No description available.

The thesis proposes a number of techniques for elaborating requirements constructively from high-level goals. The techniques are based on the KAOS goal-oriented method for requirements engineering. This method consists in identifying goals and refining them into subgoals until the latter can be assigned as responsibilities of single agents such as humans, devices and software. Domain properties and assumptions about the software environment are also used during the goal refinement process. The method supports the exploration of alternative goal refinements and alternative responsibility assignments of goals to agents. It also supports the identification and resolution of conflicts between goals, and the identification and resolution of exceptional agent behaviors, called obstacles, that violate goals and assumptions produced during the goal refinement process. The thesis enriches the KAOS framework through three kinds of techniques: (a) techniques for identifying agents, goal refinements, and alternative responsibility assignments, and for deriving agent interfaces from such responsibility assignments; (b) techniques for deriving operational requirements from goal specifications; (c) techniques for generating obstacles to the satisfaction of idealized goals and assumptions, and for generating alternative obstacle resolutions. The result is a coherent body of systematic techniques for requirements elaboration that are both theoretically well-founded (a formal model of agent is defined) and effective in practice (the techniques are validated on two real case studies of significant size: the London ambulance despatching system, and the Bay Area Rapid Transit train system).

The thesis proposes a number of techniques for elaborating requirements constructively from high-level goals. The techniques are based on the KAOS goal-oriented method for requirements engineering. This method consists in identifying goals and refining them into subgoals until the latter can be assigned as responsibilities of single agents such as humans, devices and software. Domain properties and assumptions about the software environment are also used during the goal refinement process. The method supports the exploration of alternative goal refinements and alternative responsibility assignments of goals to agents. It also supports the identification and resolution of conflicts between goals, and the identification and resolution of exceptional agent behaviors, called obstacles, that violate goals and assumptions produced during the goal refinement process. The thesis enriches the KAOS framework through three kinds of techniques: (a) techniques for identifying agents, goal refinements, and alternative responsibility assignments, and for deriving agent interfaces from such responsibility assignments; (b) techniques for deriving operational requirements from goal specifications; (c) techniques for generating obstacles to the satisfaction of idealized goals and assumptions, and for generating alternative obstacle resolutions. The result is a coherent body of systematic techniques for requirements elaboration that are both theoretically well-founded (a formal model of agent is defined) and effective in practice (the techniques are validated on two real case studies of significant size: the London ambulance despatching system, and the Bay Area Rapid Transit train system).

This thesis aims to develop a knowledge structuring framework to organize knowledge according to means-ends relationships. Means-ends relationships are particularly relevant in technology and goal-oriented domains such as the geo-engineering domain, where technical problems are identified, and solutions proposed. The proposed goal oriented representation in this thesis does not replace current classification methodologies. In this project, a small corpus of research publications from a technology domain is used to help construct the framework. The main means-ends relationships from the articles are manually extracted and represented in a graphical model showing which problems are approached, by which solutions proposed, and in which publications. Proposed solutions can lead to new problems which are in turn addressed by solutions proposed in other publications. A metamodel is derived to capture the important concepts and relationships relevant for this purpose. The metamodel, and the framework have undergone several iterations before finalization.

This thesis aims to develop a knowledge structuring framework to organize knowledge according to means-ends relationships. Means-ends relationships are particularly relevant in technology and goal-oriented domains such as the geo-engineering domain, where technical problems are identified, and solutions proposed. The proposed goal oriented representation in this thesis does not replace current classification methodologies. In this project, a small corpus of research publications from a technology domain is used to help construct the framework. The main means-ends relationships from the articles are manually extracted and represented in a graphical model showing which problems are approached, by which solutions proposed, and in which publications. Proposed solutions can lead to new problems which are in turn addressed by solutions proposed in other publications. A metamodel is derived to capture the important concepts and relationships relevant for this purpose. The metamodel, and the framework have undergone several iterations before finalization.

Background: Requirements engineering (RE) is the process of identifying, defining, documenting, and validating requirements. However, RE approaches are usually not applied to AI-based systems due to their ambiguity and is still a growing subject. Research also shows that the quality of ML-based systems is affected due to the lack of a structured RE process. Hence, there is a need to apply RE techniques in the development of ML-based systems.  Objectives: This research aims to identify the practices and challenges concerning RE techniques for AI-based systems in autonomous driving and then to identify a suitable RE approach to overcome the identified challenges. Further, the thesis aims to check the feasibility of the selected RE approach in developing a prototype AI-based classification system for road markings.  Methods: A combination of research methods has been used for this research. We apply techniques of interviews, case study, and a rapid literature review. The case company is Scania CV AB. A literature review is conducted to identify the possible RE approaches that can overcome the challenges identified through interviews and discussions with the stakeholders. A suitable RE approach, GR4ML, is found and used to develop and validate an AI-based classification system for road markings.  Results: Results indicate that RE is a challenging subject in autonomous driving. Several challenges are faced at the case company in eliciting, specifying, and validating requirements for AI-based systems, especially in autonomous driving. Results also show that the views in the GR4ML framework were suitable for the specification of system requirements and addressed most challenges identified at the case company. The iterative goal-oriented approach maintained flexibility during development. Through the system's development, it was identified that the Random Forest Classifier outperformed the Logistic Regressor and Support Vector Machine for the road markings classification.  Conclusions: The validation of the system suggests that the goal-oriented requirements engineering approach and the GR4ML framework addressed most challenges identified in eliciting, specifying, and validating requirements for AI-based systems at the case company. The views in the GR4ML framework provide a good overview of the functional and non-functional requirements of the lower-level systems in autonomous driving. However, the GR4ML framework might not be suitable for validation of higher-level AI-based systems in autonomous driving due to their complexity.

