In quantified safety engineering, mathematical probability models are used to predict the risk of failure or hazardous events in systems. Markov processes have commonly been utilized to analyze the safety of systems modeled as discrete-state stochastic processes. In continuous time Markov models, transition time between states are exponentially distributed. Semi-Markov processes expand this modeling framework by allowing transition time between states to follow any distribution. This master thesis project seeks to extend the semi-Markov modeling framework even further by allowing hierarchical states, which further relaxes Markov-assumptions by allowing models to keep memory even in state transition. To achieve this, the master thesis proposes a method using the phase-type distribution to replace Markov-chains of states to a single state. For application purposes, it is shown how semi-Markov chains with phase-type distributed transitions can be evaluated by a method using the Laplace-Stieltjes transform. Furthermore, to replace semi-Markov chains, a method to approximate these by the phase-type distribution is presented. This is done by deriving the moments of the time to absorption in a semi-Markov process with a method using the Laplace-Stieltjes transform, and fitting a phase-type distribution with these moments. To evaluate the methods, some case studies are performed on appropriate models. Analytical results are compared with Monte-Carlo simulations and Laplace-transform inverse methods. The results are used to show how hierarchical semi-Markov models can be replaced in an exact manner, and how semi-Markov models can be replaced approximately with varying accuracy. An important conclusion is that by enabling hierarchical modeling, it is possible to predict the safety of systems which demand a more realistic model, as relaxing Markov assumptions allows for more complexity. / Matematiska sannolikhetsmodeller används inom kvantifierad säkerhetsteknik för att utvärdera risken för fel eller farliga olyckor i system. Ett vanligt sätt att analysera säkerheten i system som kan modelleras som stokastiska processer med diskreta tillstånd är att använda Markovprocesser. I tidskontinuerliga Markovprocesser är tidsövergången mellan tillstånd exponentialfördelade. Semi-Markov processer utökar denna modelleringsteknik ytterligare genom att tillåta tidsövergångar som är fördelade enligt alla möjliga fördelningar. Detta examensarbete har som mål att utöka modelleringsmöjligheterna med Semi-Markov processer genom att tillåta hierarkiska tillstånd, som därmed ytterligare utmanar antaganden inom Markov-modeller genom att bibehålla minne efter tillståndsövergång. För att uppnå detta föreslås i denna rapport en metod som använder phase-type-fördelningen för att byta ut Markovkedjor med ett enda tillstånd. För att tillämpa metoden visas hur semi-Markov kedjor kan utvärderas med hjälp av Laplace-Stieltjes-transformen. För att kunna ersätta semi-Markov kedjor med samma metod presenteras även en approximationsmetod för att åter igen använda phase-type-fördelningen. Detta görs genom att använda Laplace-Stieltjes-transformen för att generera momenten av tiden till absorption i semi-Markov processer, och anpassa dessa till momenten av en phase-type-fördelning. För att utvärdera metoderna presenteras en del exempel. Analytiska resultat jämförs med Monte-Carlo simulering och inverteringsmetoder för Laplace-transformen. Resultaten används för att visa hur hierarkiska Markov modeller kan ersättas exakt, och hur semi-Markov processer kan approximeras med varierande noggrannhet. En viktig slutsats är att genom att tillåta hierarkisk modellering är det möjligt att utvärdera säkerheten i system som kräver mer realistiska modeller, då detta öppnar upp för mer komplexitet.

The work presents a novel voltage biasing design that helps the CMOS RF circuits resilient to variability and reliability. The biasing scheme provides resilience through the threshold voltage (VT) adjustment, and at the mean time it does not degrade the PA performance. Analytical equations are established for sensitivity of the resilient biasing under various scenarios. Power Amplifier (PA) and Low Noise Amplifier (LNA) are investigated case by case through modeling and experiment. PTM 65nm technology is adopted in modeling the transistors within these RF blocks. A traditional class-AB PA with resilient design is compared the same PA without such design in PTM 65nm technology. Analytical equations are established for sensitivity of the resilient biasing under various scenarios. A traditional class-AB PA with resilient design is compared the same PA without such design in PTM 65nm technology. The results show that the biasing design helps improve the robustness of the PA in terms of linear gain, P1dB, Psat, and power added efficiency (PAE). Except for post-fabrication calibration capability, the design reduces the majority performance sensitivity of PA by 50% when subjected to threshold voltage (VT) shift and 25% to electron mobility (μn) degradation. The impact of degradation mismatches is also investigated. It is observed that the accelerated aging of MOS transistor in the biasing circuit will further reduce the sensitivity of PA. In the study of LNA, a 24 GHz narrow band cascade LNA with adaptive biasing scheme under various aging rate is compared to LNA without such biasing scheme. The modeling and simulation results show that the adaptive substrate biasing reduces the sensitivity of noise figure and minimum noise figure subject to process variation and iii device aging such as threshold voltage shift and electron mobility degradation. Simulation of different aging rate also shows that the sensitivity of LNA is further reduced with the accelerated aging of the biasing circuit. Thus, for majority RF transceiver circuits, the adaptive body biasing scheme provides overall performance resilience to the device reliability induced degradation. Also the tuning ability designed in RF PA and LNA provides the circuit post-process calibration capability.

