el valor correspondiente a una norma de calidad o a un valor umbral, una investigación confirma, entre otros puntos, la falta de riesgo significativo para el medio ambiente. Analizar el comportamiento estadístico de los datos procedentes de la red de seguimiento y control puede resultar considerablemente complejo, debido al sesgo positivo que suelen presentar dichos datos y a su distribución asimétrica, debido a la existencia de valores anómalos y diferentes tipos de suelos y mezclas de contaminantes. Además, la distribución de determinados componentes en el agua subterránea puede presentar concentraciones por debajo del límite de detección o no ser estacionaria debida a la existencia de tendencias lineales o estacionales. En el primer caso es necesario realizar estimaciones de esos valores desconocidos, mediante procedimientos que varían en función del porcentaje de valores por debajo del límite de detección y el número de límites de detección aplicables. En el segundo caso es necesario eliminar las tendencias de forma previa a la realización de contrastes de hipótesis sobre los residuos. Con esta tesis se ha pretendido establecer las bases estadísticas para el análisis riguroso de los datos de las redes de calidad con objeto de realizar la evaluación del estado químico de las masas de agua subterránea para la determinación de tendencias al aumento en la concentración de contaminantes y para la detección de empeoramientos significativos, tanto en los casos que se ha fijado un estándar de calidad por el organismo medioambiental competente como en aquéllos que no ha sido así. Para diseñar una metodología que permita contemplar la variedad de casos existentes, se han analizado los datos de la Red Oficial de Seguimiento y Control del Estado Químico de las Aguas Subterráneas del Ministerio de Agricultura, Alimentación y Medio Ambiente (Magrama). A continuación, y dado que los Planes Hidrológicos de Cuenca son la herramienta básica de las Directivas, se ha seleccionado la Cuenca del Júcar, dada su designación como cuenca piloto en la estrategia de implementación común (CIS) de la Comisión Europea. El objetivo principal de los grupos de trabajo creados para ello se dirigió a implementar la Directiva Derivada de Agua Subterráneas y los elementos de la Directiva Marco del Agua relacionadas, en especial la toma de datos en los puntos de control y la preparación del primer Plan de Gestión de Cuencas Hidrográficas. Dada la extensión de la zona y con objeto de analizar una masa de agua subterránea (definida como la unidad de gestión en las Directivas), se ha seleccionado una zona piloto (Plana de Vinaroz Peñiscola) en la que se han aplicado los procedimientos desarrollados con objeto de determinar el estado químico de dicha masa. Los datos examinados no contienen en general valores de concentración de contaminantes asociados a fuentes puntuales, por lo que para la realización del estudio se han seleccionado valores de concentración de los datos más comunes, es decir, nitratos y cloruros. La estrategia diseñada combina el análisis de tendencias con la elaboración de intervalos de confianza cuando existe un estándar de calidad e intervalos de predicción cuando no existe o se ha superado dicho estándar. De forma análoga se ha procedido en el caso de los valores por debajo del límite de detección, tomando los valores disponibles en la zona piloto de la Plana de Sagunto y simulando diferentes grados de censura con objeto de comparar los resultados obtenidos con los intervalos producidos de los datos reales y verificar de esta forma la eficacia del método. El resultado final es una metodología general que integra los casos existentes y permite definir el estado químico de una masa de agua subterránea, verificar la existencia de impactos significativos en la calidad del agua subterránea y evaluar la efectividad de los planes de medidas adoptados en el marco del Plan Hidrológico de Cuenca. ABSTRACT Groundwater protection is a priority of the EU environmental policy. As a result, it has established a framework for prevention and control of pollution, which includes provisions for assessing the chemical status of waters and reducing the presence of contaminants in it. The measures include: • criteria for assessing the chemical status of groundwater bodies • criteria for identifying significant upward trends and sustained concentrations of contaminants and define starting points for reversal of such trends • preventing and limiting indirect discharges of pollutants as a result of percolation through soil or subsoil. The basic tools for the development of such policies are the Water Framework Directive and Groundwater Daughter Directive. According to them, the groundwater bodies are considered in good status if: • measured or predicted concentration of nitrate does not exceed 50 mg / l and the active ingredients of pesticides, their metabolites and reaction products do not exceed 0.1 mg / l (0.5 mg / l for total of pesticides measured) • the concentration of certain hazardous substances is below the threshold set by the Member States concerned, at least, of ammonium, arsenic, cadmium, chloride, lead, mercury, sulphates, trichloroethylene and tetrachlorethylene • the concentration of other contaminants fits the definition of good chemical status set out in Annex V of the Framework Directive on water policy • If the value corresponding to a quality standard or a threshold value is exceeded, an investigation confirms, among other things, the lack of significant risk to the environment. Analyzing the statistical behaviour of the data from the monitoring networks may be considerably complex due to the positive bias which often presents such information and its asymmetrical distribution, due to the existence of outliers and different soil types and mixtures of pollutants. Furthermore, the distribution of certain components in groundwater may have concentrations below the detection limit or may not be stationary due to the existence of linear or seasonal trends. In the first case it is necessary to estimate these unknown values, through procedures that vary according to the percentage of values below the limit of detection and the number of applicable limits of detection. In the second case removing trends is needed before conducting hypothesis tests on residuals. This PhD thesis has intended to establish the statistical basis for the rigorous analysis of data quality networks in order to conduct the evaluation of the chemical status of groundwater bodies for determining upward and sustained trends in pollutant concentrations and for the detection of significant deterioration in cases in which an environmental standard has been set by the relevant environmental agency and those that have not. Aiming to design a comprehensive methodology to include the whole range of cases, data from the Groundwater Official Monitoring and Control Network of the Ministry of Agriculture, Food and Environment (Magrama) have been analysed. Then, since River Basin Management Plans are the basic tool of the Directives, the Júcar river Basin has been selected. The main reason is its designation as a pilot basin in the common implementation strategy (CIS) of the European Commission. The main objective of the ad hoc working groups is to implement the Daughter Ground Water Directive and elements of the Water Framework Directive related to groundwater, especially the data collection at control stations and the preparation of the first River Basin Management Plan. Given the size of the area and in order to analyze a groundwater body (defined as the management unit in the Directives), Plana de Vinaroz Peñíscola has been selected as pilot area. Procedures developed to determine the chemical status of that body have been then applied. The data examined do not generally contain pollutant concentration values associated with point sources, so for the study concentration values of the most common data, i.e., nitrates and chlorides have been selected. The designed strategy combines trend analysis with the development of confidence intervals when there is a standard of quality and prediction intervals when there is not or the standard has been exceeded. Similarly we have proceeded in the case of values below the detection limit, taking the available values in Plana de Sagunto pilot area and simulating different degrees of censoring in order to compare the results obtained with the intervals achieved from the actual data and verify in this way the effectiveness of the method. The end result is a general methodology that integrates existing cases to define the chemical status of a groundwater body, verify the existence of significant impacts on groundwater quality and evaluate the effectiveness of the action plans adopted in the framework of the River Basin Management Plan.

