This dissertation addresses two important problems in reusing intellectual properties (IPs) in the form of reusable design or verification components. The first problem is associated with fast and effective integration of reusable design components into a System-on-chip (SoC), so faster design turn-around time can be achieved, leading to faster time-to-market. The second problem has the same goals of faster product design cycle, but emphasizes on verification model reuse, rather than design component reuse. It specifically addresses reuse of reusable verification IPs to enable a "write once, use many times" verification strategy. This dissertation is accordingly divided into part I and part II which are related but describe the two problems and our solutions to them. These two related but distinctive problems faced by system design companies have been tackled through a unique approach which hither-to-fore only have been used in the software engineering domain. This approach is called metamodeling, which allows creating customized meta-language to describe the syntax and semantics for a modeling domain. It provides a way to create, transform and analyze domain specific languages, which are themselves described by metamodels, and the transformation and processing of models in such languages are also described by metamodels. This makes machine based interpretation and translation from these models an easier and formal task. In part I, we consider the problem of rapid system-level model integration of existing reusable components such that (i) the required architecture of the SoC can be expressed formally, (ii) automatic selection of components from an IP library to match the need of the system being integrated can be done, (iii) integrability of the components is provable, or checkable automatically, and (iv) structural and behavioral type systems for each component can be utilized through inferencing and matching techniques to ensure their compatibility. Our solutions include a component composition language, algorithms for component selection, type matching and inferencing algorithms, temporal property based behavioral typing, and finally a software system on top of an existing metamodeling environment. In part II, we use the same metamodeling environment to create a framework for modeling generative verification IPs. Our main contributions relate to INTEL's microprocessor verification environment, and our solution spans various abstraction levels (System, architectural, and microarchitecture) to perform verification. We provide a unified language that can be used to model verification IPs at all abstraction levels, and verification collaterals such as testbenches, simulators, and coverage monitors can be generated from these models, thereby enhancing reuse in verification. / Ph. D.

System-on-Chip (SoC) and other complex distributed hardware/software systems contain heterogeneous components whose behavior are best captured by different models of computations (MoCs). As a result, any system design framework for such systems requires the capability to express heterogeneous MoCs. Although a number of system level design languages (SLDL)s and frameworks have proliferated over the last few years, most of them are lacking in multiple ways. Some of the SLDLs and system design frameworks we have worked with are SpecC, Ptolemy II, SystemC-H, etc. From our analysis of these, we identify their following shortcomings: First, their dependence on specific programming language artifacts (Java or C/C++) make them less amenable to formal analysis. Second, the refinement strategies proposed in the design flows based on these languages lack formal semantics underpinnings making it difficult to prove that refinements preserve correctness, and third, none of the available SLDLs are easily customizable by users. In our work, we address these problems as follows: To alleviate the first problem, we follow Axel Jantsch's paradigm of function-based semantic definitions of MoCs and formulate a functional programming framework called SML-Sys. We illustrate through a number of examples how to model heterogenous computing systems using SML-Sys. Our framework provides for formal reasoning due to its formal semantic underpinning inherited from SML's precise denotational semantics. To handle the second problem and apply refinement strategies at a higher-level, we propose a refinement methodology and provide a semantics preserving transformation library within our framework. To address the third shortcoming, we have developed EWD, which allows users to customize MoC-specific visual modeling syntax defined as a metamodel. EWD is developed using a metamodeling framework GME (Generic Modeling Environment). It allows for automatic design-time syntactic and semantic checks on the models for conformance to their metamodel. Modeling in EWD facilitates saving the model in an XML-based interoperability language (IML) we defined for this purpose. The IML format is in turn automatically translated into Standard ML, or Haskell models. These may then be executed and analyzed either by our existing model analysis tools SMLSys, or the ForSyDe environment. We also generate SMV-based template from the XML representation to obtain verification models. / Master of Science

Le développement de systèmes logiciels implique l'intégration de différents langages pour modéliser l'organisation et le comportement des composants d'une application. De plus, s'il existe des modèles de processus décrivant les différentes activités pour passer d'une spécification à une réalisation, il n'existe pas de processus général expliquant comment relier de façon rationnelle langages et activités. Par ailleurs, l'Ingénierie Dirigée par les Modèles propose des concepts et des outils pour spécifier et combiner différents langages; ceci en introduisant les concepts de méta­modèles comme moyen pour spécifier des langages, et de transformation de modèles pour les mettre en relation. Un exemple de métamodèle est donné par le standard SPEM (Software Process Engineering Metamodel) dédié à la modélisation de processus de développement logiciel.Dans ce contexte, la thèse propose de tirer profit des éléments précédents pour élaborer des processus de développements spécifiques et orientés modèles conformes à un schéma conceptuel dérivé du schéma de SPEM dans lequel des activités d'un processus peuvent exploiter des métamodèles et des transformations. Un guide méthodologique pour aborder l'ingénierie de processus spécifiques complète le schéma proposé. Les intérêts de la proposition résident alors dans une meilleure capitalisation des connaissances pour un domaine ainsi qu'une réduction des efforts de développement. Finalement, la thèse propose d'élaborer de tels processus spécifiques pour des applications logicielles relevant du domaine des Systèmes à Évènements Discrets. / Software development usually requires the usage and integration of different modeling languages for modeling the behavior and the organization of the application's components. There are many process models describing the different activities of software development but there is no generic process that explains how to relate these languages to the activities. In this context, Model Driven Engineering offers concepts and tools to specify and to combine different languages with meta­ models for specifying languages and model transformations to put together various meta-models. An example of such a meta-model is given by SPEM (Software Process Engineering Metamodel) which serves as a modeling language for describing software development processes.The thesis proposes to take advantage of the preceding concepts to define specific model-drivendevelopment processes; this by defining a conceptual scheme, derived from SPEM, in which the activities of a process use meta-models and transformations. From this scheme, the components of a specific process are identified what leads to a methodological approach for the engineering of domain specific processes. Thus, the major benefits of the approach are: knowledge capitalization and a reduction of development efforts. Finally, the thesis applies this proposition to derive aspecific software engineering process dedicated to the design of Discrete Event Systems.

This thesis consists of three chapters, covering topics in both the design and modeling aspects of computer experiments as well as their engineering applications. The first chapter systematically develops a new class of space-filling designs for computer experiments by splitting two-level factorial designs into multiple layers. The new design is easy to generate, and our numerical study shows that it can have better space-filling properties than the optimal Latin hypercube design. The second chapter proposes a novel modeling approach for approximating computationally expensive functions that are not second-order stationary. The new model is a composite of two Gaussian processes, where the first one captures the smooth global trend and the second one models local details. The new predictor also incorporates a flexible variance model, which makes it more capable of approximating surfaces with varying volatility. The third chapter is devoted to a two-stage sequential strategy which integrates analytical models with finite element simulations for a micromachining process.

A persistent challenge in the design of composite materials is the ability to fabricate materials that simultaneously display high stiffness and high loss factors for the creation of structural elements capable of passively suppressing vibro-acoustic energy. Relevant recent research has shown that it is possible to produce composite materials whose macroscopic mechanical stiffness and loss properties surpass those of conventional composites through the addition of trace amounts of materials displaying negative stiffness (NS) induced by phase transformation [R. S. Lakes, et al., Nature, 410, pp. 565-567, (2001)]. The present work investigates the ability to elicit NS behavior without employing physical phenomena such as inherent nonlinear material behavior (e.g., phase change or plastic deformation) or dynamic effects, but rather the controlled buckling of small-scale structural elements, metamaterials, embedded in a continuous viscoelastic matrix. To illustrate the effect of these buckled elements, a nonlinear hierarchical multiscale material model is derived which estimates the macroscopic stiffness and loss of a composite material containing pre-strained microscale structured inclusions. The nonlinear multiscale model is then utilized in a set-based hierarchical design approach to explore the design space over a wide range of inclusion geometries. Finally, prototype NS inclusions are fabricated using an additive manufacturing technique and tested to determine quasi-static inclusion stiffness which is compared with analytical predictions. / text

