Prompt 6 — Discrete Event Simulation and Mission Modeling
Conduct a comprehensive review of discrete-event simulation, agent-based simulation, system dynamics, mission engineering, campaign analysis, operational research, and defense simulation methodologies. Evaluate how enterprise architectures can be transformed into executable simulation models capable of representing schedules, resources, dependencies, risks, and organizational behavior.
Generated on: June 07, 2026
Now I have gathered comprehensive research material across all the key areas. Let me create an outline and generate the final literature review.
Comprehensive Literature Review: Discrete-Event Simulation, Agent-Based Simulation, Mission Engineering, and Cyber-Physical System Integration
1. Foundations of Discrete-Event Simulation and DEVS Formalism
Discrete-Event System Specification (DEVS) provides a rigorous mathematical framework for modeling and simulating complex discrete-event systems with proven applications across diverse domains [1]. The DEVS formalism enables comprehensive modeling of system structure, behavior, and information aspects, offering distinct advantages including completeness, verifiability, extensibility, and maintainability [1]. This foundational approach has evolved through multiple implementations across programming languages (C++, Java, Python), creating a robust ecosystem for interoperable modeling and simulation [1].
Advanced discrete-event simulation architectures now support parallel and distributed execution, dramatically improving computational efficiency for large-scale models [2]. Parallel discrete-event simulation (PDES) employs sophisticated synchronization mechanisms—both optimistic and conservative—to accelerate simulations while maintaining correctness [3]. The xDEVS framework exemplifies this evolution, combining traditional DEVS implementations with cloud deployment capabilities, achieving speedups of 15.95× on distributed systems while maintaining model reusability and semantic rigor [2].
DEVS extends naturally to modeling industrial systems within the context of Industry 4.0. Discrete event systems can replicate complex production dynamics, incorporating stochastic variables and variability endemic to manufacturing environments [4]. Real-world validation studies demonstrate that discrete-event models achieve less than 5% deviation from actual plant data, confirming their precision in capturing manufacturing process behavior [5]. This validation credibility positions discrete-event simulation as a primary decision-support tool for industrial optimization and waste reduction initiatives.
2. Agent-Based Simulation and Emergent Behavior Modeling
Agent-based modeling (ABM) represents a paradigm shift from centralized system representation toward autonomous, heterogeneous entities that exhibit adaptive behavior and emergent dynamics [6]. Unlike traditional top-down modeling, ABM captures decentralized decision-making, inter-agent interactions, and the emergence of collective phenomena from microscopic behavioral rules. This approach proves particularly valuable in modeling systems where individual actors pursue conflicting objectives, exhibit bounded rationality, and operate within complex social or operational networks [7].
Military and defense applications demonstrate the power of agent-based simulation for operational effectiveness analysis. Agent-based modeling of UAV swarm precision strikes reveals how emergent collective behavior affects target destruction rates and attrition, accounting for communication, maneuverability, firepower, and sensing capabilities [8]. Similarly, agent-based simulations enable evaluation of network-centric warfare scenarios where communication failures directly impact combat effectiveness [9]. These simulations provide quantitative measures of operational performance under varying environmental conditions, directly supporting force composition and tactical planning decisions.
Multi-agent simulation platforms enable sophisticated behavioral modeling where agents maintain individual state variables, exhibit goal-directed behavior, and interact through explicit communication mechanisms [10]. The FLAME-GPU framework accelerates agent-based transport models by leveraging GPU acceleration, achieving 100× improvements over traditional CPU-based simulations while scaling to model vehicle populations of tens of thousands [10]. This computational advancement democratizes large-scale agent-based modeling, making realistic population-level simulations feasible for urban planning, epidemiology, and logistics optimization.
3. System Dynamics and Feedback-Based Complex System Modeling
System Dynamics (SD) provides a complementary approach to discrete-event and agent-based simulation, emphasizing causal feedback loops, time-delayed responses, and nonlinear relationships in complex systems [6]. SD excels at capturing aggregate-level phenomena, policy impacts, and long-term system trajectories—particularly relevant for strategic planning, resource management, and organizational behavior modeling.
Hybrid simulation—combining discrete-event, system dynamics, and agent-based approaches—addresses limitations inherent in single-method modeling by integrating macroscopic aggregate-level dynamics with microscopic behavioral heterogeneity [6]. In operational research contexts, hybrid models have been successfully applied to healthcare systems, supply chain management, and manufacturing environments where both continuous aggregate flows and discrete event sequences matter [6]. Software vulnerability risk contagion in supply chains demonstrates SD&#39;s effectiveness at modeling cascading effects and defense strategy impacts across multi-layer network structures [11].