Submitted by Marlene Santos (marlene.bc.ufg@gmail.com) on 2016-08-09T17:19:49Z No. of bitstreams: 2 Dissertação - Heyde Francielle do Carmo França - 2016.pdf: 7287432 bytes, checksum: 9138c675f605c1734af600ab0faf3141 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2016-08-10T11:33:59Z (GMT) No. of bitstreams: 2 Dissertação - Heyde Francielle do Carmo França - 2016.pdf: 7287432 bytes, checksum: 9138c675f605c1734af600ab0faf3141 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2016-08-10T11:33:59Z (GMT). No. of bitstreams: 2 Dissertação - Heyde Francielle do Carmo França - 2016.pdf: 7287432 bytes, checksum: 9138c675f605c1734af600ab0faf3141 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2016-03-29 / The Goal Oriented Requirements Engineering (GORE) approach represents users’ needs through goals with focus on capturing the real intentions of stakeholders. Based on the GORE technique, the i* modeling language represents system’s and organization’s goals and brings several advantages. Despite that, the i* language faces problems regarding the quality of models, which include typical mistakes of misuse of i* constructs, the presence of ambiguities on the interpretation of those constructs, and the complexity of the resulting i* models. The aim of this work is to present an ontology-based solution for i* models in order to reduce the most well-known errors while constructing such models. To achieve this goal was accomplished initially a literature search, followed by an experimental research to produce the proposed solution This solution includes the extension of an ontology called OntoiStar+ with OWL restrictions to ensure that frequent mistakes in i* models are not found. Besides, the TAGOOn+ tool was also extended to validate i* models in the iStarML language and convert those to an OWL representation.To perform the tests were modeled two different domains, Media Shop and on universities, using these domains case studies have been reproduced and measured results. Results demonstrate an approximate coverage of 70% of those common errors with extension of OntoiStar+ and more than 80% with extension of TAGOOn+ tool. / A abordagem de Engenharia de Requisitos Orientada a Metas (do Inglês, GORE) representa as necessidades dos usuários através de metas e intenções, focando em capturar a real intenção dos stakeholders. Baseada na técnica GORE, a linguagem de modelagem i* representa metas do sistema e da organização e traz diversas vantagens. Apesar disso, a linguagem i* apresenta problemas relacionados à qualidade dos modelos, que incluem erros típicos de mau uso dos construtores, à presença de ambiguidades na interpretação dos construtores e à complexidade dos modelos resultantes. Assim, o objetivo desta dissertação é apresentar uma solução baseada em ontologia visando a redução de erros comuns em modelos de requisitos construídos na linguagem i*. Para atingir tal objetivo foi realizada inicialmente uma pesquisa bibliográfica, seguida de uma pesquisa experimental para produzir a solução proposta. Esta solução foi implementada realizando a extensão de um ontologia chamada OntoiStar+, na qual foram inseridas restrições na linguagem OWL para garantir que os erros frequentes de modelos i* não sejam reproduzidos. Foi realizada também a extensão da ferramenta TAGOOn+ para validação de modelos i* escritos em iStarML e conversão para modelos em OWL. Para realização dos testes foram modelados dois domínios diferentes, o Media Shop e um sobre universidades, usando estes domínios foram reproduzidos estudos de casos e mensurados os resultados. Os testes realizados em ambas soluções geraram resultados satisfatórios. Os resultados demonstraram uma cobertura de mais de 70% dos erros mais comuns com a extensão da OntoiStar+ e mais de 80% com a extensão da ferramenta TAGOOn+ .