Development of safety performance functions for empirical Bayes estimation of crash reduction factors

Crash reduction factors (CRFs) are used to estimate the potential number of traffic crashes expected to be prevented from investment in safety improvement projects. The method used to develop CRFs in Florida has been based on the commonly used before-and-after approach. This approach suffers from a widely recognized problem known as regression-to-the-mean (RTM). The Empirical Bayes (EB) method has been introduced as a means to addressing the RTM problem. This method requires the information from both the treatment and reference sites in order to predict the expected number of crashes had the safety improvement projects at the treatment sites not been implemented. The information from the reference sites is estimated from a safety performance function (SPF), which is a mathematical relationship that links crashes to traffic exposure. The objective of this dissertation was to develop the SPFs for different functional classes of the Florida State Highway System. Crash data from years 2001 through 2003 along with traffic and geometric data were used in the SPF model development. SPFs for both rural and urban roadway categories were developed. The modeling data used were based on one-mile segments that contain homogeneous traffic and geometric conditions within each segment. Segments involving intersections were excluded. The scatter plots of data show that the relationships between crashes and traffic exposure are nonlinear, that crashes increase with traffic exposure in an increasing rate. Four regression models, namely, Poisson (PRM), Negative Binomial (NBRM), zero-inflated Poisson (ZIP), and zero-inflated Negative Binomial (ZINB), were fitted to the one-mile segment records for individual roadway categories. The best model was selected for each category based on a combination of the Likelihood Ratio test, the Vuong statistical test, and the Akaike's Information Criterion (AIC). The NBRM model was found to be appropriate for only one category and the ZINB model was found to be more appropriate for six other categories. The overall results show that the Negative Binomial distribution model generally provides a better fit for the data than the Poisson distribution model. In addition, the ZINB model was found to give the best fit when the count data exhibit excess zeros and over-dispersion for most of the roadway categories. While model validation shows that most data points fall within the 95% prediction intervals of the models developed, the Pearson goodness-of-fit measure does not show statistical significance. This is expected as traffic volume is only one of the many factors contributing to the overall crash experience, and that the SPFs are to be applied in conjunction with Accident Modification Factors (AMFs) to further account for the safety impacts of major geometric features before arriving at the final crash prediction. However, with improved traffic and crash data quality, the crash prediction power of SPF models may be further improved.