Thesis (MEng)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Standards define target reliability levels that govern the safety of designed structures. These target levels should be around an economic optimum for the class of structure under consideration. However, society may have safety requirements in excess of that required to achieve an economic optimum. The LQI criterion can be used to determine society’s willingness to invest in safety, thereby defining a minimum acceptable safety- or reliability level. This thesis determines economically optimised reliability levels for reliability class two concrete structures in South Africa, over a range of typical input parameters. Rackwitz’s (2000) approach is used here, adjusted for the South African context. The structure is described using a simple limit state function, defined as the difference between load and resistance, with resistance a function of a global safety parameter. South African construction costs, costs of increasing safety, failure costs and discount rates are used in the objective function for economic optimisation. Life Quality Index (LQI) theory is used as a basis to derive society’s willingness to pay (SWTP) for safety and the corresponding reliability level is found by applying the LQI criterion. In the South African context the derivation of SWTP presents some challenges, which is discussed. Situations where the minimum required reliability would exceed the economically optimum reliability level are discussed. Various reliability based cost optimization case studies are conducted covering a broad range of typical concrete design situations. From these case studies a range of target reliability indices are derived for typical concrete structural components and failure modes. Obtained values are compared to current South African target levels of reliability provided by the South African loading code and recommendations are made. The approach used by Rackwitz (2000) is compared with results obtained from case studies and used as basis to estimate optimum reliability levels for other types of buildings. Functions are written in MATLAB to allow replication of the study for others seeking to derive optimum reliability indices. / AFRIKAANSE OPSOMMING: Standaarde spesifiseer teiken betroubaarheidsvlakke wat die veiligheidsvlak van ontwerpte strukture bepaal. Hierdie teikenvlak moet rondom die ekonomiese optimum wees vir die klas van struktuur onder oorweging. Die samelewing verkies moontlik ‘n hoër veiligheidsvlak as wat deur die ekonomiese optimum dikteer word. Die LKI (Lewens Kwaliteit Indeks) maatstaf kan gebuik word om die samelewing se bereidwilligheid om in veiligheid te belê te bepaal en sodoende ‘n minimum veiligheidsvlak bepaal. Hierdie tesis bepaal die ekonomiese optimum betroubaarheidsvlak vir klas twee beton strukture in Suid-Afrika vir wisselende parameters. Rackwitz (2000) se benadering word in hierdie studie gebruik en is aangepas vir Suid-Afrikaanse omstandighede. Die struktuur word beskryf deur ‘n eenvoudige limiet staat funksie, gedefinieer as die verskil tussen die las en weerstand, met die weerstand as die funksie van ‘n globale veiligheidsparameter. Suid-Afrikaanse konstruksie koste, veiligheidsvermedering koste, falingskoste en diskonteer koerse word gebruik vir optimering. Die LKI teorie word gebruik om SBB (Samelewing Bereidheid om te Belê) vir veiligheid af te lei en die ooreenkomstige betroubaarheidsvlak word bepaal deur die LKI maatstaf toe te pas. In die afleiding hiervan vir Suid-Afrikaanse omstandighede is sekere uitdagings teegekom wat bespreek word. Situasies waar die minimum betroubaarheidsvlak hoer is as die ekonomiese optimum word bespreek. Verskillende betroubaarheids gebaseerde optimering gevalstudies word gedoen op tipiese beton struktuur elemente. Van hierdie gevalstudies is optimum betroubaarheidsindekse vir die tipiese beton elemente en galingsmodusie afgelei. Die betroubaarheidsindekse word vergelyk met huidige betroubaarheidsindekse soos wat voorgeskryf is in die Suid-Afrikaanse laskode (SANS10160-1(2011)). Rackwitz (2000) se benadering word vergelyk met die resultate van die gevallestudies en word gebruik as basis om optimum betroubaarheidsvlakke vir ander tipes geboue te voorspel. MATLAB funksies is geprogrameer om minimum en optimum betroubaarheidsindekse af te lei.