Les méthodes traditionnelles de dimensionnement à la fatigue s’appuient sur l’utilisation de coefficients dits de “sécurité” dans le but d’assurer l’intégrité de la structure en couvrant les incertitudes inhérentes à la fatigue. Ces méthodes de l’ingénieur ont le mérite d’être simples d’application et de donner des solutions heureusement satisfaisantes du point de vue de la sécurité. Toutefois, elles ne permettent pas au concepteur de connaître la véritable marge de sécurité de la structure et l’influence des différents paramètres de conception sur la fiabilité. Les approches probabilistes sont envisagées dans cette thèse afin d’acquérir ces informations essentielles pour un dimensionnement optimal de la structure vis-à-vis de la fatigue. Une approche générale pour l’analyse probabiliste en fatigue est proposée dans ce manuscrit. Elle s’appuie sur la modélisation des incertitudes (chargement, propriétés du matériau, géométrie, courbe de fatigue) et vise à quantifier le niveau de fiabilité de la structure étudiée pour un scénario de défaillance en fatigue. Les méthodes classiques de fiabilité nécessitent un nombre important d’évaluations du modèle mécanique de la structure et ne sont donc pas envisageables lorsque le calcul du modèle est coûteux en temps. Une famille de méthodes appelée AK-RM (Active learning and Kriging-based Reliability Methods) est précisément proposée dans ces travaux de thèse afin de résoudre le problème de fiabilité avec un minimum d’évaluations du modèle mécanique. L’approche générale est appliquée à deux cas-tests fournis par SNECMA dans le cadre du projet ANR APPRoFi. / Traditional procedures for designing structures against fatigue are grounded upon the use of so-called safety factors in an attempt to ensure structural integrity while masking the uncertainties inherent to fatigue. These engineering methods are simple to use and fortunately, they give satisfactory solutions with regard to safety. However, they do not provide the designer with the structure’s safety margin as well as the influence of each design parameter on reliability. Probabilistic approaches are considered in this thesis in order to acquire this information, which is essential for an optimal design against fatigue. A general approach for probabilistic analysis in fatigue is proposed in this manuscript. It relies on the modelling of the uncertainties (load, material properties, geometry, and fatigue curve), and aims at assessing the reliability level of the studied structure in the case of a fatigue failure scenario. Classical reliability methods require a large number of calls to the mechanical model of the structure and are thus not applicable when the model evaluation is time-demanding. A family of methods named AK-RM (Active learning and Kriging-based Reliability methods) is proposed in this research work in order to solve the reliability problem with a minimum number of mechanical model evaluations. The general approach is applied to two case studies submitted by SNECMA in the frame of the ANR project APPRoFi.