System dynamics enables the formalization of enterprise-level decision-making processes and organizational learning dynamics. Economic and mathematical modeling of human resource development in hybrid work environments demonstrates how SD can quantify the nonlinear relationship between competency development, risk, and organizational efficiency—guiding resource allocation under budget constraints [12]. This approach translates qualitative management theory into executable simulation models, facilitating evidence-based decision support for workforce transformation initiatives.
4. Mission Engineering and Operational Planning Frameworks
Mission engineering represents the systematic application of systems engineering principles to military and operational domains, translating strategic objectives into executable plans spanning resources, schedules, risks, and organizational structures [13]. Digital transformation in project and operations management demonstrates how enterprise architectures integrating AI, real-time analytics, and cloud infrastructure reduce operational expenses by 25% while improving schedule precision by 40% [13]. These metrics reflect the tangible value of mission-centric planning transformed through digital tools.
Human performance modeling architectures bridge operational planning and human factors analysis. The integration of Systems Modeling Language (SysML) with discrete-event simulation tools like IMPRINT creates traceability from system design models to human performance predictions, enabling evaluation of mission impacts on operator workload and effectiveness [14]. This architecture maintains SysML as the authoritative source of truth while enabling quantitative human factors analysis through simulation.
Hierarchical scheduling mechanisms for logistics and resource management address operational efficiency under dynamic mission requirements. Research on UAV logistics operations using hierarchical channel management demonstrates that explicit task classification, channel allocation, and conflict resolution strategies increase channel utilization by 25% while reducing conflict rates by 35% [15]. These scheduling improvements directly translate to mission capacity and operational resilience.
5. Defense Simulation, Campaign Analysis, and Wargaming
Professional military wargaming and computer-based defense simulation have evolved into disciplined methodologies for exploring strategic options, training personnel, and analyzing operational alternatives [16]. The Angloval tactical military scenario provides standardized environmental representations for network-centric warfare experimentation, including node mobility, radio propagation models, and realistic battalion-sized operation dynamics [17]. Such standardized scenarios enable reproducible comparative analysis across different networking protocols, tactics, and organizational structures.
Military simulation produces massive datasets—&quot;big data&quot; from high-resolution models simulating complex battlefield dynamics—requiring specialized frameworks for generation, collection, processing, and analysis [18]. Two-tier architectures distinguish between representation data (model structure, parameters) and operational data (simulation results, outcome metrics), with layered service-oriented architectures enabling flexible data query and analysis [18]. These big data management frameworks transform military simulation from isolated studies into organizational knowledge systems.
Contemporary military AI applications span tactical autonomy, intelligence analysis, and decision support systems. Intelligent air-defense command systems leverage multi-agent architectures to achieve coordinated decision-making in distributed, heterogeneous environments without central coordination [7]. Meanwhile, AI-driven wargaming and scenario generation systems automatically produce situational injects, opponent behaviors, and performance feedback aligned with MITRE ATT&CK attack frameworks [19]. These AI-enhanced simulations provide scalable, repeatable training experiences that adapt to trainee performance in real time.
6. Enterprise Architecture and Transformation into Executable Models
Enterprise Architecture (EA) frameworks define organizational structure, information flows, process dependencies, and governance models but frequently remain abstract representations disconnected from operational execution [13]. Transforming EA into executable simulation models requires systematic mapping of architectural elements to simulation entities, constraints, and decision rules.
Building Information Modeling (BIM) and discrete-event simulation integration demonstrates the power of structured data-driven modeling for complex systems [20]. Game-theoretic modeling of client-contractor conflicts combines BIM-generated project data (volumes, resources, dependencies) with AI-based forecasting to identify equilibrium contract conditions that balance cost, schedule, and profitability [20]. This integration transforms static architectural documentation into dynamic executable models supporting strategic analysis and optimization.
Model-Based Systems Engineering (MBSE) provides disciplined methodologies for developing executable architectures. Proof-of-concept digital twins combining LEGO-based physical systems, Raspberry Pi microcontrollers, and FlexSim discrete-event simulation demonstrate how MBSE techniques (Capella, Arcadia) support bidirectional synchronization between physical and digital domains via MQTT protocols [21]. These systems enable virtual commissioning and validation before physical deployment, reducing costly design errors and accelerating time-to-market.
