Modelling and Simulation Based Engineering for IoT

物联网(IoT)在加拿大正迅速成为商业现实。由于数十亿设备和复杂、异构系统的互联，物联网范式带来了新的软件工程挑战，要求它们在不确定和不断变化的环境中可靠地合作。软件开发框架需要发展以满足这种动态自适应系统的需要，以自主地应对(预测、检测和适应)操作、目标或环境中的内部和外部变化。最近的工业项目和研究表明，模型驱动工程(MDE)方法和工具的应用对于进一步开发复杂的、软件密集型系统是必不可少的。在MDE中，模型用于驱动整个工程或软件开发过程，模型转换用于操纵这些模型。基于模型的设计允许在最适当的抽象级别上使用最合适的形式对系统的结构、行为和交互进行建模，例如软件部分的状态图、Petri网或活动图，以及物理部分的Modelica。此外，特定于领域的语言可以最大限度地约束建模者处理手头的问题。此外，开发过程本身的显式建模允许进行分析、优化和自动化。使用这种基于模型的推动者还允许在开发过程的早期处理需求和设计错误以及替代方案。拥有协调这些(需求、分析、设计和运行时)模型的有效且高效的方法、跨多个视图维护一致性以及管理跨模型的变更传播是至关重要的。该研究计划的长期目标是开发创新的建模和基于模拟的方法、技术和软件工具，以促进自适应和经济高效的物联网系统的需求分析和设计。这一计划的长期愿景将通过几个短期的具体项目来实现。我打算提出有效和可扩展的方法和技术，用于建模语言工程、过程建模、全球可追溯性管理以及自适应物联网系统的模拟和分析。这些方法和工具将通过物联网领域的应用程序进行演示。结合使用MDE、需求工程和模拟来应对物联网系统的复杂性和挑战仍处于初级阶段。我设想这个研究计划将对模型驱动工程和物联网领域做出贡献。该研究计划还旨在通过优化设计和部署时间来适应行业需求，并最终减少资本和运营费用。

The Internet of Things (IoT) is fast becoming a commercial reality in Canada. The IoT paradigm has brought about new software engineering challenges due to the interconnection of billions of devices and complex, heterogeneous systems which are required to cooperate reliably in an uncertain and ever-changing environment. Software development frameworks need to evolve to address the needs of such dynamically adaptive systems to autonomically cope with (anticipate, detect, and adapt to) internal and external changes in the operation, goals, or environment. Recent industrial projects and research have demonstrated that the application of model-driven engineering (MDE) methods and tools are essential to further the development of complex, software-intensive systems. In MDE, models are used to drive the entire engineering or software development process, and model transformations are used to manipulate these models. Model-based design enables modelling of the structure, behaviour, and interactions of the system at the most appropriate level of abstraction using the most appropriate formalisms, such as Statecharts, Petri Nets, or Activity Diagrams for the software parts and Modelica for the physical parts. Furthermore, domain-specific languages may maximally constrain the modeller to the problem at hand. Moreover, explicit modelling of the development process itself allows for analysis, optimization and automation. Use of such model-based enablers also allows requirements and design errors and alternatives to be addressed early-on in the development process. Having an effective and efficient way of coordinating these (requirements, analysis, design, and runtime) models, maintaining consistency across multiple views, and managing the propagation of changes across the models is crucial. The long-term goal of this research program is to develop innovative modelling and simulation-based methods, techniques, and software tools to facilitate requirements analysis and design of adaptive and cost-effective IoT Systems. The long-term vision of this program will be realized through several short-term, concrete projects. I intend to propose effective and scalable methods and techniques for modelling language engineering, process modelling, global traceability management, and simulation and analysis of adaptive IoT systems. The methods and tools will be demonstrated with applications from the IoT domain. The combined use of MDE, requirements engineering, and simulation to cope with the complexities and challenges of IoT systems is still in its infancy. I envision this research program to contribute to both the model-driven engineering and IoT domains. The research program also aims to align with industry needs through optimization of design and deployment time and ultimately to a reduction in capital and operating expenses.