China
Africa
Explore chapters and articles related to this topic
DEVS as a Semantic Domain for Programmed Graph Transformation
Published in Gabriel A. Wainer, Pieter J. Mosterman, Discrete-Event Modeling and Simulation, 2018
Eugene Syriani, Hans Vangheluwe
In almost all modeling and simulation endeavors, some form of model transformation is used. Models are, for example, transformed for optimization purposes, to address new requirements, to synthesize real-time embedded code, and so on. Transformations are also commonly used to describe the semantics of a modeling formalism. In the case of operational semantics, the transformation iteratively updates the state of a model. In the case of denotational semantics, the transformation maps a model in one formalism onto a model in a known formalism, thereby defining the meaning of the original model. Model transformations can be described in many ways. Rule-based descriptions are elegant and easy to understand. Such descriptions have declarative (specifying “what” to change, not “how” to) model rewriting rules as their primitive building blocks. A rule consists of a left-hand side (LHS) pattern that is matched against a host model. If a match is found, this pattern is updated, in the host model, based on what is specified in the right-hand side (RHS) of the rule. Additionally, Negative Application Condition (NAC) patterns may be used, specifying which patterns should not be found in the host model. Because at some level of abstraction, all models can be represented as (typed, attributed) graphs, and thanks to its rigorous formal underpinning, our rule-based specification is based on the theory of graph rewriting.
A versatile and extensible solution to the integration of BIM and energy simulation
Published in Symeon E. Christodoulou, Raimar Scherer, eWork and eBusiness in Architecture, Engineering and Construction, 2017
D. Mazza, E. El Asmi, S. Robert, K. Zreik, B. Hilaire
The idea underlying model transformation is then the possibility of converting the initial PIM into a more specific model that represents a solution for the problem on a specific platform, leading to what is called Platform-Specific Model (PSM). Model-to-model transformation therefore involves mainly a transformation of the concepts (model objects) information (model data, objects features) from the initial, abstract model to a low-level, technology-specific model, thus “rearranging” the original information and objects in order to build up a solution-specific model. Obviously, the advantage of such an approach is that different PSMs, one for each specific platform, can stem from the same PIM, as shown in Figure 1.
Cyber-Physical Systems: a multi-criteria assessment for Internet-of-Things (IoT) systems
Published in Enterprise Information Systems, 2021
Edgar M. Silva, Ricardo Jardim-Goncalves
Model-driven Architecture (MDA) (Miller et al. 2003) was initially designed for software development, but it has been widely applied in system/enterprise interoperability and in business process management areas. To cover the entire system development process, the architecture presents three abstraction layers, the platform-specific model (PSM), platform-independent model (PIM) and computation independent model (CIM) (Miller et al. 2003; Singh and Sood 2009). Supported by two types of transformations, horizontal and vertical, in which a variety of possible transformation categories can be established, such as, marking, meta-model transformation, model transformation or model merging (Singh and Sood 2009). These transformations assist in the development process (more automatic) of new system functionalities and can also be used to perform system/enterprise interoperability by matching process phases or data descriptions from one company/system to another. Between the two top levels, CIM and PIM, MDA does not specify any transformations, presenting instead the use of Meta-Object Framework, an OMG standard that gives the basis for the definition of meta-models.