Méthodes de Bootstrap pour les modèles à facteurs

Cette thèse développe des méthodes bootstrap pour les modèles à facteurs qui sont couram- ment utilisés pour générer des prévisions depuis l'article pionnier de Stock et Watson (2002) sur les indices de diffusion. Ces modèles tolèrent l'inclusion d'un grand nombre de variables macroéconomiques et financières comme prédicteurs, une caractéristique utile pour inclure di- verses informations disponibles aux agents économiques. Ma thèse propose donc des outils éco- nométriques qui améliorent l'inférence dans les modèles à facteurs utilisant des facteurs latents extraits d'un large panel de prédicteurs observés. Il est subdivisé en trois chapitres complémen- taires dont les deux premiers en collaboration avec Sílvia Gonçalves et Benoit Perron. Dans le premier article, nous étudions comment les méthodes bootstrap peuvent être utilisées pour faire de l'inférence dans les modèles de prévision pour un horizon de h périodes dans le futur. Pour ce faire, il examine l'inférence bootstrap dans un contexte de régression augmentée de facteurs où les erreurs pourraient être autocorrélées. Il généralise les résultats de Gonçalves et Perron (2014) et propose puis justifie deux approches basées sur les résidus : le block wild bootstrap et le dependent wild bootstrap. Nos simulations montrent une amélioration des taux de couverture des intervalles de confiance des coefficients estimés en utilisant ces approches comparativement à la théorie asymptotique et au wild bootstrap en présence de corrélation sérielle dans les erreurs de régression. Le deuxième chapitre propose des méthodes bootstrap pour la construction des intervalles de prévision permettant de relâcher l'hypothèse de normalité des innovations. Nous y propo- sons des intervalles de prédiction bootstrap pour une observation h périodes dans le futur et sa moyenne conditionnelle. Nous supposons que ces prévisions sont faites en utilisant un ensemble de facteurs extraits d'un large panel de variables. Parce que nous traitons ces facteurs comme latents, nos prévisions dépendent à la fois des facteurs estimés et les coefficients de régres- sion estimés. Sous des conditions de régularité, Bai et Ng (2006) ont proposé la construction d'intervalles asymptotiques sous l'hypothèse de Gaussianité des innovations. Le bootstrap nous permet de relâcher cette hypothèse et de construire des intervalles de prédiction valides sous des hypothèses plus générales. En outre, même en supposant la Gaussianité, le bootstrap conduit à des intervalles plus précis dans les cas où la dimension transversale est relativement faible car il prend en considération le biais de l'estimateur des moindres carrés ordinaires comme le montre une étude récente de Gonçalves et Perron (2014). Dans le troisième chapitre, nous suggérons des procédures de sélection convergentes pour les regressions augmentées de facteurs en échantillons finis. Nous démontrons premièrement que la méthode de validation croisée usuelle est non-convergente mais que sa généralisation, la validation croisée «leave-d-out» sélectionne le plus petit ensemble de facteurs estimés pour l'espace généré par les vraies facteurs. Le deuxième critère dont nous montrons également la validité généralise l'approximation bootstrap de Shao (1996) pour les regressions augmentées de facteurs. Les simulations montrent une amélioration de la probabilité de sélectionner par- cimonieusement les facteurs estimés comparativement aux méthodes de sélection disponibles. L'application empirique revisite la relation entre les facteurs macroéconomiques et financiers, et l'excès de rendement sur le marché boursier américain. Parmi les facteurs estimés à partir d'un large panel de données macroéconomiques et financières des États Unis, les facteurs fortement correlés aux écarts de taux d'intérêt et les facteurs de Fama-French ont un bon pouvoir prédictif pour les excès de rendement.