Este trabalho propõe um método com base no planejamento estatístico de testes de confiabilidade para proporcionar um tratamento quantitativo aos dois primeiros níveis da abordagem de padrões baseados em metas (Goal-Based Standards). O uso do método é demonstrado em dois exemplos práticos, através da sua aplicação na meta de confiabilidade dos componentes do sistema de pós-tratamento de gases de exaustão, em motores a diesel de veículos automotivos e marítimos. Os fundamentos do trabalho incluem a teoria do planejamento estatístico de testes de confiabilidade, os princípios do GBS, tipos e características de metas de confiabilidade, a ferramenta de avaliação de segurança formal (FSA), e da análise do modo e efeito da falha (FMEA). A originalidade da contribuição do trabalho está na combinação das técnicas de planejamento de testes de confiabilidade com os princípios do GBS para quantificar o nível de atendimento das metas de confiabilidade. Dessa forma, é possível ajustar o programa de testes na verificação das metas de segurança, aumentando o conhecimento sobre o desempenho do sistema. O método é de simples aplicação permitindo a sua aplicação expedita e integrada aos processos de desenvolvimento de sistemas. / This research proposes a probabilistic method based on reliability test planning in order to provide a quantitative approach to the Tier I and II of the Goal-Based Standards (GBS). The use of the method is exhibited through its application in the reliability goal of the components of the exhaust gases post-treatment system in diesel engines for automotive and marine vehicles. The fundamentals of this work includes the theory of statistical design of reliability tests, the GBS principles, a brief description of the types and characteristics of reliability goals, the tool of Formal Safety Assessment (FSA) recommended for marítima projects, and the risk qualitative methodology based on Failure Mode Effect Analysis (FMEA). This theoretical framework was applied to two practical examples, one in the automotive industry and the other in the shipbuilding sector. The originality of this research lies in the combination of the reliability test planning techniques with the principles of GBS, in order to verify, from the whole project tests planning, the reliability goals of the basic components as a fundamental step in the assessment of system safety. The testing program can be adjusted to focus on verifying the safety goals, so that the level of knowledge about the product performance before it is delivered to the client. This work also offers the benefit of providing an easy method of system safety assessment, which allows a large number of companies to use it and to integrate it into their product development process.

As companhias estão inseridas em um contexto global de preocupação sustentável, onde questões ambientais e sociais acompanham as preocupações econômicas. Faz-se necessário englobar, de alguma forma, nas análises de viabilidade econômica ou ainda em modelagens de Vida Econômica e Reposição de Ativos, variáveis que possam representar aspectos sociais e ambientais. Nesse panorama entra a norma PAS-55 - Publicly Avaiable Specification - 55, publicada pela British Standards Institution, que visa aperfeiçoar o sistema de gestão de ativos físicos de maneira a permitir maiores resultados econômicos, sociais e ambientais, provenientes da utilização dos equipamentos, nas empresas. Os procedimentos previstos na norma PAS-55 sugerem o alinhamento de conhecimentos oriundos da gestão de manutenção, que compõem o arcabouço da Engenharia de Confiabilidade, a conhecimentos de Engenharia Econômica, de maneira a compor ferramentas e métodos para tomar decisões em relação aos ativos físicos de empresas. O objetivo do presente trabalho é propor um modelo de substituição de ativos que considere variáveis de Confiabilidade e Sustentabilidade, juntamente com variáveis financeiras, por meio da Análise Envoltória de Dados. Os resultados apresentam um modelo conceitual de substituição de ativos, com inclusão de variáveis socioambientais e aplicação DEA e, também, um estudo de caso com dados reais que avalia a substituição de um equipamento pertencente a uma empresa produtora de ácido fosfórico. / Companies are placed in a global context of sustainable concern where environmental and social issues accompanying economic concerns. It is necessary to include, in some way, the economic feasibility analysis or in modeling of Economic Life and Asset Replacement, variables that may pose social and environmental aspects. In this scenario enters the PAS-55 standard - Publicly Avaiable Specification - 55, published by the British Standards Institution, which aims to improve the management system of physical assets in order to allow greater economic results, social and environmental issues arising from the use of equipment, in companies. The procedures laid down in PAS-55 standard suggests the alignment of knowledge from maintenance management, which make up the framework of Reliability Engineering, the knowledge of economic engineering in order to make tools and methods to make decisions in relation to physical assets companies. The objective of this study is to propose a model of asset replacement that considers variables Reliability and Sustainability, along with financial variables, using data envelopment analysis. The results present a conceptual model of asset replacement, with the inclusion of social and environmental variables and DEA application and also a case study with real data that assesses the replacement of equipment belonging to a producer of phosphoric acid.

