Software Product Line Engineering (SPLE) has been proven to reduce development and maintenance costs, improving the time-to-market, and increasing the quality of product variants developed from a product family via systematic reuse of its core assets. SPLE has been successfully used in the development of safety-critical systems, especially in automotive and aerospace domains. Safety-critical systems have to be developed according to safety standards, which demands safety analysis, Fault Tree Analysis (FTA), and assurance cases safety engineering artefacts. However, performing safety analysis, FTA, and assurance case construction activities from scratch and manually for each product variant is time-consuming and error-prone, whereas variability in safety engineering artefacts can be automatically managed with the support of variant management techniques. As safety is context-dependent, context and design variation directly impact in the safety properties changing hazards, their causes, the risks posed by these hazards to system safety, risk mitigation measures, and FTA results. Therefore, managing variability in safety artefacts from different levels of abstraction increases the complexity of the variability model, even with the support of variant management techniques. To achieve an effective balance between benefits and complexity in adopting an SPLE approach for safety-critical systems it is necessary to distinguish between reusable safety artefacts, whose variability should be managed, and those that should be generated from the reused safety artefacts. On the other hand, both industry and safety standards have recognized the use of model-based techniques to support safety analysis and assurance cases. Compositional safety analysis, design optimization, and model-based assurance cases are examples of techniques that have been used to support the generation of safety artefacts required to achieve safety certification. This thesis aims to propose a model-based approach that integrates model-based development, compositional safety analysis, and variant management techniques to support the systematic reuse and generation of safety artefacts in safety-critical software product line engineering. The approach contributes to reduce the effort and costs of performing safety analysis and assessment for a particular product variant, since such analysis is performed from the reused safety artefacts. Thus, variant-specific fault trees, Failure Modes and Effects Analysis (FMEA), and assurance case artefacts required to achieve safety certification can be automatically generated with the support the model-based safety analysis and assurance case construction techniques. / Engenharia de Linha de Produtos de Software (ELPS) contribui para a redução dos custos de desenvolvimento e de manutenção, a melhoria do time-to-market, e o aumento da qualidade de produtos desenvolvidos a partir de uma família de produtos por meio do reuso sistemático dos ativos principais da linha de produtos. A ELPS vem sendo utilizada com sucesso no desenvolvimento de sistemas embarcados críticos, especificamente nos domínios de sistemas automotivos e aeroespaciais. Sistemas embarcados críticos devem ser desenvolvidos de acordo com os requisitos definidos em padrões de segurança, que demandam a produção de artefatos de análise de segurança, árvores de falhas e casos de segurança. Entretanto, a realização de atividades de análise de segurança, análise de árvores de falhas e construção de casos de segurança de forma manual para cada produto de uma linha de produtos é uma tarefa demorada e propensa a erros. O gerenciamento de variabilidade em artefatos de análise de segurança pode ser automatizado com o apoio de técnicas de gerenciamento de variabilidades. Em virtude de safety ser uma propriedade dependente de contexto, a variabilidade no projeto e contexto inerente uma linha de produtos software impacta na definição de propriedades de segurança do sistema, modificando as ameaças à segurança do sistema, suas causas e riscos, medidas de mitigação aplicáveis, e resultados de análise de árvore de falhas. Dessa forma, gerenciar variabilidades em artefatos relacionados à safety em diferentes níveis de abstração aumenta a complexidade do modelo de variabilidade mesmo com o apoio de técnicas de gerenciamento de variabilidades. Para alcançar o equilíbrio eficaz entre os benefícios e a complexidade da adoção de uma abordagem de ELPS para o desenvolvimento de sistemas embarcados críticos é necessário fazer a distinção entre artefatos de safety reusáveis, em que a variabilidade deve ser gerenciada, e artefatos de safety que devem ser gerados a partir de artefatos reusáveis. Por outro lado, tanto a indústria quanto os padrões de segurança têm reconhecido o uso de técnicas dirigidas a modelos para apoiar a análise segurança e a construção de casos de segurança. Técnicas de análise de segurança composicional e otimização de projeto, e de construção de casos de segurança dirigido a modelos vêm sendo utilizadas para apoiar a geração de artefatos de safety requeridos para certificação. O objetivo desta tese é a proposta de uma abordagem dirigida a modelos que integra técnicas de desenvolvimento dirigido a modelos, análise de segurança composicional e otimização de projeto, e construção de casos de segurança dirigido a modelos para apoiar o reuso sistemático e a geração de artefatos de safety em engenharia de linhas de produtos de sistemas embarcados críticos. A abordagem proposta reduz o esforço e os custos de análise e avaliação de segurança para produtos de uma linha de produtos, uma vez que tal análise é realizada a partir de artefatos de safety reusados. Assim, artefatos como análises de árvores de falhas e de modos de falha e efeitos, e casos de segurança requeridos para certificação podem ser gerados automaticamente com o apoio de técnicas dirigidas a modelos.

This work deals with the and preemptive road users behaviour safety analysis problem. Safety analysis is based on a processing of road users trajectories obtained from processed aerial videos captured by drons. A system for traffic conflicts detection from spatial-temporal data is presented in this work. The standard approach for pro-active traffic conflict indicators evaluation was extended by simulating traffic objects movement in the scene using Ackerman steering geometry in order to get more accurate results.

Component failures can result in catastrophic behaviors in safety critical embedded systems, sometimes resulting in loss of life. Component failures can be treated as off nominal behaviors (ONBs) with respect to the components and sub systems involved in an embedded system. A lot of research is being carried out to tackle the problem of ONBs. These approaches are mainly focused on the states (i.e., desired and undesired states of a system at a given point of time to detect ONBs). In this paper, an approach is discussed to detect component failures and critical components of an embedded system. The approach is based on fault tree analysis (FTA), applied to the requirements specification of embedded systems at design time to find out the relationship between individual component failures and overall system failure. FTA helps in determining both qualitative and quantitative relationship between component failures and system failure. Analyzing the system at design time helps in detecting component failures and critical components and helps in devising strategies to mitigate component failures at design time and improve overall safety and reliability of a system.

This thesis deals with design construction of laboratory model of air press and also proposes of safety measures for this machine. Before the design itself, norms concerning machinery construction were surveyed. There are two norms mentioned in the text ČSN EN 61 508 and ČSN EN ISO 13489-1. Next part shows practical demonstration of machine safety analysis using SISTEMA software tool. Changes to the safety measures were made according to the result of this analysis. Important part of this thesis is also design of control software for the air press in RSlogix5000 environment using Ladder logic and also created visualization of pressing process and show up on to visualization panel PanelView 700.