MRI-based assessment of the pineal gland in a large population of children aged 0-5 years and comparison with pineoblastoma: part I, the solid gland.

INTRODUCTION: Differentiation between normal solid (non-cystic) pineal glands and pineal pathologies on brain MRI is difficult. The aim of this study was to assess the size of the solid pineal gland in children (0-5 years) and compare the findings with published pineoblastoma cases. METHODS: We retrospectively analyzed the size (width, height, planimetric area) of solid pineal glands in 184 non-retinoblastoma patients (73 female, 111 male) aged 0-5 years on MRI. The effect of age and gender on gland size was evaluated. Linear regression analysis was performed to analyze the relation between size and age. Ninety-nine percent prediction intervals around the mean were added to construct a normal size range per age, with the upper bound of the predictive interval as the parameter of interest as a cutoff for normalcy. RESULTS: There was no significant interaction of gender and age for all the three pineal gland parameters (width, height, and area). Linear regression analysis gave 99 % upper prediction bounds of 7.9, 4.8, and 25.4 mm(2), respectively, for width, height, and area. The slopes (size increase per month) of each parameter were 0.046, 0.023, and 0.202, respectively. Ninety-three percent (95 % CI 66-100 %) of asymptomatic solid pineoblastomas were larger in size than the 99 % upper bound. CONCLUSION: This study establishes norms for solid pineal gland size in non-retinoblastoma children aged 0-5 years. Knowledge of the size of the normal pineal gland is helpful for detection of pineal gland abnormalities, particularly pineoblastoma.

Méthodes de Bootstrap pour les modèles à facteurs

Cette thèse développe des méthodes bootstrap pour les modèles à facteurs qui sont couram- ment utilisés pour générer des prévisions depuis l'article pionnier de Stock et Watson (2002) sur les indices de diffusion. Ces modèles tolèrent l'inclusion d'un grand nombre de variables macroéconomiques et financières comme prédicteurs, une caractéristique utile pour inclure di- verses informations disponibles aux agents économiques. Ma thèse propose donc des outils éco- nométriques qui améliorent l'inférence dans les modèles à facteurs utilisant des facteurs latents extraits d'un large panel de prédicteurs observés. Il est subdivisé en trois chapitres complémen- taires dont les deux premiers en collaboration avec Sílvia Gonçalves et Benoit Perron. Dans le premier article, nous étudions comment les méthodes bootstrap peuvent être utilisées pour faire de l'inférence dans les modèles de prévision pour un horizon de h périodes dans le futur. Pour ce faire, il examine l'inférence bootstrap dans un contexte de régression augmentée de facteurs où les erreurs pourraient être autocorrélées. Il généralise les résultats de Gonçalves et Perron (2014) et propose puis justifie deux approches basées sur les résidus : le block wild bootstrap et le dependent wild bootstrap. Nos simulations montrent une amélioration des taux de couverture des intervalles de confiance des coefficients estimés en utilisant ces approches comparativement à la théorie asymptotique et au wild bootstrap en présence de corrélation sérielle dans les erreurs de régression. Le deuxième chapitre propose des méthodes bootstrap pour la construction des intervalles de prévision permettant de relâcher l'hypothèse de normalité des innovations. Nous y propo- sons des intervalles de prédiction bootstrap pour une observation h périodes dans le futur et sa moyenne conditionnelle. Nous supposons que ces prévisions sont faites en utilisant un ensemble de facteurs extraits d'un large panel de variables. Parce que nous traitons ces facteurs comme latents, nos prévisions dépendent à la fois des facteurs estimés et les coefficients de régres- sion estimés. Sous des conditions de régularité, Bai et Ng (2006) ont proposé la construction d'intervalles asymptotiques sous l'hypothèse de Gaussianité des innovations. Le bootstrap nous permet de relâcher cette hypothèse et de construire des intervalles de prédiction valides sous des hypothèses plus générales. En outre, même en supposant la Gaussianité, le bootstrap conduit à des intervalles plus précis dans les cas où la dimension transversale est relativement faible car il prend en considération le biais de l'estimateur des moindres carrés ordinaires comme le montre une étude récente de Gonçalves et Perron (2014). Dans le troisième chapitre, nous suggérons des procédures de sélection convergentes pour les regressions augmentées de facteurs en échantillons finis. Nous démontrons premièrement que la méthode de validation croisée usuelle est non-convergente mais que sa généralisation, la validation croisée «leave-d-out» sélectionne le plus petit ensemble de facteurs estimés pour l'espace généré par les vraies facteurs. Le deuxième critère dont nous montrons également la validité généralise l'approximation bootstrap de Shao (1996) pour les regressions augmentées de facteurs. Les simulations montrent une amélioration de la probabilité de sélectionner par- cimonieusement les facteurs estimés comparativement aux méthodes de sélection disponibles. L'application empirique revisite la relation entre les facteurs macroéconomiques et financiers, et l'excès de rendement sur le marché boursier américain. Parmi les facteurs estimés à partir d'un large panel de données macroéconomiques et financières des États Unis, les facteurs fortement correlés aux écarts de taux d'intérêt et les facteurs de Fama-French ont un bon pouvoir prédictif pour les excès de rendement.

