|
Welcome to the home page of the 20th Computer Automated Multi-Paradigm Modelling (CAMPaM) workshop.
This is the CAMPaM 2.0 reboot of the succesful series of CAMPaM workshops held at McGill's Bellairs Research Institute on the beautiful island of Barbados. This workshop aims to further the state-of-the-art in (Computer Automated) Multi-Paradigm Modelling. Multi-Paradigm Modelling (MPM) advocates modelling every part and aspect of a system explicitly, at the most appropriate level(s) of abstraction, using the most appropriate modelling formalism(s). This, realizing that explicitly modelling workflow/processes as is the modelling of artefacts. In addition to modelling workflows, modelling languages' engineering, including model transformation, and the study of their semantics, are used to realize MPM. MPM is seen as an effective answer to the challenges of designing and evolving Cyber-Physical Systems (CPS) and Systems of Systems (SoS). We aim to tackle open scientific problems and explore future directions of MPM by bringing together a small group with a mixture of senior world experts and junior researchers in this field for an intense one-week, by invitation only workshop in a relatively secluded location. |
|
Organizers
The Topics
Some tentative topics:
- Experiment modelling and experiment/data replicability. There are currently Electronic Lab Notebooks, but they do not capture the complexity of experiments (both real-world and software/simulation/data driven, and combinations thereof). Explicitly modelling, recording and mining experiments is crucial for replicability. This is a basis for making model validity explicit. One important aspect of experiment replicability is traceability. This is often in the form of keeping track of how an artefact was obtained. This relating relating a trace of (experimentation) activities and artefacts to the process/workflow that gave rise to that trace. It is only this backward link that will explain whether two artefacts are for example related via version relationship, and if so, whether a new version was needed to fix a bug, due to a new or changing requirement, to optimize the system, etc. Another important aspect of meaningful experiment replicability (and collaboration, both between experimentation stages and between different stakeholders) is explainability of concepts used. There is a link with model management and with testing. A test is a particular kind of experiment, be it on software or on a physical system, or combinations thereof. There is also the interesting topic (related to the MPM topic) of front-loading of testing by early-stage experiments, usually on simplified or incomplete models.
- MPM = multi-* (with * = view, component, formalism, abstraction, ...). We need to delve deeper, and here too, there is a link with model management. Topics of particular interest are consistency and traceability (see experiment modelling) between the many related artefacts.
- Focusing on the FTG(+PM), continuing the work started in the MPM4CPS COST action.
Some workflow patterns and types of modelling languages (formalisms, such as differential equations and Statecharts), are used successfully in CPS engineering practice.
In analogy with software Design Patterns, we believe there exist useful modelling paradigm (formalisms + workflow) patterns that should be made explicit.
We should bring together researchers and practitioners whose expertise covers design, optimization, deployment, formal methods, simulation, digital twins etc., all combining different formalisms.
The concrete goals are to:
- work on the foundation of FTG(+PM): what are the most appropriate languages and architectures to support this (a Modelverse?)?
- Catalogue useful properties of modelling formalisms and workflows.
- Classify different formalisms and workflows based on these properties and link them with different types of CPS problems. This, starting from a CPS case summary by each participant, charting used formalisms and relationships between these, as well as workflows.
