Modelling languages are essential in many disciplines to express knowledge in a precise way. Furthermore, some domains require families of notations (rather than individual languages) that account for variations of a language. Some examples of language families include those to define automata, Petri nets, process models or software architectures. Several techniques have been proposed to engineer families of languages, but they often neglect the language's concrete syntax, especially if it is graphical.

Quantum programming languages support the design of quantum applications. However, to create such programs, one needs to understand the fundamental characteristics of quantum computing and quantum information theory. Furthermore, quantum algorithms frequently make use of abstract operations with a hidden low-level realization (e.g., Quantum Fourier Transform). Thus, turning from elementary quantum operations to a higher-level view of quantum circuit design not only reduces the development effort but also lowers the entry barriers for non-quantum computing experts. To this end, this article proposes a modeling language and design framework for quantum circuits. This allows the definition of composite operators to advocate a higher-level quantum algorithm design, together with automated code generation for the circuit execution. To demonstrate the benefits of the proposed approach, coined Composition-based Quantum Circuit Designer , we applied it for realizing the Quantum Counting algorithm and the Quantum Approximate Optimization Algorithm. Our evaluation results show that, compared to an existing state-of-the-art editor, the proposed approach allows for the realization of both quantum algorithms on a high level with a substantially reduced development effort. In particular, the proposed approach shows constant scaling when increasing the size of the investigated quantum circuits and a lower change criticality when evolving existing quantum circuits.

Many recommenders for modelling tasks have recently appeared. They use a variety of recommendation methods,tailored to concrete modelling languages. Typically, recommenders are created as independent programs, and subsequently need to be integrated within a modelling tool, incurring in high development effort. Moreover, it is currently not possible to reuse a recommender created for a modelling language with a different notation, even if they are similar. To attack these problems, we propose a methodology to reuse and integrate recommenders into modelling tools. It considers four orthogonal dimensions: the target modelling language, the tool, the recommendation source, and the recommended items. To make homogeneous the access to arbitrary recommenders, we propose a reference recommendation service that enables indexing recommenders, investigating their properties, and obtaining recommendations likely coming from several sources. Our methodology is supported by IronMan, an Eclipse plugin that automates the integration of recommenders within Sirius and tree-based editors, and can bridge recommenders created for a modelling language for their reuse with a different one. We evaluate the power of the tool by reusing 2 recommenders for 4 different languages, and integrating them into 6 modelling tools.

Petri nets are commonly used to represent concurrent systems. However, they lack support for modelling and analysing system families, like variants of controllers, different variations of a process model, or the possible configurations of a flexible assembly line. To facilitate modelling potentially large collections of similar systems, in this paper, we enrich Petri nets with variability mechanisms based on product line engineering. Moreover, we present methods for the efficient analysis of the place and transition invariants in all defined versions of a Petri net. Efficiency is achieved by analysing the system family as a whole, instead of analysing each possible net variant separately. For this purpose, we lift the notion of incidence matrix to the product line level, and rely on constraint solving techniques. We present tool support and evaluate the benefits of our techniques on synthetic and realistic examples, achieving in some cases speed-ups of two orders of magnitude with respect to analysing each net variant separately.

Runtime monitoring is critical for ensuring safe operation and for enabling self-adaptive behavior of Cyber-Physical Systems (CPS). Monitors are established by identifying runtime properties of interest, creating probes to instrument the system, and defining constraints to be checked at runtime. For many systems, implementing and setting up a monitoring platform can be tedious and time-consuming, as generic monitoring platforms do not adequately cover domain-specific monitoring requirements. This situation is exacerbated when the System under Monitoring (SuM) evolves, requiring changes in the monitoring platform. Most existing approaches lack support for the automated generation and setup of monitors for diverse technologies and do not provide adequate support for dealing with system evolution. In this paper, we present GRuM (Generating CPS Runtime Monitors), a framework that combines model-driven techniques and runtime monitoring, to automatically generate a customized monitoring platform for a given SuM. Relevant properties are captured in a Domain Model Fragment, and changes to the SuM can be easily accommodated by automatically regenerating the platform code. To demonstrate the feasibility and performance we evaluated GRuM against two different systems using TurtleBot robots and Unmanned Aerial Vehicles. Results show that GRuM facilitates the creation and evolution of a runtime monitoring platform with little effort and that the platform can handle a substantial amount of events and data.

To test a system efficiently, we need to know how good are the defined test cases and to localize detected faults in the system. Measuring test coverage can address both concerns as it is a popular metric for test quality evaluation and, at the same time, is the foundation of advanced fault localization techniques. However, for Domain-Specific Languages (DSLs), coverage metrics and associated tools are usually manually defined for each DSL representing costly, error-prone, and non-reusable work.

With the increase of technologies such as the Internet of Things (IoT) and Cyber-Physical Systems, a huge amount of data is generated by current systems. To gain insights from this data, it must be combined with meta-information about its origins. Therefore, Digital Twins (DTs), as a common representation of a system and its data, are currently gaining traction in both industry and academia. However, these DTs have of course to be evolvable in order to reflect the high need of flexibility of the systems to support extensions, adaptations, customizations, etc. Evolving the DT representations currently not only involves a lot of manual effort, but might also lead to loss of data if not done correctly. To provide dedicated evolution support, we propose a dedicated framework for realizing evolution strategies between the schema, instance, and data level of a DT. In particular, we present a fluent API which allows the flexible but systematic manipulation of DTs during runtime and demonstrate its usage for a use case.