Incremental criterion prediction of personality facets over factors: obtaining unbiased estimates and confidence intervals

 Many researchers have argued that higher order models of personality such as the Five Factor Model are insufficient, and that facet-level analysis is required to better understand criteria such as well-being, job performance, and personality disorders. However, common methods in the extant literature used to estimate the incremental prediction of facets over factors have several shortcomings. This paper delineates these shortcomings by evaluating alternative methods using statistical theory, simulation, and an empirical example. We recommend using differences between Olkin-Pratt adjusted r-squared for factor versus facet regression models to estimate the incremental prediction of facets and present a method for obtaining confidence intervals for such estimates using double adjusted-. r-squared bootstrapping. We also provide an R package that implements the proposed methods.

The cystatin C/creatinine ratio, a marker of glomerular filtration quality: associated factors, reference intervals, and prediction of morbidity and mortality in healthy seniors.

The ratio of cystatin C (cysC) to creatinine (crea) is regarded as a marker of glomerular filtration quality associated with cardiovascular morbidities. We sought to determine reference intervals for serum cysC-crea ratio in seniors. Furthermore, we sought to determine whether other low-molecular weight molecules exhibit a similar behavior in individuals with altered glomerular filtration quality. Finally, we investigated associations with adverse outcomes. A total of 1382 subjectively healthy Swiss volunteers aged 60 years or older were enrolled in the study. Reference intervals were calculated according to Clinical & Laboratory Standards Institute (CLSI) guideline EP28-A3c. After a baseline exam, a 4-year follow-up survey recorded information about overall morbidity and mortality. The cysC-crea ratio (mean 0.0124 ± 0.0026 mg/μmol) was significantly higher in women and increased progressively with age. Other associated factors were hemoglobin A1c, mean arterial pressure, and C-reactive protein (P < 0.05 for all). Participants exhibiting shrunken pore syndrome had significantly higher ratios of 3.5-66.5 kDa molecules (brain natriuretic peptide, parathyroid hormone, β2-microglobulin, cystatin C, retinol-binding protein, thyroid-stimulating hormone, α1-acid glycoprotein, lipase, amylase, prealbumin, and albumin) and creatinine. There was no such difference in the ratios of very low-molecular weight molecules (urea, uric acid) to creatinine or in the ratios of molecules larger than 66.5 kDa (transferrin, haptoglobin) to creatinine. The cysC-crea ratio was significantly predictive of mortality and subjective overall morbidity at follow-up in logistic regression models adjusting for several factors. The cysC-crea ratio exhibits age- and sex-specific reference intervals in seniors. In conclusion, the cysC-crea ratio may indicate the relative retention of biologically active low-molecular weight compounds and can independently predict the risk for overall mortality and morbidity in the elderly.

Use of Bayesian geostatistical prediction to estimate local variations in Schistosoma haematobium infection in western Africa

