1963–1994

Simulation-Based Global Optimization

Contemporary themes

During 1963 through 1994, Modeling and Simulation established itself as a unifying methodological framework across engineering, hydrology, traffic, and materials science, emphasizing robust numerical methods, simulation design, and experiment thinking. Stochastic optimization and simulated annealing emerged as core tools for solving hard engineering problems within simulation contexts, enabling probabilistic exploration of large design spaces and driving the broader metaheuristics movement. Advances in subgrid and turbulence modeling, multiscale approaches, and discrete-kinetic methods such as lattice-based schemes expanded the fidelity and efficiency of fluid and gas simulations. Engineering design optimization increasingly relied on design-of-experiments and heuristic control embedded in simulation workflows, linking theoretical models to practical engineering practice.

• Stochastic optimization and simulated annealing emerged as a core approach for solving hard engineering optimization problems, including TSP, parameter design, and scheduling in simulation contexts[1][5].
• Simulation serves as a unifying methodological framework across domains like hydrology, reservoirs, grain growth, and traffic, emphasizing numerical methods and simulation design across disciplines[6][7][8][10][11][20].
• Subgrid and turbulence modeling highlight multiscale approaches to flow phenomena, leveraging turbulence theory and subgrid procedures to improve simulations[7][12][19].
• Lattice-gas and Boltzmann-based discrete kinetic methods provide a computational framework for simulating fluids and gas dynamics[13][14].
• Engineering design optimization and planning employ design-of-experiments and heuristic control within simulation contexts[9][15][18].

Influential works

The period produced foundational texts and methodologies that defined Modeling and Simulation as a discipline. Simulated annealing catalyzed a paradigm shift toward global optimization by balancing exploration and exploitation through cooling schedules. The development of discrete-event simulation as a standard practice, together with modeling languages and toolkits associated with structural equation modeling and model dimension estimation, established lasting conventions for modeling, analysis, and data-driven decision making. Collectively, these innovations positioned Simulation as a central driver of cross-domain research and complex-system analysis for years to come.

• Simulated annealing: theory and applications. (1988) introduced the simulated annealing algorithm for global optimization, using probabilistic moves and cooling schedules to escape local optima; it catalyzed the metaheuristics revolution across engineering and computing[3].
• A cellular automaton model for freeway traffic (1992) introduced the cellular automaton modeling approach for highway flow; the Nagel-Schreckenberg rules yield realistic jam formation and wave-like dynamics, catalyzing CA-based traffic flow research and broader complex-systems modeling[11].
• Simulation Modeling and Analysis. (1983) standardizes the practice of discrete-event simulation, detailing model-building, input modeling, experimental design, and output analysis; it shaped education and industry practice in Modeling and Simulation (M&S) for decades[8].
• Lisrel 8: Structural Equation Modeling With the Simplis Command Language (1993) popularized Structural Equation Modeling (SEM); the SimplISL language enabled latent-variable specification and estimation, transforming social sciences and marketing research and establishing LISREL as a standard tool[2].
• Estimating the Dimension of a Model (1978) proposed how to determine the number of parameters a model should have; the insight informed early model-selection criteria and balance between fit and complexity, influencing modern statistical modeling and data analysis[1].

1995–2001

Integrated Complex Systems Modeling

Contemporary themes

During this period, integrated modeling of climate, atmosphere, and environmental processes unified efforts across aerosol dynamics, dust emission, land-surface physics, and nested climate modules to project climate and air quality. Methodological rigor intensified through model selection, evaluation, and verification/validation practices across domains, accompanied by the cross-domain adoption of social-system simulation approaches using structural equation modeling concepts. High-performance numerical methods and parallel computing underpinned scalable simulations, complemented by rigorous domain-specific CFD verification and validation.

• Integrated climate and atmospheric environmental modeling emerges as a unifying pattern, coupling aerosol dynamics, dust emission, land-surface physics, and nested climate modules to project climate and air quality[4][5][8][9][12][15][18][19].
• Model selection, evaluation, and sequential procedures drive a methodological pattern for identifying the best simulated system, paired with verification/validation practices across domains[1][2][10][11][16].
• Simulation of social and complex systems is advanced via structural equation modeling concepts and social simulations, indicating cross-domain adoption of complex-system approaches[1][7][20].
• High-performance numerical methods and parallel computing underpin scalable simulations, including parallel Monte Carlo techniques and large-eddy simulation contexts that stress performance and validation[3][16].
• Domain-specific CFD and engineering verification/validation emphasize rigorous, ship-scale CFD applications and systematic validation methodologies across engineering contexts[3][11].