Industrial standards define domain-specific languages that are frequently used for developing control software. For instance, IEC 61499 standardizes a graphical modeling language that includes a platform-independent application model. The application is composed of Function Blocks. A runtime can execute the model by implementing the semantics that is described in the standard in natural language. By defining an interpreter for IEC 61499 models, we can directly execute them without prior code generation. This enables providing feedback directly on the model level. We present an interpreter for Basic Function Blocks, which encapsulate a state-based Execution Control Chart. An existing EMF meta-model for IEC 61499 was extended with an operational semantics implemented in Java and Xtend. The test cases are defined either in Java or as an interface model. Such a model is standardized in IEC 61499 as Service Sequences. We evaluate our interpreter by executing the Basic Function Blocks that are defined in the standard and compare our results to those of the open-source runtime 4diac FORTE. As a practical use case, we show how developers can use the interpreter for unit testing self-defined Basic Function Blocks.

With JSON's increasing adoption, the need for structural constraints and validation capabilities led to JSON Schema, a dedicated meta-language to specify languages which are in turn used to validate JSON documents. Currently, the standardisation process of JSON Schema and the implementation of adequate tool support (e.g., validators and editors) are work in progress. However, the periodic issuing of newer JSON Schema drafts makes tool development challenging. Nevertheless, many JSON Schemas as language definitions exist, but JSON documents are still mostly edited in basic text-based editors. To tackle this challenge, we investigate in this paper how Model-Driven Engineering (MDE) methods for language engineering can help in this area. Instead of re-inventing the wheel of building up particular technologies directly for JSON, we study how the existing MDE infrastructures may be utilized for JSON. In particular, we present a bridge between the JSONware and Modelware technical spaces to exchange languages and documents. Based on this bridge, our approach supports language engineers, domain experts, and tool providers in editing, validating, and generating tool support with enhanced capabilities for JSON schemas and their documents. We evaluate our approach with Shipyard, a JSON Schema-based language for the workflow specification for Keptn, an open-source tool for DevOps automation of cloud-native applications. The results of the case study show that proper editors and language evolution support from MDE can be reused and, at the same time, the surface syntax of JSON is maintained.

Model-driven optimization has gained much interest in the last years which resulted in several dedicated extensions for in-place model transformation engines. The main idea is to exploit domain-specific languages to define models which are optimized by applying a set of model transformation rules. Objectives are guiding the optimization processes which are currently mostly realized by meta-heuristic searchers such as different kinds of Genetic Algorithms. However, meta-heuristic search approaches are currently challenged by reinforcement learning approaches for solving optimization problems. In this new ideas paper, we apply for the first time reinforcement learning for in-place model transformations. In particular, we extend an existing model-driven optimization approach with reinforcement learning techniques. We experiment with value-based and policy-based techniques. We investigate several case studies for validating the feasibility of using reinforcement learning for model-driven optimization and compare the performance against existing approaches. The initial evaluation shows promising results but also helped in identifying future research lines for the whole model transformation community.

The description of families of production systems usually relies on the use of variability modelling. This aspect of modelling is gaining increasing interest with the emergence of Industry 4.0 to facilitate the product development as new requirements appear. As a consequence, there are several emerging modelling techniques able to apply variability in different domains. In this paper, we introduce an approach to establish product system families in AutomationML. Our approach is based on the definition of feature models describing the variability space, and on the assignment of presence conditions to AutomationML model elements. These conditions (de-)select the model elements depending on the chosen configuration. This way, it is possible to model a large set of model variants in a compact way using one single model. To realize our approach, we started from an existing EMF-based AutomationML workbench providing graphical modelling support. From these artifacts, we synthesized an extended graphical modelling editor with variability support, integrated with FeatureIDE. Furthermore, we validated our approach by creating and managing a production system family encompassing six scenarios of the Pick and Place Unit Industry 4.0 demonstrator.

Model-based development approaches, such as Model-Driven Engineering (MDE), heavily rely on the use of modelling languages to achieve and automate software development tasks. To enable the definition of model variants (e.g., supporting the compact description of system families), one solution is to combine MDE with Software Product Lines. However, this is technically costly as it requires adapting many MDE artefacts associated to the modelling language -- especially the meta-models and graphical environments. To alleviate this situation, we propose a method for the automated injection of variability into graphical modelling languages. Given the meta-model and graphical environment of a particular language, our approach permits configuring the allowed model variability, and the graphical environment is automatically adapted to enable creating models with variability. Our solution is implemented atop the Eclipse Modeling Framework and Sirius, and synthesizes adapted graphical editors integrated with FeatureIDE.

Model-Driven Engineering (MDE) promotes the use of models to conduct all phases of software development in an automated way. Models are frequently defined using Domain- Specific Modelling Languages (DSMLs), which many times need to be developed for the domain at hand. However, while constructing DSMLs is a recurring activity in MDE, there is scarce support for gathering, reusing and enacting knowledge for their design and implementation. This forces the development of every new DSML to start from scratch. To alleviate this problem, we propose the construction of DSMLs and their modelling environments aided by patterns which gather knowledge of specific domains, design alternatives, concrete syntax, dynamic semantics and functionality for the modelling environment. They may have associated services, realized via components. Our approach is supported by a tool that enables the construction of DSMLs through the application of patterns, and synthesizes a graphical modelling environment according to them.

The effectiveness of a MOOC as a learning environment could be encouraged by an active students' participation in social and collaborative learning processes. In this sense, both social media platforms and collaborative learning services have showed several successful experiences with massive projects. However, in this context, the lack of a pedagogical model that can be coupled with a multiple media and distributed resources increases the complexity of processes such as monitoring and assessment, which are fundamental in a learning environment. In consequence, this can produce the instructors' workload when they script and perform learning activities into these scenarios. This paper presents several reflections about monitoring and assessment processes from two collaborative learning systems approaches and proposes some ideas about their application in a MOOC context, with the aim to reduce the instructors' workload. Additionally, other considerations about massive collaborative learning experiences are presented.