Catálogo de publicaciones - libros

We extend our approach for verifying properties of graph transformation systems using suitable abstractions. In the original approach properties are specified as formulae of a propositional temporal logic whose atomic predicates are monadic second-order graph formulae. We generalize this aspect by considering more expressive logics, where edge quantifiers and temporal modalities can be interleaved, a feature which allows, e.g., to trace the history of objects in time. This requires the use of graph transition systems, a generalization of transition systems where states and transitions are mapped to graphs and graph morphisms, respectively, and of a corresponding notion of abstraction. After characterizing fragments of the logic which can be safely checked on the approximations, we show how the verification of the logic over graph transformation systems can be reduced to the verification of a logic over suitably defined Petri nets.

The theory of abstract algebraic logic aims at drawing a strong bridge between logic and universal algebra, namely by generalizing the well known connection between classical propositional logic and Boolean algebras. Despite of its successfulness, the current scope of application of the theory is rather limited. Namely, logics with a many-sorted language simply fall out from its scope. Herein, we propose a way to extend the existing theory in order to deal also with many-sorted logics, by capitalizing on the theory of many-sorted equational logic. Besides showing that a number of relevant concepts and results extend to this generalized setting, we also analyze in detail the examples of first-order logic and the paraconsistent logic of da Costa.

We present a notion of module acquired from developing an algebraic framework for service-oriented modelling. More specifically, we give an account of the notion of module that supports the composition model of the SENSORIA Reference Modelling Language (SRML). The proposed notion is independent of the logic in which properties are expressed and components are programmed. Modules in SRML are inspired in concepts proposed for Service Component Architecture (SCA) and Web Services, as well the modules that have been proposed for Algebraic Specifications, namely by H. Ehrig and F. Orejas, among others; they include interfaces for required (imported) and provided (exported) services, as well as a number of components (body) whose orchestrations ensure how given behavioural properties of the provided services are guaranteed assuming that the requested services satisfy required properties.

Communities of autonomous units are rule-based and graph-transformational devices to model data-processing systems that may consist of distributed and mobile components. The components may communicate and interact with each other, they may link up to ad-hoc networks. In this paper, we introduce and investigate the parallel-process semantics of communities of autonomous units.

We connect the algebraic specification language with a variety of automated first-order provers. The heart of this connection is an institution comorphism from to (softly typed first-order logic); the latter is then translated to the provers’ input syntaxes. We also describe a GUI integrating the translations and the provers into the Heterogeneous Tool Set. We report on experiences with provers, which led to fine-tuning of the translations. This framework can also be used for checking consistency of specifications.

We introduce two institutions for the process algebra , one for the traces model, and one for the stable failures model. The construction is generic and should be easily instantiated with further models. As a consequence, we can use structured specification constructs like renaming, hiding and parameterisation (that have been introduced over an arbitrary institution) also for . With a small example we demonstrate that structuring indeed makes sense for .

During model-driven software development, we are inevitably confronted with design models that contain a wide variety of inconsistencies. Interactive and automated support for detecting and resolving these inconsistencies is therefore indispensable. In this paper, we report on an iterative inconsistency resolution process. Our approach relies on the underlying formalism of graph transformation. We exploit the mechanism of critical pair analysis to analyse dependencies and conflicts between inconsistencies and resolutions, to detect resolution cycles and to analyse the completeness of resolutions. The results of this analysis are integrated in the iterative inconsistency resolution process and can help the software engineer to develop and evolve models in presence of inconsistencies.

We propose to extend the algebraic-coalgebraic specification language by full coalgebraic modal logic based on predicate liftings for functors. This logic is more general than the modal logic previously used in and supports the specification of a variety of modal logics, such as graded modal logic, majority logic, and probabilistic modal logic. thus becomes a modern modal language that covers a wide range of Kripke and non-Kripke semantics of modal logics via the coalgebraic interpretation.

SVL is the core of a slicing-based verification environment for UML statechart models. We present an overview of the SVL software architecture. Special attention is paid to the slicing approach. Slicing reduces the complexity of the verification approach, based on removing pieces of the model that are not of interest during verification. In [18] a slicing algorithm has been proposed for statecharts, but it was not able to handle orthogonal regions efficiently. We optimize this algorithm by removing false dependencies, relying on the broadcasting mechanism between different parts of the statechart model.

We present a model theoretic analysis of synchronization of deterministic CSP processes. We show that there is co-amalgamation within the indexed coalgebraic reconstruction of CSP developed in [14]. Synchronization, however, can not be characterized in terms of co-amalgamation. We show that synchronization can be described, nevertheless, as a pullback construction within the corresponding fibred algebraic setting. Analyzing and generalizing the transition between the indexed and the fibred setting we show that for a wide range of signature embeddings :→ the -algebras, traditionally considered as parameter algebras, can be considered also as signatures, instead.