Influential works

Structural Equation Modeling (SEM) established a canonical workflow for model specification, fit assessment, and interpretation, shaping cross-disciplinary practice. The companion texts Structural Equation Modeling With Lisrel, Prelis, and Simplis and the partial least squares approach expanded practical tools and reproducibility for SEM analyses. The MATLAB ODE Suite provided a unified collection of numerical integrators for ordinary differential equations, while Business Dynamics: Systems Thinking and Modeling for a Complex World popularized system dynamics as a rigorous framework for exploring feedback-rich systems.

• Principles and Practice of Structural Equation Modeling (1998) introduced a canonical synthesis of SEM methods—model specification, fit assessment, and interpretation—providing practical guidelines that normalized SEM workflows across disciplines[1].
• Structural Equation Modeling With Lisrel, Prelis, and Simplis: Basic Concepts, Applications, and Programming (1998) bridged theory and software by detailing LISREL-family tools and programming approaches, accelerating applied SEM adoption and reproducibility[3].
• The partial least squares approach for structural equation modeling (1998) popularized a component-based SEM alternative, enabling robust analyses with smaller samples and complex constructs; its enduring influence pervades marketing and social science research[10].
• Business Dynamics: Systems Thinking and Modeling for a Complex World (2000) popularized system dynamics as a rigorous framework for modeling feedback-rich organizations, translating stock-and-flow simulations into actionable business insights; it shaped education and practice in modeling complex systems[6].
• The MATLAB ODE Suite (1997) delivered a unified, accessible collection of numerical integrators for ordinary differential equations, dramatically expanding the practical toolkit for simulating dynamical systems and validating models across disciplines[5].

2002–2008

End-to-End Multiscale Simulation

Contemporary themes

During 2002–2008, modeling and simulation consolidated into integrated, end-to-end platforms that supported cross-domain, full-system simulations and streamlined engineering workflows. Multiscale environmental and climate modeling leveraged ensemble uncertainty quantification and cross-domain coupling, enabling large-ensemble analyses across atmospheric and oceanic systems. Agent-based and human-centric simulation became a core paradigm for social dynamics and healthcare, complemented by optimization, metamodelling, and experimental design to efficiently explore complex spaces and guide decisions.

• Simulation platforms and infrastructure enable cross-domain, end-to-end modeling through full-system simulators, standard dissemination venues, and integrated engineering workflows[7][15][20].
• Multiscale environmental and climate modeling with ensemble uncertainty quantification and cross-domain coupling, illustrating large-ensemble analyses, atmospheric/oceanic models, and multiscale numerical methods[4][10][13][16][17].
• Agent-based and human-centric simulation as a core paradigm for social dynamics and healthcare systems, spanning methodological foundations and domain-specific applications[6][9][14].
• Optimization, metamodelling and experimental design in simulation studies, focusing on efficient exploration, surrogate modeling, and advanced guidance concepts[1][5][8][11][12].
• Numerical methods and scalable modeling innovations driving accuracy and performance in engineering and geoscience simulations, including adaptive/scale-aware methods and robust numerical solvers[16][17][18][20].

Influential works

The period cemented the shift toward modular, extensible simulation ecosystems, enabling easier integration of heterogeneous models and data. It established cross-domain interoperability and standardized workflows, laying the groundwork for broader adoption of ensemble uncertainty, optimization-driven experimentation, and end-to-end pipelines in both research and industry. The innovations during this time catalyzed a long-lasting emphasis on scalable numerical methods, surrogate modeling, and automated design of experiments that continued to shape subsequent generations of simulation practice.

• Geant4—a simulation toolkit introduced a modular, object-oriented framework for Monte Carlo particle transport, enabling precise detector and instrument simulations across high-energy physics and medical applications, catalyzing cross-disciplinary adoption and ongoing ecosystem growth[2]
• First-principles simulation: ideas, illustrations and the CASTEP code presents CASTEP, a plane-wave DFT code enabling reliable ab initio materials simulations; popularized first-principles predictions for solids and spurred the broader computational materials science boom[5]
• YALMIP : a toolbox for modeling and optimization in MATLAB introduces a high-level modeling language that translates user specifications into convex and semidefinite programs, accelerating prototyping, education, and solver development in optimization[3]
• Model Predictive Control outlines an optimization-based control paradigm using receding horizon decisions to satisfy constraints and optimize performance; its practical success in industry and academia spurred broad research on real-time optimization and control theory[9]

2009–2015

Standardized Modeling Lifecycle

Contemporary themes

The period centralized a unified modeling lifecycle from conceptual design through uncertainty quantification, validation, and transparent reporting, reinforced by formal guidelines. