Dissertação (mestrado)—Universidade de Brasília, Instituto de Ciências Exatas, Departamento de Ciência da Computação, 2015. / Submitted by Ana Cristina Barbosa da Silva (annabds@hotmail.com) on 2015-04-27T15:56:38Z No. of bitstreams: 1 2015_DaniloFilgueiraMendonça.pdf: 15442097 bytes, checksum: 3fd8f92515216f0962560e658342894f (MD5) / Approved for entry into archive by Raquel Viana(raquelviana@bce.unb.br) on 2015-05-14T19:16:39Z (GMT) No. of bitstreams: 1 2015_DaniloFilgueiraMendonça.pdf: 15442097 bytes, checksum: 3fd8f92515216f0962560e658342894f (MD5) / Made available in DSpace on 2015-05-14T19:16:39Z (GMT). No. of bitstreams: 1 2015_DaniloFilgueiraMendonça.pdf: 15442097 bytes, checksum: 3fd8f92515216f0962560e658342894f (MD5) / Um contexto de operação estático não é a realidade para muitos sistemas de software atualmente. Variações de contextos impõe novos desafios ao desenvolvimento de sistemas seguros, o que inclui a ativação de falhas apenas em contextos específicos de operação. A engenharia de requisitos orientada a objetivos (GORE) explicita o ‘por quê’ dos requisitos de um sistema, isto é, a intencionalidade por trás de objetivos do sistema e os meios de se atingi-los. Um Runtime goal model (RGM) adiciona especificação de comportamento ao modelo de objetivos convencional, enquanto um Contextual goal model (CGM) especifica efeitos de contextos sobre objetivos, meios e métricas de qualidade. Visando uma verificação formal da dependabilidade de um Contextual-Runtime goal model (CRGM), nesse trabalho é proposta uma nova abordagem para a análise de dependabilidade orientada a objetivos baseada na técnica de verificação probabilística de modelos. Em particular, são definidas regras para a transformação de um CRGM para um modelo cadeia de Makov de tempo discreto (DTMC) com o qual se possa verificar a confiabilidade de se satisfazer um ou mais objetivos do sistema. Adicionalmente, para diminuir o esforço de análise e aumentar a usabilidade de nossa proposta, um gerador automatizado de código CRGM para DTMC foi implementado e integrado com sucesso à ferramenta gráfica que dá suporte às fases de modelagem e análise de objetivos da metodologia TROPOS. A verificação contextual de dependabilidade resultante reflete os requisitos no CRGM, que podem representar: o projeto de um sistema, cuja verificação ocorreria em fase de projetos; ou um sistema em execução, cujo comportamento pode ser verificado em tempo de execução como parte de uma análise de auto-adaptação com foco em dependabilidade. / A static and stable operation environment is not a reality for many systems nowadays. Context variations impose many threats to systems safety, including the activation of context specific failures. Goal-oriented requirements engineering (GORE) brings forward the ‘why’ of system requirements, i.e., the intentionality behind system goals and the means to meet then. A runtime goal model adds a behaviour specification layer to a conventional design goal model, and a contextual goal model specifies the context effects over system goals, means and qualitative metrics. In order to formally verify the dependability of a CRGM, we propose a new goal-oriented dependability analysis based on the probabilistic model checking technique. In particular, we define rules for the transformation of a CRGM into a DTMC model that can be verified for the reliability of the fulfilment of one or more system goals. Also, to mitigate the analysis overhead and increase the usability of our proposal, we have successfully implemented and integrated a CRGM to DTMC code generator to the graphical tool that supports the goal modelling and analysis phases of the TROPOS development methodology. The resulting contextual dependability verification reflects the system requirements in a CRGM, which may represent: a system-to-be, whose verification would take place at design-time; or a running system, whose behaviour can be verified at runtime as part of a self-adaptation analysis targeting dependability.

