No description available.

Made available in DSpace on 2014-07-29T14:57:46Z (GMT). No. of bitstreams: 1 Dissert parte 1 Glauber Boff.pdf: 3627301 bytes, checksum: a9c5bc5ed01e465f7ee93a59c0da1439 (MD5) Previous issue date: 2010-04-12 / The development of Information Systems (IS) should be performed so that changes in the client s business, which may be common, may be reflected in the system, without reducing its consistency and reliability. To achieve these features, portability and maintainability, which are challenges for Software Engineering, are very important. There are two essential requirements which, when present in the development process, make it easier to deal with the system complexity. The first requirement states that it should be possible to describe the IS business rules (BR) using a high level language, generating a single, implementation independent model. The second requirement demands that it should be possible to generate software source code from the abstract BR model, and to integrate this code to the IS. This work presents an approach to meet these requirements. This approach covers the specification, modeling, implementation and evaluation of business rules in software to facilitate the development and maintenance of complex IS. Instead of hard-coding the BR into the applications, this approach adopts OCL as a platform independent high-level language to define a single BR model for the IS. To store the rules safely throughout the IS life cycle, a centralized repository in the database was developed, which enables control over all BR versions. During implementation, rules in OCL are automatically converted to SQL and stored in the database as stored procedures for further evaluation by a specific component of the application. This approach facilitates the maintenance of the IS by the centralized and high-level description of all BR. To demonstrate the feasibility of the proposed approach, an empirical validation was performed in the context of a research project funded by CNPq, which aims to optimize the production of beef cattle. / O desenvolvimento de Sistemas de Informação (SI) deve ser realizado de forma que as mudanças no negócio do cliente, que podem ser frequentes, possam ser refletidas no sistema, sem que ele perca sua consistência e confiabilidade. Para se atingir essas características, a portabilidade e manutenibilidade, que são desafios para a Engenharia de Software, são importantes. Há dois requisitos essenciais que, quando presentes no processo de desenvolvimento, facilitam o tratamento da complexidade do sistema. O primeiro requisito diz que deveria ser possível descrever as regras de negócio (RN) usando uma linguagem de alto nível, gerando um modelo único e independente de implementação. O segundo requisito demanda que deveria ser possível gerar código fonte a partir do modelo abstrato de RN e integrá-lo ao SI. Neste trabalho é apresentada uma abordagem para atender a esses requisitos. Esta abordagem abrange especificação, modelagem, implementação e avaliação de regras de negócio em software de forma a facilitar o desenvolvimento e a manutenção de SI complexos. Em vez de codificar as RN diretamente nas aplicações, adotou-se a linguagem OCL, uma linguagem abstrata e independente de plataforma, para definir um modelo único de RN para o SI. Para armazenar as regras ao longo do ciclo de vida do SI de forma segura e de fácil acesso foi criado um repositório centralizado no banco de dados, que permite realizar o controle de todas as versões das RN. Na implementação, regras em OCL são convertidas automaticamente para SQL e armazenadas em banco de dados como procedimentos armazenados para posterior avaliação por um componente específico da aplicação. Esta abordagem facilita a manutenção do SI pela descrição centralizada e em alto nível de todas as RN de um sistema. Para demonstrar a viabilidade da abordagem proposta, uma validação empírica foi realizada no contexto de um projeto de pesquisa apoiado pelo CNPq que visa a otimização da produção de gado de corte.

Les langages dédiés de modélisation (DSMLs) sont de plus en plus utilisés dans les phases amont du développement des systèmes complexes, en particulier pour les systèmes critiques embarqués. L’objectif est de pouvoir raisonner très tôt dans le développement sur ces modèles et, notamment, de conduire des activités de vérification et validation (V and V). Une technique très utilisée est la vérification des modèles comportementaux par exploration exhaustive (model-checking) en utilisant une sémantique de traduction pour construire un modèle formel à partir des modèles métiers pour réutiliser les outils performants disponibles pour les modèles formels. Définir cette sémantique de traduction, exprimer les propriétés formelles à vérifier et analyser les résultats nécessite une expertise dans les méthodes formelles qui freine leur adoption et peut rebuter les concepteurs. Il est donc nécessaire de construire pour chaque DSML, une chaîne d’outils qui masque les aspects formels aux utilisateurs. L’objectif de cette thèse est de faciliter le développement de telles chaînes de vérification. Notre contribution inclut 1) l’expression des propriétés comportementales au niveau métier en s’appuyant sur TOCL (Temporal Object Constraint Language), une extension temporelle du langage OCL; 2) la transformation automatique de ces propriétés en propriétés formelles en réutilisant les éléments clés de la sémantique de traduction; 3) la remontée des résultats de vérification grâce à une transformation d’ordre supérieur et un langage de description de correspondance entre le domaine métier et le domaine formel et 4) le processus associé de mise en oeuvre. Notre approche a été validée par l’expérimentation sur un sous-ensemble du langage de modélisation de processus de développement SPEM, et sur le langage de commande d’automates programmables Ladder Diagram, ainsi que par l’intégration d’un langage formel intermédiaire (FIACRE) dans la chaîne outillée de vérification. Ce dernier point permet de réduire l’écart sémantique entre les DSMLs et les domaines formels. / Domain specific Modeling Languages (DSMLs) are increasingly used at the early phases in the development of complex systems, in particular, for safety critical systems. The goal is to be able to reason early in the development on these models and, in particular, to fulfill verification and validation activities (V and V). A widely used technique is the exhaustive behavioral model verification using model-checking by providing a translational semantics to build a formal model from DSML conforming models in order to reuse powerful tools available for this formal domain. Defining a translational semantics, expressing formal properties to be assessed and analysing such verification results require such an expertise in formal methods that it restricts their adoption and may discourage the designers. It is thus necessary to build for each DSML, a toolchain which hides formal aspects for DSML end-users. The goal of this thesis consists in easing the development of such verification toolchains. Our contribution includes 1) expressing behavioral properties in the DSML level by relying on TOCL (Temporal Object Constraint Language), a temporal extension of OCL; 2) An automated transformation of these properties on formal properties while reusing the key elements of the translational semantics; 3) the feedback of verification results thanks to a higher-order transformation and a language which defines mappings between DSML and formal levels; 4) the associated process implementation. Our approach was validated by the experimentation on a subset of the development process modeling language SPEM, and on Ladder Diagram language used to specify programmable logic controllers (PLCs), and by the integration of a formal intermediate language (FIACRE) in the verification toolchain. This last point allows to reduce the semantic gap between DSMLs and formal domains.