Objective We aimed to predict sub-national spatial variation in numbers of people infected with Schistosoma haematobium, and associated uncertainties, in Burkina Faso, Mali and Niger, prior to implementation of national control programmes. Methods We used national field survey datasets covering a contiguous area 2,750 × 850 km, from 26,790 school-aged children (5–14 years) in 418 schools. Bayesian geostatistical models were used to predict prevalence of high and low intensity infections and associated 95% credible intervals (CrI). Numbers infected were determined by multiplying predicted prevalence by numbers of school-aged children in 1 km2 pixels covering the study area. Findings Numbers of school-aged children with low-intensity infections were: 433,268 in Burkina Faso, 872,328 in Mali and 580,286 in Niger. Numbers with high-intensity infections were: 416,009 in Burkina Faso, 511,845 in Mali and 254,150 in Niger. 95% CrIs (indicative of uncertainty) were wide; e.g. the mean number of boys aged 10–14 years infected in Mali was 140,200 (95% CrI 6200, 512,100). Conclusion National aggregate estimates for numbers infected mask important local variation, e.g. most S. haematobium infections in Niger occur in the Niger River valley. Prevalence of high-intensity infections was strongly clustered in foci in western and central Mali, north-eastern and northwestern Burkina Faso and the Niger River valley in Niger. Populations in these foci are likely to carry the bulk of the urinary schistosomiasis burden and should receive priority for schistosomiasis control. Uncertainties in predicted prevalence and numbers infected should be acknowledged and taken into consideration by control programme planners.

Asset life prediction and maintenance decision-making using a non-linear non-Gaussian state space model

Estimating and predicting degradation processes of engineering assets is crucial for reducing the cost and insuring the productivity of enterprises. Assisted by modern condition monitoring (CM) technologies, most asset degradation processes can be revealed by various degradation indicators extracted from CM data. Maintenance strategies developed using these degradation indicators (i.e. condition-based maintenance) are more cost-effective, because unnecessary maintenance activities are avoided when an asset is still in a decent health state. A practical difficulty in condition-based maintenance (CBM) is that degradation indicators extracted from CM data can only partially reveal asset health states in most situations. Underestimating this uncertainty in relationships between degradation indicators and health states can cause excessive false alarms or failures without pre-alarms. The state space model provides an efficient approach to describe a degradation process using these indicators that can only partially reveal health states. However, existing state space models that describe asset degradation processes largely depend on assumptions such as, discrete time, discrete state, linearity, and Gaussianity. The discrete time assumption requires that failures and inspections only happen at fixed intervals. The discrete state assumption entails discretising continuous degradation indicators, which requires expert knowledge and often introduces additional errors. The linear and Gaussian assumptions are not consistent with nonlinear and irreversible degradation processes in most engineering assets. This research proposes a Gamma-based state space model that does not have discrete time, discrete state, linear and Gaussian assumptions to model partially observable degradation processes. Monte Carlo-based algorithms are developed to estimate model parameters and asset remaining useful lives. In addition, this research also develops a continuous state partially observable semi-Markov decision process (POSMDP) to model a degradation process that follows the Gamma-based state space model and is under various maintenance strategies. Optimal maintenance strategies are obtained by solving the POSMDP. Simulation studies through the MATLAB are performed; case studies using the data from an accelerated life test of a gearbox and a liquefied natural gas industry are also conducted. The results show that the proposed Monte Carlo-based EM algorithm can estimate model parameters accurately. The results also show that the proposed Gamma-based state space model have better fitness result than linear and Gaussian state space models when used to process monotonically increasing degradation data in the accelerated life test of a gear box. Furthermore, both simulation studies and case studies show that the prediction algorithm based on the Gamma-based state space model can identify the mean value and confidence interval of asset remaining useful lives accurately. In addition, the simulation study shows that the proposed maintenance strategy optimisation method based on the POSMDP is more flexible than that assumes a predetermined strategy structure and uses the renewal theory. Moreover, the simulation study also shows that the proposed maintenance optimisation method can obtain more cost-effective strategies than a recently published maintenance strategy optimisation method by optimising the next maintenance activity and the waiting time till the next maintenance activity simultaneously.

Latent degradation indicators estimation and prediction : a Monte Carlo approach

Asset health inspections can produce two types of indicators: (1) direct indicators (e.g. the thickness of a brake pad, and the crack depth on a gear) which directly relate to a failure mechanism; and (2) indirect indicators (e.g. the indicators extracted from vibration signals and oil analysis data) which can only partially reveal a failure mechanism. While direct indicators enable more precise references to asset health condition, they are often more difficult to obtain than indirect indicators. The state space model provides an efficient approach to estimating direct indicators by using indirect indicators. However, existing state space models to estimate direct indicators largely depend on assumptions such as, discrete time, discrete state, linearity, and Gaussianity. The discrete time assumption requires fixed inspection intervals. The discrete state assumption entails discretising continuous degradation indicators, which often introduces additional errors. The linear and Gaussian assumptions are not consistent with nonlinear and irreversible degradation processes in most engineering assets. This paper proposes a state space model without these assumptions. Monte Carlo-based algorithms are developed to estimate the model parameters and the remaining useful life. These algorithms are evaluated for performance using numerical simulations through MATLAB. The result shows that both the parameters and the remaining useful life are estimated accurately. Finally, the new state space model is used to process vibration and crack depth data from an accelerated test of a gearbox. During this application, the new state space model shows a better fitness result than the state space model with linear and Gaussian assumption.