This thesis investigates the patterning of the Algonquian “central agreement”, i.e. the primary person-number agreement marking, from a diachronic and comparative perspective. The central agreement patterns differ in the two orders: in the conjunct it is fusional and often portmanteau while in the independent it is discontinuous and non-portmanteau. In addition to these differences, there are also some commonalities, such as a pattern in which 1p consistently outranks 2p in both orders. This thesis shows that the differences between the two orders can be taken to reflect variation in the features of the syntactic probe and different morphological spell-out rules, while the shared properties follow from the underlying structure of φ-features. In particular, it is proposed that an additional person feature under the [plural] node causes first person plural to be privileged over second person plural in the competition among vocabulary items in post-syntactic spellout. / October 2016

This thesis presents a preliminary goal oriented modeling of situation awareness in a command and control system. Tropos, an agent oriented software development methodology, has been used for modeling. Use of Tropos allows us to represent, at the knowledge level, the Command and Control actors along with their goals and interdependencies. Through refinement we aim to derive an architectural design for the Situation Awareness component of an Air Defense Command and Control system. This work suggests that goal oriented methodologies can be successfully used in the modeling of the complex systems at the requirement analysis phase. By analyzing dependencies between Command and Control entities, it should be possible to improve the modularity of the Command and Control system architecture.

While several approaches exist for modeling goals and business processes in organizations, the relationships between these two views are often not well defined. This inhibits the effective reuse of available knowledge in models. This thesis aims to address this issue through the introduction of a Goal-oriented Pattern Family (GoPF) framework that helps constructing business process models from organization goals while expanding these goals, establishing traceability relationships between the goal and process views, and improving reusability. Methods for extracting domain knowledge as patterns, which are composed of goal model building blocks, process model building blocks, and their relationships, and for maintaining the patterns over time are also presented. The GoPF framework provides the infrastructure for defining pattern families, i.e., collections of related patterns for particular domains. The foundation of GoPF is formalized as a profile of the User Requirements Notation, a standard modeling language that supports goals, scenarios, and links between them. A method for the use of GoPF is defined and then illustrated through a case study that targets the improvement of patient safety in healthcare organizations. The framework and the extraction/maintenance methods are also validated against another case study involving aviation security in a regulatory environment. The GoPF framework is expected to have a positive impact on the scientific community through the formalization, evolution, and reuse of patterns in domain-specific business domains. From an industrial viewpoint, this framework will also help intermediary organizations (such as consulting firms) who are required to repeatedly create and document goal and process models for other organizations in their business domain.

While several approaches exist for modeling goals and business processes in organizations, the relationships between these two views are often not well defined. This inhibits the effective reuse of available knowledge in models. This thesis aims to address this issue through the introduction of a Goal-oriented Pattern Family (GoPF) framework that helps constructing business process models from organization goals while expanding these goals, establishing traceability relationships between the goal and process views, and improving reusability. Methods for extracting domain knowledge as patterns, which are composed of goal model building blocks, process model building blocks, and their relationships, and for maintaining the patterns over time are also presented. The GoPF framework provides the infrastructure for defining pattern families, i.e., collections of related patterns for particular domains. The foundation of GoPF is formalized as a profile of the User Requirements Notation, a standard modeling language that supports goals, scenarios, and links between them. A method for the use of GoPF is defined and then illustrated through a case study that targets the improvement of patient safety in healthcare organizations. The framework and the extraction/maintenance methods are also validated against another case study involving aviation security in a regulatory environment. The GoPF framework is expected to have a positive impact on the scientific community through the formalization, evolution, and reuse of patterns in domain-specific business domains. From an industrial viewpoint, this framework will also help intermediary organizations (such as consulting firms) who are required to repeatedly create and document goal and process models for other organizations in their business domain.

While several approaches exist for modeling goals and business processes in organizations, the relationships between these two views are often not well defined. This inhibits the effective reuse of available knowledge in models. This thesis aims to address this issue through the introduction of a Goal-oriented Pattern Family (GoPF) framework that helps constructing business process models from organization goals while expanding these goals, establishing traceability relationships between the goal and process views, and improving reusability. Methods for extracting domain knowledge as patterns, which are composed of goal model building blocks, process model building blocks, and their relationships, and for maintaining the patterns over time are also presented. The GoPF framework provides the infrastructure for defining pattern families, i.e., collections of related patterns for particular domains. The foundation of GoPF is formalized as a profile of the User Requirements Notation, a standard modeling language that supports goals, scenarios, and links between them. A method for the use of GoPF is defined and then illustrated through a case study that targets the improvement of patient safety in healthcare organizations. The framework and the extraction/maintenance methods are also validated against another case study involving aviation security in a regulatory environment. The GoPF framework is expected to have a positive impact on the scientific community through the formalization, evolution, and reuse of patterns in domain-specific business domains. From an industrial viewpoint, this framework will also help intermediary organizations (such as consulting firms) who are required to repeatedly create and document goal and process models for other organizations in their business domain.

<p>This thesis emphasizes on surveying the state-of-the-art in software requirements specification with a focus on, autonomic, self-adapting software systems. Since various requirements are brought forward accord with environments, modeling requirements for adaptive software systems may be changed at run-time. Nowadays, Keep All Objectives Satisfied (KAOS) is an effective method to build goal model. Various manipulations, such as change, remove, active and de-active goals, appear new goals, could mediate conflicts among goals in adaptive software system. At specification time, specifications of event sequences to be monitored are generated from requirements specification.</p>

This thesis emphasizes on surveying the state-of-the-art in software requirements specification with a focus on, autonomic, self-adapting software systems. Since various requirements are brought forward accord with environments, modeling requirements for adaptive software systems may be changed at run-time. Nowadays, Keep All Objectives Satisfied (KAOS) is an effective method to build goal model. Various manipulations, such as change, remove, active and de-active goals, appear new goals, could mediate conflicts among goals in adaptive software system. At specification time, specifications of event sequences to be monitored are generated from requirements specification.

[ES] La ingeniería de requisitos es fundamental para el desarrollo de software ya que se encarga de obtener, analizar, especificar, validar y gestionar los requisitos del software. La definición incorrecta de los requisitos puede causar retrasos y sobrecostes. Además, es importante identificar los requisitos más relevantes y cuáles deberían implementarse primero considerando las limitaciones de recursos y teniendo cuenta distintos puntos de vista. Los modelos de objetivos destacan para obtener y modelar requisitos de alto nivel ya que se centran en los objetivos de los stakeholders y las relaciones entre ellos. Estos modelos ayudan a identificar los requisitos ya que capturan las motivaciones e intenciones de los stakeholders con respecto al software. Si bien existen diversas técnicas de análisis de modelos de objetivos, la mayoría se centran en la satisfacibilidad, identificando la compatibilidad entre objetivos gestionando así un problema importante de la ingeniería de requisitos. Las técnicas de modelos de objetivos adoptan un enfoque neutral con respecto al valor, considerando todos los objetivos igual de importantes. Si bien existen técnicas que utilizan la priorización de importancia junto con satisfacción, el enfoque principal es la satisfacción, haciendo que la importancia para los stakeholders quede relegado a un segundo plano. En consecuencia, los resultados pueden ayudar con la compatibilidad de los requisitos, pero no a identificar los más relevantes. El objetivo de esta tesis doctoral es definir y evaluar una técnica de análisis de modelos de objetivos que priorice los objetivos de un modelo de objetivos en función del valor que aportan según las preferencias de los stakeholders y las relaciones entre los objetivos, permitiendo identificar qué objetivos aportan más valor a los stakeholders. Esta técnica no pretende sustituir a las ya existentes sino complementarlas ya que no identifica la compatibilidad entre los objetivos sino cuáles proporcionan más valor. La técnica propuesta (VeGAn) hace uso de la lógica difusa para tratar la incertidumbre en la asignación de importancia relativa y de una técnica de toma de decisiones multicriterio ampliamente utilizada en la industria (TOPSIS) para calcular el valor, la cual se ve enriquecida por una técnica de propagación sistemática que provee información adicional sobre la interacción entre los objetivos en base a las relaciones del modelo de objetivos. Además, se ha desarrollado una herramienta que da soporte tecnológico a VeGAn automatizando la técnica y proporcionado una serie de funciones: i) importar modelos de objetivos de otras herramientas; ii) realizar validaciones sobre los datos introducidos; iii) ordenar los resultados según varios; y iv) almacenar información sobre distintos análisis de un modelo de objetivos para realizar comparaciones del valor obtenido. VeGAn se evaluó mediante un estudio de caso y una familia de experimentos. El estudio de caso verificó la aplicabilidad sistemática y consistente de la técnica, obteniendo retroalimentación para mejoras. La familia de experimentos involucró a 172 sujetos, estudiantes de ingeniería en informática y másteres en ingeniería del software donde se analizó la precisión de la priorización, tiempo de priorización, satisfacción percibida, facilidad de uso percibida, utilidad percibida e intención de uso percibida de los participantes aplicando VeGAn en oposición a GRL-Quant. El análisis estadístico indicó que ambas técnicas son similares salvo para la satisfacción percibida a favor de VeGAn. Esta tesis doctoral contribuye al ámbito de la ingeniería de requisitos y en especial a la elicitación temprana de requisitos mediante el uso de modelos de objetivos proporcionando una técnica de análisis que complementa a las ya existentes ayudando a priorizar aquellos objetivos que proporcionan más valor teniendo en cuenta el punto de vista de diversos stakeholders. / [CA] L'enginyeria de requisits és fonamental per al desenvolupament de programari ja que s'encarrega d'obtindre, analitzar, especificar, validar i gestionar els requisits del programari. La definició incorrecta dels requisits pot causar retards i sobrecostos. A més, és important identificar els requisits més rellevants i quins haurien d'implementar-se primer considerant les limitacions de recursos i tenint compte diferents punts de vista. Els models d'objectius destaquen per a obtindre i modelar requisits d'alt nivell ja que se centren en els objectius dels stakeholders i les relacions entre ells. Estos models ajuden a identificar els requisits ja que capturen les motivacions i intencions dels stakeholders respecte al programari. Si bé existixen diverses tècniques d'anàlisis de models d'objectius, la majoria se centren en la satisfacibilitat, identificant la compatibilitat entre objectius gestionant així un problema important de l'enginyeria de requisits. Les tècniques de models d'objectius adopten un enfocament neutral respecte al valor, considerant tots els objectius igual d'importants. Si bé existixen tècniques que utilitzen la priorització d'importància juntament amb satisfacció, l'enfocament principal és la satisfacció, fent que la importància per als stakeholders quede relegat a un segon pla. En conseqüència, els resultats poden ajudar amb la compatibilitat dels requisits, però no a identificar els més rellevants. L'objectiu d'esta tesi doctoral és definir i avaluar una tècnica d'anàlisi de models d'objectius que prioritze els objectius d'un model d'objectius en funció del valor que aporten segons les preferències dels stakeholders i les relacions entre els objectius, permetent identificar quins objectius aporten més valor als stakeholders. Esta tècnica no pretén substituir a les ja existents sinó complementar-les ja que no identifica la compatibilitat entre els objectius sinó quins proporcionen més valor. La tècnica proposada (VeGAn) fa ús de la lògica difusa per a tractar la incertesa en l'assignació d'importància relativa i d'una tècnica de presa de decisions multicriteri àmpliament utilitzada en la indústria (TOPSIS) per a calcular el valor, la qual es veu enriquida per una tècnica de propagació sistemàtica que proveïx informació addicional sobre la interacció entre els objectius sobre la base de les relacions del model d'objectius. A més, s'ha desenvolupat una ferramenta que dona suport tecnològic a VeGAn automatitzant la tècnica i proporcionat una sèrie de funcions: i) importar models d'objectius d'altres ferramentes; ii) realitzar validacions sobre les dades introduïdes; iii) ordenar els resultats segons varis; i iv) emmagatzemar informació sobre diferents anàlisis d'un model d'objectius per a realitzar comparacions del valor obtingut. VeGAn es va avaluar mitjançant un estudi de cas i una família d'experiments. L'estudi de cas va verificar l'aplicabilitat sistemàtica i consistent de la tècnica, obtenint retroalimentació per a millores. La família d'experiments va involucrar a 172 subjectes, estudiants d'enginyeria en informàtica i màsters en enginyeria del programari on es va analitzar la precisió de la priorització, temps de priorització, satisfacció percebuda, facilitat d'ús percebuda, utilitat percebuda i intenció d'ús percebuda dels participants aplicant VeGAn en oposició a GRL-Quant. L'anàlisi estadística va indicar que totes dues tècniques són similars excepte per a la satisfacció percebuda a favor de VeGAn. Esta tesi doctoral contribuïx a l'àmbit de l'enginyeria de requisits i especialment a la elicitació primerenca de requisits mitjançant l'ús de models d'objectius proporcionant una tècnica d'anàlisi que complementa a les ja existents ajudant a prioritzar aquells objectius que proporcionen més valor tenint en compte el punt de vista de diversos stakeholders. / [EN] Requirements engineering is fundamental to software development as it is responsible for eliciting, analysing, specifying, validating and managing software requirements. Incorrect definition of requirements can cause delays and cost overruns. Furthermore, it is important to identify the most relevant requirements and which ones should be implemented first considering resource constraints and taking into account different points of view. Goal models stand out at eliciting and modelling high-level requirements as they focus on stakeholder goals and the relationships between them. These models help identify requirements by capturing the motivations and intentions of stakeholders with respect to the software. Although there are a variety of goal-oriented analysis techniques, most focus on satisfiability, identifying compatibility between goals and thus managing an important requirements engineering problem. Goal-oriented analysis techniques adopt a value-neutral approach, considering all goals equally important. While there are techniques that use prioritisation of importance along with satisfaction, the main focus is on satisfaction, pushing importance to stakeholders into the background. As a result, the results may help with the compatibility of requirements, but not in identifying the most relevant ones. The objective of this PhD thesis is to define and evaluate a goal-oriented analysis technique that prioritises the goals of a goal model according to the value they provide according to the preferences of stakeholders and the relationships between the goals, making it possible to identify which goals provide the most value to stakeholders. This technique is not intended to replace existing techniques but to complement them, as it does not identify compatibility between goals but rather which ones provide the most value. The proposed technique (VeGAn) makes use of fuzzy logic to deal with uncertainty in assigning relative importance and a widely used multi-criteria decision-making technique in industry (TOPSIS) to calculate value, which is enriched by a systematic propagation technique that provides additional information on the interaction between goals based on the relationships of the goal model. In addition, a tool has been developed to provide technological support to VeGAn by automating the technique and providing a series of functions: i) importing goal models from other tools; ii) performing validations on the data entered; iii) sorting the results according to several; and iv) storing information on different analyses of a goal model in order to make comparisons of the value obtained. VeGAn was evaluated through a case study and a family of experiments. The case study verified the systematic and consistent applicability of the technique, obtaining feedback for improvements. The family of experiments involved 172 subjects, undergraduate students in computer engineering and masters in software engineering. The aim of the family was to analyse the participants' prioritisation accuracy, prioritisation time, perceived satisfaction, perceived ease of use, perceived usefulness, and perceived intention to use by applying VeGAn as opposed to GRL-Quant, a widely used goal modelling analysis method. Statistical analysis indicated that both techniques are similar except for perceived satisfaction in favour of VeGAn. This PhD thesis contributes to the field of requirements engineering and in particular to the early elicitation of requirements through the use of goal models by providing a goal-oriented analysis technique that complements the existing ones by helping to prioritise those goals that provide the most value taking into account the point of view of various stakeholders. / Cano Genovés, C. (2024). Análisis de modelos de objetivos basado en valor mediante lógica difusa y toma de decisiones multicriterio [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/207550

