DESIGN AND IMPLEMENTATION OF UNIFYING FRAMEWORK FOR ENGINEERING DEVS DISTRIBUTIONS SIMULATION

Abstract

The increasing complexity of modern engineering systems has significantly expanded the role of modeling and simulation (M&S) in the design, analysis, and evaluation of real-world processes. Simulation engineering provides an effective means of representing complex systems through computational models, simulation environments, and execution platforms, enabling researchers and engineers to evaluate system behavior without the risks and costs associated with physical experimentation. Among the various modeling formalisms available, the Discrete Event System Specification (DEVS) framework has emerged as one of the most widely adopted approaches for developing modular, reusable, and scalable simulation models. DEVS offers a structured methodology that clearly separates model specification from simulation execution, making it particularly suitable for parallel and distributed simulation environments.

Over the years, considerable progress has been made in applying DEVS to model complex systems across diverse application domains. Researchers have developed numerous DEVS-based modeling approaches, simulation environments, and execution platforms capable of supporting distributed and high-performance simulations. Despite these advancements, one of the persistent challenges in distributed simulation engineering is the absence of a unified mechanism for managing the migration and interoperability of simulation models, execution engines, and computing platforms. As simulation requirements evolve, organizations often need to transfer simulation models between different software environments or deploy them across heterogeneous distributed computing infrastructures. The lack of standardized migration strategies makes this process complex, time-consuming, and prone to compatibility issues.

Another major challenge is the diversity of existing DEVS simulation environments. Many DEVS simulators have been independently developed using different architectures, programming techniques, communication mechanisms, and execution strategies without adhering to a common development framework or standardized engineering methodology. This heterogeneity limits interoperability among simulation tools, complicates model reuse, and hinders collaboration between researchers and software developers. Furthermore, important design decisions—such as event scheduling mechanisms, synchronization techniques, communication protocols, execution platforms, simulation architectures, and implementation algorithms—are frequently addressed using ad hoc approaches, resulting in inconsistent simulator development practices.

This research addresses these limitations by proposing a comprehensive framework that unifies the engineering process for DEVS-based distributed simulation systems. The study introduces a systematic methodology that supports the design, development, integration, and deployment of heterogeneous DEVS simulation environments while promoting consistency, flexibility, and interoperability throughout the simulation engineering lifecycle.

The first major contribution of this research is the development of a conceptual and formalized implementation guide for DEVS simulation engineering. Based on an extensive review of existing DEVS implementation strategies, a comprehensive taxonomy of fundamental concepts, architectural components, and implementation processes is developed. This taxonomy organizes the essential elements involved in constructing DEVS simulators, including model representation, event management, simulation coordination, communication protocols, execution mechanisms, synchronization strategies, and software architecture. Formal definitions are provided to establish a common understanding of these concepts, thereby facilitating standardized development practices and supporting future DEVS standardization initiatives.

The second major contribution is the design and implementation of the Model-Driven Distributed Simulation Engineering Framework (MD2SEF), a layered engineering framework that applies Model-Driven Engineering (MDE) principles to the development of distributed DEVS simulations. The proposed framework provides an integrated environment that enables developers to model DEVS systems at multiple abstraction levels, specify simulation environments, configure distributed execution platforms, integrate heterogeneous simulation components, and automatically generate executable distributed simulation code. By separating conceptual system design from implementation details, the framework significantly improves software maintainability, portability, and scalability.

The MD2SEF framework incorporates multiple engineering layers, each responsible for a specific aspect of the simulation development process. These layers collectively support model specification, simulation environment configuration, distributed execution management, component integration, and automated code generation. This structured architecture simplifies the development of complex distributed simulation applications while reducing implementation effort and minimizing the risk of design inconsistencies.

A key feature of the proposed framework is its ability to integrate heterogeneous DEVS simulators developed using different technologies and execution strategies. Through standardized model representations and model transformation techniques, the framework promotes interoperability among independently developed simulation systems, enabling greater reuse of existing simulation assets and facilitating collaboration across multidisciplinary development teams.

Furthermore, the application of Model-Driven Engineering within the framework enhances development agility by enabling automatic transformation of high-level conceptual models into executable distributed simulation applications. This automation reduces manual coding requirements, accelerates software development, improves model consistency, and allows simulation systems to adapt more easily to changing operational requirements or technological advancements.

The proposed framework also provides a systematic approach for managing the migration of simulation models, simulation engines, and execution platforms. This capability ensures that distributed simulation systems remain flexible and adaptable throughout their lifecycle, allowing organizations to respond efficiently to evolving computational environments, infrastructure upgrades, and changing simulation objectives.

The results of this research demonstrate that the proposed unifying framework significantly improves the engineering of distributed DEVS simulation systems by providing standardized development guidelines, enhanced interoperability, automated model transformation, and efficient integration of heterogeneous simulation components. The framework offers a scalable and reusable solution for simulation engineering that supports both academic research and industrial applications involving complex distributed systems.

Overall, this study contributes to the advancement of distributed simulation engineering by establishing a unified, model-driven framework that simplifies DEVS simulator development while improving portability, interoperability, maintainability, and development efficiency. The proposed approach provides a strong foundation for future research in cloud-based simulation, digital twins, cyber-physical systems, high-performance computing, and intelligent simulation environments.