7. Digital Twins and Cyber-Physical System Simulation
Digital twin technology—the bidirectional real-time linkage between physical and virtual system representations—has emerged as a critical enabler for mission-critical operations, predictive maintenance, and organizational decision support [22]. Digital twins integrate real-time sensor data, physics-based simulation models, machine learning algorithms, and interactive visualization into unified platforms that mirror physical system state while enabling what-if analysis and remote operations.
Cyber-physical system modeling demands rigorous approaches to verification and validation across multiple abstraction levels. The quintuple helix conceptual framework addresses DT verification through systematic mapping across problem domains (actual systems), solution domains (virtual systems), and execution domains (bridging mechanisms) at multiple meta-model levels [23]. High-precision DT construction encompasses data acquisition, preliminary model establishment, model refinement through data fusion, and iterative validation cycles [24]. Real-time Kalman filtering and clustering algorithms enhance data fusion efficiency, improving accuracy while managing computational load.
Industrial cyber-physical systems increasingly require zero-trust security architectures combined with anomaly detection driven by DT reference models. Digital twins enable continuous validation of system behavior against normal operating envelopes, detecting process-level attacks that evade network-only intrusion detection [25]. Real-time DT frameworks for SCADA systems achieve 96.3% attack detection F1-scores with sub-500-millisecond latency, substantially outperforming conventional rule-based approaches [25].
Defense and aerospace applications demonstrate DT maturity in mission-critical environments. Spacecraft digital twins constructed through mechanism-data fusion enable on-orbit mission simulation, real-time interactive monitoring, and rapid operating condition prediction, supporting critical space operations including rendezvous, docking, and crewed operations [26]. Similarly, 6G-enabled DT frameworks for real-time cyber-physical systems achieve 0.8ms end-to-end latency using terahertz communications and intelligent reflecting surfaces, enabling ultra-reliable industrial control [27].
8. Model Validation, Verification, and Continuous Evolution
Model validation remains a critical bottleneck in DT deployment. Existing validation techniques from model-based design can be reused for continuous validation of digital twins, detecting anomalies and parameter drift through systematic comparison of predicted versus actual behavior [28]. Parameter estimation based on historical data enables automatic DT evolution as physical systems degrade or are modified, maintaining representativeness throughout system lifecycle.
Reachability analysis for formal verification of cyber-physical models addresses safety-critical applications. Data-driven dynamic sensitivity approaches enable analysis of black-box models (e.g., FMI-compliant components) without complete source code access, computing reachable states and probabilistic performance guarantees from simulation-based measurements [29]. These techniques scale to complex nonlinear systems while providing mathematically sound safety certificates.
Hardware-in-the-loop (HIL) validation establishes ground truth through synchronized testing of real control systems against high-fidelity simulated environments. Cyber-physical testbed co-simulation of power systems demonstrates real-time validation of frequency control and protection schemes using dual real-time simulators (Opal-RT, Typhoon HIL) communicating via ANSI C37.118 protocols [30]. Such testbeds validate system behavior under both nominal and emergency conditions before deployment.
9. Resource Scheduling, Dependencies, and Risk Modeling in Simulation
Complex operational systems exhibit multidimensional interdependencies spanning resource constraints, temporal scheduling, communication networks, and organizational hierarchies. Discrete-event simulation naturally captures these dependencies through explicit modeling of resource queues, preemption rules, and priority schemes. Two-tier hub-and-spoke architectures for freight logistics integrate GIS-based routing with stochastic demand simulation, enabling optimization of last-mile delivery under realistic variability [31].
Risk modeling within simulation frameworks enables probabilistic analysis of mission success under uncertainty. Evolutionary game theory combined with system dynamics provides mathematical formalization of multi-stakeholder decision-making under policy constraints and financial incentives [32]. Cross-regional coordinated dispatch of emergency supplies demonstrates how tripartite game models quantify the impact of governance mechanisms on cooperation equilibria—critical for disaster response planning.
Workforce restoration time analysis under stochastic demand and organizational decision latencies demonstrates the integration of queuing theory, heavy-tailed distributions, and decision-theoretic models into unified probabilistic frameworks [33]. These approaches capture how organizational delays and attrition volatility interact to determine system recovery capacity—directly relevant to mission-critical personnel management.
10. Integration of Machine Learning, AI, and Data-Driven Optimization
Modern simulation frameworks increasingly embed machine learning (ML) capabilities for automated parameter tuning, surrogate model development, and decision optimization. Reinforcement learning within agent-based electricity market simulations (ASSUME framework) enables adaptive bidding strategy exploration and market dynamics analysis without specifying dominant strategies a priori [34]. This approach shifts simulation from predictive analysis toward exploratory learning-based optimization.