The growing need for computer software in different sectors of activity, (health, agriculture, industries, education, aeronautic, science and telecommunication) together with the increasing reliance of the society as a whole on information technology, is placing a heavy and fast growing demand on complex and high quality software systems. In this regard, the anticipation has been on non-functional requirements (NFRs) engineering and formal methods. Despite their common objective, these techniques have in most cases evolved separately. NFRs engineering proceeds firstly, by deriving measures to evaluate the quality of the constructed software (product-oriented approach), and secondarily by improving the engineering process (process-oriented approach). With the ability to combine the analysis of both functional and non-functional requirements, Goal-Oriented Requirements Engineering (GORE) approaches have become de facto leading requirements engineering methods. They propose through refinement/operationalisation, means to satisfy NFRs encoded in softgoals at an early phase of software development. On the other side, formal methods have kept, so far, their promise to eliminate errors in software artefacts to produce high quality software products and are therefore particularly solicited for safety and mission critical systems for which a single error may cause great loss including human life. This thesis introduces the concept of Complementary Non-functional action (CNF-action) to extend the analysis and development of NFRs beyond the traditional goals/softgoals analysis, based on refinement/operationalisation, and to propagate the influence of NFRs to other software construction phases. Mechanisms are also developed to integrate the formal technique Z/Object-Z into the standardised User Requirements Notation (URN) to formalise GRL models describing functional and non-functional requirements, to propagate CNF-actions of the formalised NFRs to UCMs maps, to facilitate URN construction process and the quality of URN models. / School of Computing / D. Phil (Computer Science)