Hydrologic drought prediction under climate change: Uncertainty modeling with Dempster-Shafer and Bayesian approaches

Representation and quantification of uncertainty in climate change impact studies are a difficult task. Several sources of uncertainty arise in studies of hydrologic impacts of climate change, such as those due to choice of general circulation models (GCMs), scenarios and downscaling methods. Recently, much work has focused on uncertainty quantification and modeling in regional climate change impacts. In this paper, an uncertainty modeling framework is evaluated, which uses a generalized uncertainty measure to combine GCM, scenario and downscaling uncertainties. The Dempster-Shafer (D-S) evidence theory is used for representing and combining uncertainty from various sources. A significant advantage of the D-S framework over the traditional probabilistic approach is that it allows for the allocation of a probability mass to sets or intervals, and can hence handle both aleatory or stochastic uncertainty, and epistemic or subjective uncertainty. This paper shows how the D-S theory can be used to represent beliefs in some hypotheses such as hydrologic drought or wet conditions, describe uncertainty and ignorance in the system, and give a quantitative measurement of belief and plausibility in results. The D-S approach has been used in this work for information synthesis using various evidence combination rules having different conflict modeling approaches. A case study is presented for hydrologic drought prediction using downscaled streamflow in the Mahanadi River at Hirakud in Orissa, India. Projections of n most likely monsoon streamflow sequences are obtained from a conditional random field (CRF) downscaling model, using an ensemble of three GCMs for three scenarios, which are converted to monsoon standardized streamflow index (SSFI-4) series. This range is used to specify the basic probability assignment (bpa) for a Dempster-Shafer structure, which represents uncertainty associated with each of the SSFI-4 classifications. These uncertainties are then combined across GCMs and scenarios using various evidence combination rules given by the D-S theory. A Bayesian approach is also presented for this case study, which models the uncertainty in projected frequencies of SSFI-4 classifications by deriving a posterior distribution for the frequency of each classification, using an ensemble of GCMs and scenarios. Results from the D-S and Bayesian approaches are compared, and relative merits of each approach are discussed. Both approaches show an increasing probability of extreme, severe and moderate droughts and decreasing probability of normal and wet conditions in Orissa as a result of climate change. (C) 2010 Elsevier Ltd. All rights reserved.

Proposal and validation of methodologies for the categorisation of continuous variables in the development of prediction models

158 p.

Prediction in dynamic models with time-dependent conditional variances

This paper considers forecasting the conditional mean and variance from a single-equation dynamic model with autocorrelated disturbances following an ARMA process, and innovations with time-dependent conditional heteroskedasticity as represented by a linear GARCH process. Expressions for the minimum MSE predictor and the conditional MSE are presented. We also derive the formula for all the theoretical moments of the prediction error distribution from a general dynamic model with GARCH(1, 1) innovations. These results are then used in the construction of ex ante prediction confidence intervals by means of the Cornish-Fisher asymptotic expansion. An empirical example relating to the uncertainty of the expected depreciation of foreign exchange rates illustrates the usefulness of the results. © 1992.

A framework for state-based time-series analysis and prediction

Time-series analysis and prediction play an important role in state-based systems that involve dealing with varying situations in terms of states of the world evolving with time. Generally speaking, the world in the discourse persists in a given state until something occurs to it into another state. This paper introduces a framework for prediction and analysis based on time-series of states. It takes a time theory that addresses both points and intervals as primitive time elements as the temporal basis. A state of the world under consideration is defined as a set of time-varying propositions with Boolean truth-values that are dependent on time, including properties, facts, actions, events and processes, etc. A time-series of states is then formalized as a list of states that are temporally ordered one after another. The framework supports explicit expression of both absolute and relative temporal knowledge. A formal schema for expressing general time-series of states to be incomplete in various ways, while the concept of complete time-series of states is also formally defined. As applications of the formalism in time-series analysis and prediction, we present two illustrating examples.