Hybrid architectures combining discrete-event simulation with reinforcement learning demonstrate how RL agents can optimize within DES environments. The combination enables exploration of routing strategies in factories while maintaining accurate process flow and resource contention dynamics [35]. Commercial COTS DES software (Simio) now integrates ML algorithms without requiring external programming environments, democratizing combined simulation-ML approaches.
Process mining for digital twin development systematically extracts process models from historical execution data, creating high-quality behavioral models for anomaly detection and what-if analysis [36]. This data-driven approach to model construction overcomes manual specification bottlenecks, enabling rapid DT deployment in complex industrial environments.
AI-enhanced wargaming platforms integrate large language models with scenario orchestration and automatic inject generation to create adaptive, responsive training experiences [37]. These systems accelerate military decision-making processes while providing structured feedback aligned with empirical threat models, enabling more effective preparation for adversarial encounters.
Summary Table: Simulation Methodologies and Application Domains
Methodology
Key Characteristics
Primary Applications
Scalability
Validation Complexity
Discrete-Event Simulation
Event-driven, deterministic timing, explicit resource contention
Manufacturing, logistics, aerospace, defense
High (parallel)
Medium
Agent-Based Simulation
Autonomous heterogeneous agents, emergent behavior, decentralized control
Epidemiology, military ops, traffic, financial markets
High (GPU acceleration)
High
System Dynamics
Continuous aggregate flows, feedback loops, policy analysis
Strategic planning, HR development, resource management
Medium
Low
Hybrid Simulation
Combines DES+SD+ABM
Healthcare systems, supply chains, complex adaptive systems
Medium
Very High
Digital Twin
Real-time bidirectional physical-virtual coupling, ML integration
Cyber-physical systems, predictive maintenance, operational monitoring
Medium-High
Very High
References and Key Insights
The reviewed literature demonstrates that enterprise architectures can be transformed into executable simulation models through systematic application of DEVS formalism, hybrid modeling approaches, and digital twin technologies. Rigorous verification and validation methodologies—combining formal reachability analysis, hardware-in-the-loop testing, and continuous model evolution—establish trustworthiness in mission-critical applications. Emerging integration of machine learning, multi-agent reinforcement learning, and large language models within simulation frameworks opens new frontiers for automated optimization and adaptive decision support.
The synthesis reveals that no single methodology dominates across all application contexts; rather, problem characteristics—including aggregation level, actor heterogeneity, temporal resolution requirements, and data availability—determine optimal methodological choice. Defense simulations increasingly employ hybrid approaches integrating discrete-event operations models, system-level dynamics, and AI-driven tactical autonomy to capture multi-resolution phenomena. Digital twins represent a qualitative leap beyond traditional simulation by embedding real-time bidirectional coupling, enabling operational adaptation and knowledge accumulation throughout system lifecycle.
Future directions emphasize standardization (IEC 61499 function blocks, OPC-UA protocols), cross-platform interoperability, automated model construction from architecture specifications, and federated digital twin architectures supporting decentralized optimization across distributed systems. The convergence of simulation methodologies with blockchain-based provenance tracking, quantum computing for combinatorial optimization, and neuromorphic computing for real-time edge processing promises transformative capabilities for 21st-century mission engineering and operational research.
References
[1] 	J. L. Risco-Martín, S. Mittal, K. Henares, R. Cárdenas, and P. Arroba, “xDEVS: A toolkit for interoperable modeling and simulation of formal discrete event systems,” Software, Practice & Experience, Nov. 2022, doi: 10.1002/spe.3168.
[2] 	J. L. Risco-Martín, K. Henares, S. Mittal, L. F. Almendras, and K. Olcoz, “A unified cloud-enabled discrete event parallel and distributed simulation architecture,” Simulation modelling practice and theory, Mar. 2022, doi: 10.1016/j.simpat.2022.102539.
[3] 	A. Piccione, P. Andelfinger, and A. Pellegrini, “Hybrid speculative synchronisation for parallel discrete event simulation,” SIGSIM Principles of Advanced Discrete Simulation, June 2023, doi: 10.1145/3573900.3591124.
[4] 	G. Barakat, L. D’Agati, G. Tricomi, F. Longo, A. Puliafito, and G. Merlino, “Discrete event modeling and simulation approaches for IIoT,” Wseas Transactions on Business and Economics, Dec. 2024, doi: 10.37394/23207.2024.21.202.