Spacecraft fulfill a myriad of critical functions on orbit, from defense and intelligence to science, navigation, and telecommunication. Spacecraft can also cost several hundred millions of dollars to design and launch, and given that physical access for maintenance remains difficult if not impossible to date, designing high reliability and survivability into these systems is an engineering and financial imperative. While reliability is recognized as an essential attribute for spacecraft, little analysis has been done pertaining to actual field reliability of spacecraft and their subsystems. This thesis consists of two parts. The first part fills the gap in the current understanding of spacecraft failure behavior on orbit through extensive statistical analysis and modeling of anomaly and failure data of Earth-orbiting spacecraft. The second part builds on these results to develop a novel theoretical basis (interdependent multi-layer network approach) and algorithmic tools for the analysis of survivability of spacecraft and space-based networks. Space-based networks (SBNs) allow the sharing of on-orbit resources, such as data storage, processing, and downlink. Results indicate and quantify the incremental survivability improvement of the SBN over the traditional monolith architecture. A trade-space analysis is then conducted using non-descriptive networkable subsystems/technologies to explore survivability characteristics of space-based networks and help guide design choices.

Reliability prediction of components operating in complex systems has historically been conducted in a statistically isolated manner. Current physics-based, i.e. mechanistic, component reliability approaches focus more on component-specific attributes and mathematical algorithms and not enough on the influence of the system. The result is that significant error can be introduced into the component reliability assessment process. The objective of this study is the development of a framework that infuses the influence of the system into the process of conducting mechanistic-based component reliability assessments. The formulated framework consists of six primary steps. The first three steps, identification, decomposition, and synthesis, are qualitative in nature and employ system reliability and safety engineering principles for an appropriate starting point for the component reliability assessment. The most unique steps of the framework are the steps used to quantify the system-driven local parameter space and a subsequent step using this information to guide the reduction of the component parameter space. The local statistical space quantification step is accomplished using two newly developed multivariate probability tools: Multi-Response First Order Second Moment and Taylor-Based Inverse Transformation. Where existing joint probability models require preliminary statistical information of the responses, these models combine statistical information of the input parameters with an efficient sampling of the response analyses to produce the multi-response joint probability distribution. Parameter space reduction is accomplished using Approximate Canonical Correlation Analysis (ACCA) employed as a multi-response screening technique. The novelty of this approach is that each individual local parameter and even subsets of parameters representing entire contributing analyses can now be rank ordered with respect to their contribution to not just one response, but the entire vector of component responses simultaneously. The final step of the framework is the actual probabilistic assessment of the component. Variations of this final step are given to allow for the utilization of existing probabilistic methods such as response surface Monte Carlo and Fast Probability Integration. The framework developed in this study is implemented to conduct the finite-element based reliability prediction of a gas turbine airfoil involving several failure responses. The framework, as implemented resulted in a considerable improvement to the accuracy of the part reliability assessment and an increased statistical understanding of the component failure behavior.

A utilização de ferramentas externas de diagnóstico guiado tem se tornado cada vez mais importante nas atividades de pós-venda da indústria automotiva. Isso se dá principalmente devido ao uso extensivo de sistemas embarcados nos veículos, tornando-os mais complexos e difíceis de diagnosticar. Atualmente, as técnicas empregadas para o desenvolvimento da ferramenta de diagnóstico guiado são fortemente dependentes da experiência do projetista e centralizadas nas peças e subsistemas do veículo, possibilitando baixo grau de flexibilidade e reaproveitamento da informação. Este trabalho propõe uma nova metodologia para o desenvolvimento da ferramenta de diagnóstico guiado, aplicado a um estudo de caso da indústria automotiva, numa arquitetura de software em três camadas: peças e componentes do veículo, informações e estratégia para o diagnóstico e uma camada de apresentação. Isso permite grande flexibilidade no projeto da ferramenta de diagnóstico guiado para diferentes modelos de veículos, fabricantes de peças e sistemas automotivos. A metodologia proposta é aplicada em um estudo de caso de diagnóstico da Volvo caminhões, mostrando o processo de adaptação da arquitetura de software de três camadas à metodologia proposta e seu impacto no custo do desenvolvimento da ferramenta de diagnóstico. / External guided diagnostic tools are increasingly important to the aftermarket business of automotive industry. It occurs mainly due to the extensive using of embedded systems in vehicles, making them more complex and difficult to diagnose. Currently, the techniques used to develop a guided diagnostic tool are strongly dependent on designer’s experience and are usually focused on parts and vehicle’s subsystems, allowing low flexibility and reduced information reusage. This paper proposes a new methodology for development of a guided diagnostic tool applied to the automotive industry. This methodology is based on a three-tier software architecture composed of vehicle’s parts and components, diagnostic information and strategy, and presentation layer. It allows great flexibility for designing a guided diagnostic tool for different vehicle models, parts OEMs and automotive systems. The proposed methodology has been applied to a case study at Volvo Trucks. The corresponding adaptation process to the three-tier software architecture is presented as well as its impact on development costs. / 5000

