Model Transformation: the Automation Booster for MBSE Andrea D’Ambrogio Department of Enterprise Engineering University of Rome Tor Vergata Rome, Italy dambro@uniroma2.it Copyright © held by the author. ABSTRACT Management Group – the same body that defines and promotes Model-based systems engineering (MBSE) is defined by UML and SysML), has introduced into the MBSE discipline the International Council on Systems Engineering (INCOSE) the formalization of the business processes describing the as "the formalized application of modeling to support system interactions among the organizations that build and/or make requirements, design, analysis, verification, and validation use of systems [5]. activities beginning in the conceptual design phase and The tutorial overviews existing methods that have been continuing throughout development and later life cycle" [1]. developed to automate the simulation-based analysis of The advantages obtained by the MBSE approach, in terms of systems and processes. Specifically, such methods enable the enhanced communications, reduced development risks, automated transformation of SysML [6][7] and BPMN [8][9] improved quality, increased productivity and enhanced models into the corresponding simulation models, which are knowledge transfer, can be further scaled up by innovative executed to yield the performance indices of interest. The model-driven approaches that treat models as the primary proposed approach can be effectively used by system designers artifacts of development, by introducing a radical shift from a and business analysts who do not have specific expertise in the strictly contemplative use of models to a more productive and fields of simulation-based analysis, without requiring the powerful model use. Model-driven engineering enhances the acquisition of specific modeling and simulation know-how and typical MBSE advantages by use of metamodeling techniques the manual effort that is typically needed to define and and model transformations, which allow system designers to implement simulation models. In addition, the extensive use of obtain a significant increase of the level of automation model-driven techniques contributes to provide a significant throughout the system lifecycle [2]. degree of customizability, which facilitates the tailoring of the A metamodel is a model used to describe a family of proposed model transformations to specific needs (e.g., to models, in other words it is a model that defines the primitives make use of alternative modeling languages and/or simulation of a modeling language, which is used to specify models at platforms) [10][11][12]. user level. As an example, the UML metamodel is the model defining the primitives (i.e., class, association, state, etc.) that REFERENCES are instantiated in standard UML models. A model [1] INCOSE, 2007. Systems Engineering Vision 2020, v.2.03, September, transformation is the specification of a set of mapping rules 2007. that are executed to transform a given model into a different [2] D. C. Schmidt, 2006. Model-driven Engineering, IEEE Computer, model, which conforms to the same or to a different February 2006, pp. 25-31. metamodel. Model-driven engineering enables a radical shift in [3] D. Gianni, A. D'Ambrogio and A. Tolk, 2014. Modeling and Simulation- terms of modeling activities, from a strictly contemplative use Based Systems Engineering Handbook, CRC Press, ISBN 9781466571457. of models to a more productive and powerful model use. [4] OMG, 2010. Object Management Group, System Modeling Language The tutorial first illustrates the principles and standards of (SysML), v.1.2. model-driven engineering and then shows a possible [5] OMG 2011. Object Management Group, Business Process Model and application to the simulation-based analysis of systems and Notation (BPMN), version 2.0. processes specified by use of SysML and BPMN, respectively [6] P. Bocciarelli, A. D'Ambrogio and G. Fabiani 2012. A Model-driven [3]. SysML (Systems Modeling Language) is the UML-based Approach to Build HLA-based Distributed Simulations from SysML Models, Proceedings of the 2nd International Conference on Simulation language that provides the modeling capability required in the and Modeling Methodologies, Technologies and Applications, Rome, systems engineering domain. SysML is now considered the Italy, 28 - 31 July, 2012, pp. 49-60. standard modeling notation adopted in the MBSE context [4]. [7] P. Bocciarelli, A. D'Ambrogio, A. Giglio and D. Gianni, 2013. A SaaS- In addition, the recent adoption of the BPMN (Business based automated framework to build and execute distributed simulations Process Modeling and Notation) standard by the OMG (Object from SysML models. Proceedings of the Winter Simulation Conference (WSC 2013), Washington, DC, USA, December 8-11, pp 1371-1382. [8] P. Bocciarelli, A. D'Ambrogio, Automated performance analysis of systems and software engineering, business process business processes, Proceedings of the 2012 Spring Simulation management, and distributed simulation. In such areas he has Multiconference, SpringSim 2012; Orlando, FL; United States, pp. 266- 274, March 26-30, 2012. participated to several projects at both European and overseas [9] P. Bocciarelli, A. D'Ambrogio, E. Paglia, A language for enabling level and has authored more than 100 journal/conference model-driven analysis of business processes, Proceedings of the 2nd papers. He is general chair of the 49th Summer Computer International Conference on Model-Driven Engineering and Software Simulation Conference (SCSC 2017), general co-chair of the Development (MODELSWARD 2014), pp. 325-332, Lisbon; Portugal; Summer Simulation Multiconference (SummerSim 2017), and January 7-9, 2014. has been general chair of the SCS/ACM/IEEE International [10] P. Bocciarelli, A. D'Ambrogio, A. Falcone, A. Garro and A. Giglio. Symposium on Theory of Modeling & Simulation (TMS 2014) 2015. A Model-Driven Approach to Enable the Distributed Simulation of Complex Systems, Proceedings of the Sixth International Conference and of the IEEE International Conference on Enabling on Complex Systems Design & Management, CSD&M 2015, pp. 171– Technologies: Infrastructure for Collaborative Enterprises 183, Paris, France, November 23–25, 2015. (WETICE 2008). He has served as program chair and/or [11] P. Bocciarelli, A. D'Ambrogio, D. Gianni, 4SEE: a model-driven member of the program committee of various international simulation engineering framework for business process analysis in a conferences, among which IEEE WETICE, SCS/ACM/IEEE SaaS paradigm, Proceedings of the 2013 Spring Simulation SpringSim, SCS/ACM SummerSim, IEEE/ACM PADS and Multiconference, SpringSim 2013, pp. 218-225, San Diego, CA, USA, April 7-10, 2013. INFORMS/ACM Winter Simulation Conference. He started [12] P. Bocciarelli, A. D'Ambrogio, A. Giglio, E. Paglia, D. Gianni, A the IEEE International Workshop on Collaborative Modeling Transformation Approach to Enact the Design-Time Simulation of and Simulation (CoMetS) in 2010 and the SCS/ACM/IEEE BPMN Models, Proceedings of the IEEE 23rd International WETICE International Symposium on Model-driven Approaches for Conference (WETICE 2014), pp. 199-204, Parma, Italy, June 23-25, Simulation Engineering (Mod4Sim) in 2011. Andrea 2014 D'Ambrogio is associate editor of the Transactions of the Society for Modeling & Simulation International AUTHOR BIOGRAPHY (SIMULATION journal), member of the editorial board the Andrea D'Ambrogio is associate professor of computer IAENG International Journal of Computer Science (IJCS) and science at the Department of Enterprise Engineering of the has been member of the editorial board of the Simulation University of Roma "Tor Vergata" (Italy) and director of the Practice and Theory (SIMPAT) journal. He is member of the post graduate Master degree in "Systems Engineering", management committee of the ICT COST Action on “Multi established at the University of Roma "Tor Vergata" in 2012. Paradigm Modelling for Cyber Physical Systems” Andrea D'Ambrogio's research interests are in the systems and (MPM4CPS). He is member of IEEE, IEEE Computer Society, software engineering field, specifically in the areas of system ACM, SCS and INCOSE. performance and dependability engineering, model-driven