[5] 	G. Herrera-Vidal, D. M. Sierra, H. C. Padilla, and J. Coronado-Hernández, “Discrete-event simulation for waste minimization and productivity enhancement in coupling manufacturing,” Applied Sciences, Feb. 2026, doi: 10.3390/app16041701.
[6] 	S. Brailsford, T. Eldabi, M. Kunc, N. Mustafee, and A. Osorio, “Hybrid simulation modelling in operational research: A state-of-the-art review,” European Journal of Operational Research, Nov. 2019, doi: 10.1016/J.EJOR.2018.10.025.
[7] 	X. Jiang, W. Wu, T. Liu, and L. Hao, “Research on the intelligent air-defense command and decision system based on multi-agent,” 2022 2nd Asia-Pacific Conference on Communications Technology and Computer Science (ACCTCS), Feb. 2022, doi: 10.1109/acctcs53867.2022.00034.
[8] 	Z. Lu, Y. Chen, L. Chen, Y. Ling, and F. Zhi, “Operational evaluation of UAV swarm precision strikes with agent based modeling and simulation,” 2024 10th International Conference on Big Data and Information Analytics (BigDIA), Oct. 2024, doi: 10.1109/BigDIA63733.2024.10808591.
[9] 	J. Lee, S. Shin, M. Park, and C. Kim, “Agent-based simulation and its application to analyze combat effectiveness in network-centric warfare considering communication failure environments,” Mathematical Problems in Engineering, Dec. 2018, doi: https://doi.org/10.1155/2018/2730671.
[10] 	M. Smilovitskiy et al., “FLAME-GPU for traffic systems: A scalable agent-based simulation framework,” Syst., May 2025, doi: 10.3390/systems13050376.
[11] 	H. Cai, Q. Xiong, and S. Lian, “Research on network security vulnerability risk contagion in software supply chain based on system dynamics,” PLoS ONE, Nov. 2025, doi: 10.1371/journal.pone.0335128.
[12] 	M. Zaslavska and S. Zaslavska, “ECONOMIC AND MATHEMATICAL MODELING OF HUMAN RESOURCE MANAGEMENT IN a HYBRID ENVIRONMENT BASED ON THE COMPETENCY-BASED APPROACH,” Economical, 2025, doi: 10.31474/1680-0044-2025-1(31)-54-66.
[13] 	S. Soni and R. Saini, “DIGITAL TRANSFORMATION OF PROJECT AND OPERATIONS MANAGEMENT IN SMART INFRASTRUCTURE AND ENGINEERING ENTERPRISES,” Archives for Technical Sciences, Dec. 2025, doi: 10.70102/afts.2025.1834.372.
[14] 	S. Slimp and M. E. Miller, “Human performance modeling architecture with HC-130J mission application,” Proceedings of the Human Factors and Ergonomics Society Annual Meeting, Aug. 2024, doi: 10.1177/10711813241261280.
[15] 	B. Chen and J. Li, “Research on UAV logistics operation mode based on low altitude air corridor hierarchical management,” 2025 Low-Altitude Economy Forum & International Conference on Low-Altitude Flight Technology and Unmanned Aerial Vehicle Application (LEF & ICLU), Sept. 2025, doi: 10.1109/LEFICLU65987.2025.11297260.
[16] 	J. Curry, “Professional wargaming: A flawed but useful tool,” Simulation & Gaming, Apr. 2020, doi: https://doi.org/10.1177/1046878120901852.
[17] 	N. Suri et al., “The angloval tactical military scenario and experimentation environment,” None, May 2018, doi: https://doi.org/10.1109/icmcis.2018.8398729.
[18] 	S. Xiao, Y. Wu, Y. Ma, Y. Cui, and G. Gong, “Military simulation big data: Background, state of the art, and challenges,” Mathematical Problems in Engineering, Jan. 2015, doi: https://doi.org/10.1155/2015/298356.
[19] 	T. Watkins, B. Davis, R. B. Ponnuru, and M. Azab, “AI-driven immersive emulation for tabletop scenarios,” Computing and Communication Workshop and Conference, Jan. 2026, doi: 10.1109/CCWC67433.2026.11393711.
[20] 	M. Borodin, T. Tkach, M. Yelisieieva, and K. Balakin, “MATHEMATICAL MODELING OF CONFLICT OF INTERESTS IN CONSTRUCTION BASED ON GAME THEORY, BIM, AND AI ANALYSIS,” Bulletin of Kharkov National Automobile and Highway University, Dec. 2025, doi: 10.30977/bul.2219-5548.2025.111.0.21.