In human-robot collaborative (HRC) scenarios where humans and robots work together sharing the same workspace, there is a risk of potential hazard that may occur. In this work, an AI-based risk analysis solution has been developed to identify any condition that may harm a robot and its environment. The information from the risk analysis is used in a risk mitigation module to reduce the possibility of being in a hazardous situation. The goal is to develop safety for HRC scenarios using different AI algorithms and to check the possibilities of improving efficiency of the system without any compromise on the safety. This report presents risk mitigation strategies that were built on top of the robot’s control system and based on the ISO 15066 standard. Each of them used semantic information (scene graph) about the robot’s environment and changed the robot’s movement by scaling speed. The first implementation of risk mitigation strategy used Fuzzy Logic System. This system analyzed the riskiest object’s properties to adjust the speed of the robot accordingly. The second implementation used Reinforcement Learning and considered every object’s properties. Three networks (fully connected network, convolutional neural network, and hybrid network) were implemented to estimate the Qvalue function. Additionally, local and edge computation architecture wereimplemented to measure the computational performance on the real robot. Each model was evaluated by measuring the safety aspect and the performance of the robot in a simulated warehouse scenario. All risk mitigation modules were able to reduce the risk of potential hazard. The fuzzy logic system was able to increase the safety aspect with the least efficiency reduction. The reinforcement learning model had safer operation but showed a more compromised efficiency than the fuzzy logic system. Generally, the fuzzy logic system performed up to 28% faster than reinforcement learning but compromised up to 23% in terms of safety (mean risk speed value). In terms of computational performance, edge computation was performed faster than local computation. The bottleneck of the process was the scene graph generation which analyzed an image to produce information for safety analysis. It took approximately 15 seconds to run the scene graph generation on the robot’s CPU and 0.3 seconds on an edge device. The risk mitigation module can be selected depending on KPIs of the warehouse operation while the edge architecture must be implemented to achieve a realistic performance. / I HRC-scenarier mellan människor och robotar där människor och robotar arbetar tillsammans och delar samma arbetsyta finns det risk för potentiell fara som kan uppstå. I detta arbete har en AI-baserad lösning för riskanalys utvecklats för att identifiera alla tillstånd som kan skada en robot och dess miljö. Informationen från riskanalys används i en riskreduceringsmodul för att minska risken för att vara i en farlig situation. Målet är att utveckla säkerhet för HRC-scenarier med olika AI-algoritmer och att kontrollera möjligheterna att förbättra systemets effektivitet utan att kompromissa med säkerheten.Denna rapport presenterar strategier för riskreducering som byggdes ovanpå robotens styrsystem och baserade på ISO 15066-standarden. Var och en av dem använder semantisk information (scendiagram) om robotens miljö och förändrar robotens rörelse genom skalning av hastighet. Den första implementetationen av riskreducerande strategi använder Fuzzy Logic System. Detta system analyserade de mest riskabla objektens egenskaper för att justera robotens hastighet i enlighet därmed. Den andra implementeringen använder förstärkningslärande och betraktade varje objekts egenskaper. Tre nätverk (fully connected network, convolutional neural network, and hybrid network) implementeras för att uppskatta Q-värde-funktionen. Dessutom implementerade vi också lokaloch edge-arkitektur för att beräkna beräkningsprestanda på den verkliga roboten. Varje modell utvärderas genom att mäta säkerhetsaspekten och robotens prestanda i ett simulerat lagerscenario. Alla riskreduceringsmoduler kunde minska risken för potentiell fara. Fuzzy logicsystem kunde öka säkerhetsaspekten med minsta effektivitetsminskning. Förstärkningsinlärningsmodellen har säkrare drift men har en mer begränsad effektivitet än det fuzzy logiska systemet. I allmänhet fungerar fuzzy logicsystem upp till 28 % snabbare än förstärkningslärande men komprometterar upp till 23 % när det gäller säkerhet (medelrisk hastighetsvärde). När det gäller beräkningsprestanda utfördes kantberäkningen snabbare än lokal beräkning. Flaskhalsen för processen var scengrafgenerering som analyserade en bild för att producera information för säkerhetsanalys. Det tog cirka 15 sekunder att köra scengrafgenerering på robotens CPU och 0,3 sekunder på en kantenhet. Modulen för riskreducering kan väljas beroende på KPI för lagerdriften medan edge-arkitekturen måste implementeras för att uppnå en realistisk prestanda.

Yes / Dynamic systems exhibit time-dependent behaviours and complex functional dependencies amongst their components. Therefore, to capture the full system failure behaviour, it is not enough to simply determine the consequences of different combinations of failure events: it is also necessary to understand the order in which they fail. Pandora temporal fault trees (TFTs) increase the expressive power of fault trees and allow modelling of sequence-dependent failure behaviour of systems. However, like classical fault tree analysis, TFT analysis requires a lot of manual effort, which makes it time consuming and expensive. This in turn makes it less viable for use in modern, iterated system design processes, which requires a quicker turnaround and consistency across evolutions. In this paper, we propose for a model-based analysis of temporal fault trees via HiP-HOPS, which is a state-of-the-art model-based dependability analysis method supported by tools that largely automate analysis and optimisation of systems. The proposal extends HiP-HOPS with Pandora, Petri Nets and Bayesian Networks and results to dynamic dependability analysis that is more readily integrated into modern design processes. The effectiveness is demonstrated via application to an aircraft fuel distribution system. / Partly funded by the DEIS H2020 project (Grant Agreement 732242).

This report delves into the early developmental phase of an autonomous vehicle designed for defense applications. Navigating diverse terrains, this unmanned ground vehicle (UGV) poses unique challenges, particularly in the absence of clearly defined directives found in typical traffic scenarios. The analysis employs the Systems-Theoretic Process Analysis (STPA) to identify and anticipate risks inherent in the conceptual stage of product development. Beyond the specific UGV case, the report explores the broader landscape of validating autonomous systems. It discusses prevalent methods, emphasizing adaptability to different contexts and stages of development. By shedding light on the risks and challenges of autonomy in vehicles and examining effective validation strategies, this report aims to contribute valuable insights to the ongoing discourse surrounding autonomous vehicle development.