Cette thèse est consacrée à la modélisation de la maintenance imparfaite par les modèles statistiques. La maintenance imparfaite est le cas intermédiaire de deux situations extrêmes : la maintenance minimale où le système est remis en état qu’il se trouve juste avant la défaillance, et la maintenance parfaite où le système est remplacé à neuf. Dans ce cadre, les expressions analytiques des grandeurs de fiabilité d’un modèle de maintenance imparfaite sont développées. La convergence du modèle est soulignée, et les lois asymptotiques dans le régime stationnaire sont proposées. Ensuite, les lois sont appliquées pour proposer des politiques de maintenance préventive avec des maintenances imparfaites. Le deuxième thème consiste à proposer une adaptation de procédure d’estimation pour les fenêtres d’observation. Seuls les événements pendant ces périodes sont observés. La modélisation et l’inférence s’appuient sur la convergence de modèles vers leurs régimes stationnaires, ou sur la modélisation de l’âge du système au début des fenêtres. Enfin, l’inférence bayésienne d’un modèle de maintenance imparfaite y est présentée. On étudie l’impact des choix des lois a priori sur la qualité des estimations par des simulations numériques. La sélection des modèles de maintenance imparfaite par le facteur de Bayes est proposée. Dans chaque partie, les modélisations statistiques sont appliquées à des données issues de l’industrie / The thesis analyses imperfect maintenance processes of industrial systems by statistical models. Imperfect maintenance is an intermediate situation of two extremes ones: minimal maintenance where the system is restored to the state immediately prior to failure, and perfect maintenance where the system is renewed after the failure. Analytical expressions of reliability quantities of an imperfect maintenance model are developed. The convergence of the model is highlighted and the asymptotic expressions are proposed. The results are applied to build some preventive maintenance policies that contain only imperfect maintenances. The second part of the thesis consists of analyzing failure data contained in observation windows. An observation window is a period of the entire functioning history that only the events occurring in this period are recorded. The modelling and the inference are based on the convergence property or the modelling of initial age. Finally, Bayesian inference of an imperfect maintenance model is presented. The impact of the choices of a priori distributions is analyzed by numerical simulations. A selection method of imperfect maintenance models using the Bayes factor is also introduced.The statistical modelling in each section is applied to real data

A construção de linhas de transmissão são obras de grande impacto ambiental e requerem grandes investimentos financeiros. Este projeto de pesquisa, visa por meio de ensaios de laboratório, conceber uma ferramenta sistêmica para o aperfeiçoamento do método de determinação da fluência em cabos condutores de energia e OPGW (Optical Ground Wire), que tem sido utilizada no Brasil durante os últimos 30 anos, e analisar os efeitos na construção das linhas aéreas de transmissão. Visa também, proporcionar ganhos de confiabilidade ao sistema de transmissão de energia elétrica, pois o projeto, a construção, a operação e a manutenção de linhas de transmissão dependem de parâmetros de desempenho mecânico dos condutores. Conclui–se que a metodologia de ensaio proposta proporciona resultados finais mais confiáveis quando comparado com o atual procedimento normatizado, utilizado no Brasil, e sua utilização em projetos de linhas aéreas de transmissão de energia podem reduzir custos de construção, aumentar a ampacidade das linhas já existentes, bem como, diminuir os riscos ao a que pessoas estão sujeitas quando expostas a campos elétricos e eletromagnéticos gerados por linhas de transmissão. / Overhead Transmission Line construction projects have a great environmental impact and require a large financial investment. This research aims, through laboratory tests, to improve the method of determining the creep of power cables and OPGW (Optical Ground Wire) that has been utilized in Brazil for the last thirty years, as well as analyze the effects on the construction of overhead transmission lines. It will provide greater reliability to the overhead transmission line system, since the design, construction, operation and maintenance of transmission lines depend on the mechanical performance of the conductors. In conclusion, the proposed methodology provides more reliable final results compared to the current standardized procedure and its use in overhead transmission line projects could reduce construction costs, increase the ampacity, as well as reduce the risks to which people are subject when exposed to electric and electromagnetic fields generated by power lines.