Made available in DSpace on 2016-06-02T19:06:05Z (GMT). No. of bitstreams: 1 5226.pdf: 3993089 bytes, checksum: a0750877fc955b7274271f1326ade7b3 (MD5) Previous issue date: 2011-08-24 / One of the benefits of Service-Oriented Architecture is to make business processes adaptable when this architecture is adopted during software development. For reaching this purpose, it is essential to have a support for services identification in order to meet the business goals. However, many available services found on web environment are too specific and can hardly be reused in different applications. This happens because there is a lack of systematic approaches for supporting generic services identification in a systematic way. Objective: Presenting a strategy for identifying generic services that support business processes. The identification is supported by Goal Diagrams and Business Process Models and is composed by a set of guidelines which assist the domain engineer in extracting the services. The identified services are generic enough to be reused in similar applications of a specific domain. Methodology: To elaborate our strategy, some domain-specific business process were analyzed, aiming at extracting key tasks and turn them into generic web services. This analysis was supported by an extended version of goal diagrams (GTR) and conventional BPM models. Results: As a proof-of-concept we applied our strategy for identifying services in the planning processes domain and we developed a real e-gov web portal based on the identified services. The web portal was used successfully by two different schools for elaborating their planning processes. Conclusion: We claim that our strategy is generic and can be applied to other business processes providing software suitability to the organization dynamics, besides the potential reuse of services in different instances of the same business process. / Um dos benefícios da Arquitetura Orientada a Serviços é tornar os processos de negócios adaptáveis quando esta arquitetura é adotada durante o desenvolvimento de software. Para atingir este propósito, é essencial ter um suporte para a identificação de serviços a fim de atender os objetivos de negócio. No entanto, muitos serviços disponíveis encontrados no ambiente web dificilmente podem ser reutilizados em diferentes aplicações. Isso acontece porque faltam abordagens sistemáticas de apoio à identificação de serviços genéricos de uma forma sistemática. Objetivo: Apresentar uma estratégia para identificar serviços genéricos que dêem suporte aos processos de negócios. A identificação é apoiada por diagramas de objetivos e modelos de processos de negócios, e é composto por um conjunto de diretrizes que auxiliam o engenheiro de domínio na extração dos serviços. Os serviços identificados são genérico o suficiente para serem reutilizados em aplicações semelhantes de um domínio específico. Metodologia: Para elaborar a nossa estratégia, um processo de negócio específico de domínio foi analisado, com o objetivo de extrair tarefas chaves deste processo e transformá-las em serviços web genéricos. Esta análise foi apoiada por uma versão estendida de diagramas de objetivos (GTR) e modelos BPM convencionais. Resultados: Como prova de conceito, aplicamos a nossa estratégia para a identificação de serviços no domínio do processo de planejamento e desenvolvemos um portal web real com base nos serviços identificados. O portal foi utilizado com sucesso por duas escolas diferentes para a elaboração de seus processos de planejamento. Conclusão: Afirmamos que nossa estratégia é genérica e pode ser aplicada a outros processos de negócios provendo a adequação do software à dinâmica organização, além do potencial de reúso de serviços em diferentes instâncias do mesmo processo de negócio.

La sécurité-innocuité est une propriété émergente des systèmes critiques de sécurité (SCS), notamment les systèmes ferroviaires. Cet aspect émergent complexifie leur processus du développement et nécessite un raisonnement judicieux permettant de diminuer les dangers. Cette thèse propose une approche ontologique qui intègre les activités de sécurité dès les premières phases de conception des SCS. Ce cadre structuré offre une harmonisation sémantique entre les domaines impliqués, tels que l'ingénierie de sécurité et l'Ingénierie des Exigences Dirigée par les Buts (IEDB). La logique métier intégrée dans cette approche est validée par des cas d'étude ferroviaires d'accidents réels et d'une mission télé-opérée. Dans un premier temps, nous avons proposé une ontologie d'analyse dysfonctionnelle appelée DAO et fondée sur l'ontologie de haut niveau UFO. DAO considère les aspects sociaux-techniques et environnementaux des SCS et intègre les différents types de fautes et de propriétés cognitives liés respectivement aux défaillances techniques et aux erreurs humaines. Le modèle conceptuel de DAO est exprimé en OntoUML et formalisé en langage OWL afin de fournir un support de raisonnement. Ensuite, un pont sémantique est établi entre les mesures de sécurité, les buts de sécurité et les exigences de sécurité par le développement d'une ontologie de gestion de sécurité orientée-but, appelée GOSMO. La gestion des décisions de sécurité s’appuie sur la réinterprétation du modèle de contrôle d'accès Or-Bac d'un point de vue sécurité-innocuité. Afin d'assurer la cohérence globale des exigences, GOSMO permet de structurer la gestion des évolutions des exigences et leur traçabilité. / Safety is an emergent property of safety critical systems (SCS), including railway systems. This emergent aspect exacerbates their development process and requires a thorough reasoning to reduce hazards. This thesis proposes an ontological approach that integrates safety activities from the early design stages of SCS. This structured framework provides a semantic harmonization between the involved domains, such as safety engineering and Goal Oriented Requirements Engineering (GORE). The business logic integrated in this approach is validated by real rail accident scenarios and a remotely operated task. At first, we proposed a dysfunctional analysis ontology called DAO and based on the high-level ontology UFO. DAO considers the socio-technical and environmental aspects of SCS and integrates the different types of faults and cognitive properties that are respectively related to technical failures and human errors. The DAO conceptual model is expressed in OntoUML and formalized in OWL language in order to provide a reasoning support. Then, a semantic bridge is established between safety measures, safety goals and safety requirements through the development of a goal-oriented security management ontology, called GOSMO. The management of safety decisions is based on the reinterpretation of the Or-Bac access control model from a safety point of view. In order to ensure the overall consistency of requirements, GOSMO allows structuring the management of requirements changes and their traceability