Orientador: Manuel de Jesus Mendes / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-08T10:49:25Z (GMT). No. of bitstreams: 1 Kamada_Aqueo_D.pdf: 3886405 bytes, checksum: 8e4f2791cf72214eac7fe3d18c8bc794 (MD5) Previous issue date: 2006 / Resumo: No atual mundo globalizado, com interações fortemente baseadas na Web, as relações entre pessoas, empresas e aos orgãos de governo estão sujeitas a mudanças cada vez mais rápidas. Esse cenário torna ainda mais crítico o já antigo problema de manutenção de sistemas, uma vez que as atualizações precisam ser realizadas para contemplar novos modelos de negócio, no contexto destas mudanças rápidas. Esta situação justifica a necessidade de uma abordagem que consiga capturar as mudanças nos negóios e rapidamente implementá-las nos sistemas computacionais. Considera-se que existem porções da lógica de negócio que são bastante voláteis e suscetíveis às mudanças e outras porções que são bastante estáveis e muito pouco suscetíveis às mudanças. Neste contexto, propõe-se um modelo para desenvolvimento rápido e execussão de serviços, baseado em regras de negócio. Neste modelo as porções voláteis são externalizadas como regras de negócio e as porções estáveis como serviços. A prova de conceito desta abordagem traduz regras modeladas em linguagem natural para regras no modelo da máquina de regras instanciada, que invoca os serviços / Abstract: In the current globalized world, characterized by strong Web based interactions, relationships between people, companies and government organizations are subject to changes that occur faster and faster. This scenario makes the old problem of system maintenance even more critical, as the updating of these systems needs to contemplate new business models in the context of these fast changes. This situation justifies the need for an approach that can capture business changes and quickly implement them into computational systems. It is considered that some business logic portions are quite volatile and susceptible to the changes and other portions are quite stable and less susceptible to the changes. In this context, a model for fast development and execution of services based on business rules is proposed. In this model the volatile portions are externalized as business rules and the stable portions as services. The proof of concept of this approach translates rules which are modeled in a natural language to rules in the model of the instantiated rule engine that invokes the services / Doutorado / Engenharia de Computação / Doutor em Engenharia Elétrica

Made available in DSpace on 2016-08-17T14:53:22Z (GMT). No. of bitstreams: 1 dissertacao de Amelia Acacia.pdf: 4459088 bytes, checksum: a31339e9163af53009c50173988fd82f (MD5) Previous issue date: 2012-06-20 / Agile principles can be used appropriately in management of generic projects, including outside the object of study of computer science and allow a quick adaptation to new domains. In this work we propose a management process model of Distance Learning Courses (DLC) within the context of the Open University of Brazil (UAB). This process model aims to manage the flow of activities in the construction and management of a DLC. As a complement of the research work, a prototype tool resulting from process modeling was developed, whose application favored the validation of the proposed model performed by an illustrative example. The prototype was developed using Eclipse Modeling Framework and Java Server Pages, addition to JPA specification associated with the Hibernate framework for persistence of data. For these goals to be achieved, were performed activities such as UAB operating environment characterization, process tailoring and construction metamodels for each phase of the model (called the Agile-UAB). As academic contributions of this study, we detach: the generation of a domain specific language, modeling Kanban agile method adapted to the UAB and the construction of a project management tool DL. We still consider as the real benefit of this research the improvement of the control flow of activities involved in the construction of a distance learning course within the context of UAB. / Os princípios ágeis podem ser usados apropriadamente no gerenciamento de projetos gerais, inclusive fora do objeto de estudo da informática, e permitem uma rápida adaptação às novas realidades. Neste trabalho, propomos um modelo de processo baseado em uma abordagem ágil para suportar a gestão de cursos de Ensino à Distância (EaD) dentro do contexto da Universidade Aberta do Brasil (UAB). Este modelo de processo tem o objetivo de controlar e efetivar o fluxo de atividades na construção e gestão de um curso EaD. Como complemento do trabalho de pesquisa, o protótipo de uma ferramenta resultante da modelagem do processo foi desenvolvido, cuja aplicação favoreceu os testes para validação do modelo de gestão proposto, realizado através de um exemplo ilustrativo. O protótipo foi desenvolvido utilizando o Eclipse Modeling Framework (EMF) e a linguagem Java Server Pages (JSP), além da especificação JPA associada ao framework Hibernate para persistência dos dados. Para que esses objetivos fossem alcançados, foram realizadas atividades como a caracterização do ambiente UAB, tailoring de processo e construção de metamodelos das fases que compõem o modelo de gestão (denominado de UAB-Ágil). Relacionamos como contribuições acadêmicas deste estudo: a geração de uma linguagem especifica de domínio, a modelagem do método ágil Kanban adaptado ao contexto da UAB e a construção de uma ferramenta de gestão de projetos EaD. Consideramos ainda como real benefício desta pesquisa a melhoria no controle do fluxo das atividades envolvidas no processo de construção de um curso à distância dentro do contexto da UAB.