Previous studies show that pavement condition properties are significant factors to enhance road safety and riding experience, and pavements with low quality might have inadequate performance in terms of safety and riding experience. Pavement Management System (PMS) databases include pavement properties for each segment of the road collected by the agencies. Understanding the impact of road characteristics on crash frequency is a key step to prevent crashes. Whereas other studies analyzed the effect of different characteristics such as International Roughness Index (IRI), Rutting Depth (RD), Annual Average Daily Traffic (AADT), this thesis analyzed the effect of Critical Condition Index (CCI) on crash frequency, in addition to the other factors identified in previous studies. Other characteristics such as Percentage of Heavy Vehicles, Road Surface Condition, Road Lighting Condition, and Driver Conditions are taken into the consideration. The scope of the study is the interstate highway system in Fairfax County, Virginia. Negative Binomial, Least Square and Nominal Logistic Models were developed, showing that the CCI value is a significant factor to predict the number of crashes, and that it has different effect for different values of AADT. The result of this study is a substantial step towards developing an integrated transportation control and infrastructure management framework. / Master of Science / Many factors cause crashes in the roads. Although there is a common sense that road characteristics such as asphalt quality are important in terms of road safety, there are few studies that scientifically prove that statement. In addition, asphalt maintenance decisions making process is mainly based on cost benefit optimization, and traffic safety is not considered at the process. The purpose of this study is to analyze crashes and road characteristics related to each crash to understand the effect of those characteristics on crash frequency, and eventually, to build a model to predict the number of crashes at each part of the road. The model can help transportation agencies to have a better understanding in terms of safety consequences of their infrastructure management plans. The scope of this study is the highway interstate system in Northern Virginia. Results suggest that pavement condition has a significant impact on crash frequency.

Traffic safety has been considered one of the most important issues in the transportation field. With consistent efforts of transportation engineers, Federal, State and local government officials, both fatalities and fatality rates from road traffic crashes in the United States have steadily declined from 2006 to 2011.Nevertheless, fatalities from traffic crashes slightly increased in 2012 (NHTSA, 2013). We lost 33,561 lives from road traffic crashes in the year 2012, and the road traffic crashes are still one of the leading causes of deaths, according to the Centers for Disease Control and Prevention (CDC). In recent years, efforts to incorporate traffic safety into transportation planning has been made, which is termed as transportation safety planning (TSP). The Safe, Affordable, Flexible Efficient, Transportation Equity Act - A Legacy for Users (SAFETEA-LU), which is compliant with the United States Code, compels the United States Department of Transportation to consider traffic safety in the long-term transportation planning process. Although considerable macro-level studies have been conducted to facilitate the implementation of TSP, still there are critical limitations in macroscopic safety studies are required to be investigated and remedied. First, TAZ (Traffic Analysis Zone), which is most widely used in travel demand forecasting, has crucial shortcomings for macro-level safety modeling. Moreover, macro-level safety models have accuracy problem. The low prediction power of the model may be caused by crashes that occur near the boundaries of zones, high-level aggregation, and neglecting spatial autocorrelation. In this dissertation, several methodologies are proposed to alleviate these limitations in the macro-level safety research. TSAZ (Traffic Safety Analysis Zone) is developed as a new zonal system for the macroscopic safety analysis and nested structured modeling method is suggested to improve the model performance. Also, a multivariate statistical modeling method for multiple crash types is proposed in this dissertation. Besides, a novel screening methodology for integrating two levels is suggested. The integrated screening method is suggested to overcome shortcomings of zonal-level screening, since the zonal-level screening cannot take specific sites with high risks into consideration. It is expected that the integrated screening approach can provide a comprehensive perspective by balancing two aspects: macroscopic and microscopic approaches.

Within the safety analysis industry there are a variety of tools used to ensure reliability and security of systems, ranging from mostly qualitative approaches to mostly quantitative. One safety analysis method that lies in between these two is called Layers Of Protection Analysis (LOPA). LOPA is known as a “semi-quantitative” approach that uses a mix of quantitative and qualitative approaches to draw conclusions. In this masters thesis the LOPA approach is demonstrated, in addition to being developed into two alternate LOPA approaches, known as MarkovLOPA and RBDLOPA. These two developed approaches use the concept of Markov chains and Reliability block diagram (RBD) respectively, to extend the applicability of the traditional LOPA methodology. Furthermore, a conventional hydraulic braking system (CHB), which includes ABS/TCS- and ESP functionality was analysed by these three methodologies. The results of the analysis show that in the analysis by LOPA and RBDLOPA 4- and 3 out of 10 scenarios need slight improvements and only 1 scenario for MarkovLOPA. Additionally, the validity of the alternative approaches are analysed by a sensitivity analysis, showing irregularities in the results, leading to the conclusion that further research and development is required prior to industrial applications of the approaches.