[21] 	B. Soykan, G. Blanc, and G. Rabadi, “A proof-of-concept digital twin for real-time simulation: Leveraging a model-based systems engineering approach,” IEEE Access, 2025, doi: 10.1109/ACCESS.2025.3557367.
[22] 	M. Singh, E. Fuenmayor, E. P. Hinchy, Y. Qiao, N. Murray, and D. M. Devine, “Digital twin: Origin to future,” Applied System Innovation, May 2021, doi: https://doi.org/10.3390/asi4020036.
[23] 	A. Perišić and B. Perišić, “Digital twins verification and validation approach through the quintuple helix conceptual framework,” Electronics, Aug. 2024, doi: 10.3390/electronics13163303.
[24] 	Y. Wang, B. Zheng, and K. Xing, “Construction of high-precision digital twin model and real-time data fusion technology,” 2025 International Conference on Digital Analysis and Processing, Intelligent Computation (DAPIC), Feb. 2025, doi: 10.1109/DAPIC66097.2025.00103.
[25] 	A. Sayghe, “Digital twin-driven intrusion detection for industrial SCADA: A cyber-physical case study,” Italian National Conference on Sensors, Aug. 2025, doi: 10.3390/s25164963.
[26] 	H. Lei, Z. Fanli, W. Wei, S. Shuai, and H. Zhou, “Digital twin method and application practice of spacecraft system driven by mechanism data,” Digital Twin, Mar. 2024, doi: 10.12688/digitaltwin.17913.1.
[27] 	V. Chakma and W. Choi, “6G-enabled digital twin framework for real-time cyber-physical systems: An experimental validation with industrial bearing fault detection,” arXiv.org, Oct. 2025, doi: 10.48550/arXiv.2510.03807.
[28] 	J. Mertens and J. Denil, “Reusing model validation methods for the continuous validation of digital twins of cyber-physical systems,” Journal of Software and Systems Modeling, Oct. 2024, doi: 10.1007/s10270-024-01225-6.
[29] 	S. Bogomolov, C. Gomes, C. Isasa, S. Soudjani, P. Stankaitis, and T. Wright, “Reachability analysis of FMI models using data-driven dynamic sensitivity,” SIMULATION, July 2024, doi: https://doi.org/10.1177/00375497241261409.
[30] 	J. Riquelme-Dominguez, F. Gonzalez-Longatt, A. F. S. Melo, J. L. Rueda, and P. Palensky, “Cyber-physical testbed co-simulation real-time: Normal and abnormal system frequency response,” IEEE transactions on industry applications, Mar. 2024, doi: 10.1109/TIA.2023.3342764.
[31] 	Z. Lyu, D. J. Pons, J. Chen, and Y. Zhang, “Developing a stochastic two-tier architecture for modelling last-mile delivery and implementing in discrete-event simulation,” Syst., Nov. 2022, doi: 10.3390/systems10060214.
[32] 	Y. Qiu, M. Shi, X. Zhao, and Y. Jing, “System dynamics mechanism of cross-regional collaborative dispatch of emergency supplies based on multi-agent game,” Complex & Intelligent Systems, Mar. 2021, doi: 10.1007/s40747-021-00303-2.
[33] 	K. Parameswari, K. Kannan, and A. Menaga, “Workforce restoration time analysis for a two grade manpower system under heavy tail distribution and dual regime decisions latency,” Scientific Reports, Mar. 2026, doi: 10.1038/s41598-026-40851-6.
[34] 	N. Harder et al., “ASSUME: An agent-based simulation framework for exploring electricity market dynamics with reinforcement learning,” SoftwareX, May 2025, doi: 10.1016/j.softx.2025.102176.
[35] 	A. Greasley, “Architectures for combining discrete-event simulation and machine learning,” International Conference on Simulation and Modeling Methodologies, Technologies and Applications, July 2020, doi: 10.5220/0009767600470058.
[36] 	F. Vitale, S. Guarino, F. Flammini, L. Faramondi, N. Mazzocca, and R. Setola, “Process mining for digital twin development of industrial cyber-physical systems,” IEEE Transactions on Industrial Informatics, Jan. 2025, doi: 10.1109/TII.2024.3465600.
[37] 	W. N. Caballero and P. R. Jenkins, “On large language models in national security applications,” Stat, Mar. 2025, doi: https://doi.org/10.1002/sta4.70057.