The need for change of direction in Software Engineering has been suggested in severalresources [Power of Models]. In recent years many organizations have focused on ModelDriven Engineering (MDE) as an approach to application design and implementation.Model Driven Architecture (MDA) was introduced by Object Management Group (OMG) in2001 in support of MDE. Models are the basic elements in MDE. The focus on MDE is onthe concept of “Everything is a model”. So far several languages, tools and platforms havebeen created for MDE.In particular, models can be developed in a distributed environment, thus once they mergetogether, conflicts and inconsistencies should be detected in a convenient way to bereconciled (automatically by software or manually by developers). This project is based onprevious works that define difference and conflict metamodels of a source model. In thisreport, we introduce the benefits of versioning systems in MDE framework. A conflictmetamodel that is generated from the input metamodel and the architecture for detectingconflicts are represented. The proposed approach is metamodel independent that meansconflict metalmodel is created for any input model which conforms to Kernel Meta MetaModel (KM3). The concept of used conflict model can be also changed regardless ofmodels. The extended idea for conflicts detection, the presented architecture for modelcode management and the tools that can detect conflicts of concurrent models can help toimprove MDE in model code management. For this report, some implementation in Eclipseplatform has been rendered while some future works are suggested. / University of L'Aquila, Project group in Moedling with Alfonso Pierantonio

Dans le cadre de l’informatique décisionnelle, la programmation physico-financière doit permettre à des acteurs d’une collectivité provenant de divers domaines de faire converger leurs problématiques vers un objectif commun. L’une des principales difficultés de la modélisation d’une programmation physico-financière est que chaque acteur exprime ses problématiques dans des échelles de temps différentes. Dans cette thèse CIFRE, un métamodèle de calcul en temps continu appliqué à la programmation physico-financière est proposé afin de permettre aux acteurs de regrouper leurs visions dans un modèle unique, tout en se plaçant sur des échelles de temps différentes. La modélisation continue développée est confrontée à la modélisation discrète (représentative de l’état de l’art) au travers d’un cas d’étude, montrant les avancées de la première vis-à-vis de la seconde. Ce métamodèle innovant a été implémenté au sein de la société MGDIS, dans le cadre d’une convention CIFRE, à l’aide d’une architecture orientée service. Cette architecture se base sur un style innovant conçu dans cette thèse afin de maximiser la capacité à paralléliser l’évaluation des modèles. La solution développée dans cette thèse a été conçue pour permettre la programmation physico-financière de gros volumes de données à l’échelle réelle. Elle a été validée sur un cas d’étude et répond aux exigences exprimées par les experts de la modélisation de programmation physico-financière de MGDIS qui ont émis un avis positif quant à l’applicabilité de la solution proposée. / In the scope of Business Intelligence, planning aims to support multiple actors in their process of converging different views and problematics from different domains to get a shared business planning model. A major difficulty in business planning is that each actor states her/his views and problematics with a different time scale. Integrating them into a unique model that represents a common state of reality becomes very costly and awkward to manage when basing the construction of these models on discrete modeling techniques used by current tools of business planning. This doctorate thesis proposes a novel solution, beyond the state-of-the-art, for addressing these issues: it conceives a novel metamodel based on a continuous time calculus. Through the developed approach, it allows multiple actors to integrate the different business logics of their planning domain in a shared model as well as to observe it from different time scales. The advantages of our solution based on continuous time against solutions based on discrete time are presented through a case study. The conceived metamodel was implemented within a real industrial set in MGDIS (a company specialized in business planning for local governments) following an innovative service oriented architecture: this architecture segregates the modeling from the evaluation to allow the parallelization of model evaluation for big volumes of data. The overall solution conceived and implemented in this thesis was designed to be a real scale prototype to be applied to real scale problems. Besides the case study, it was validated by MGDIS experts on business planning against real requirements.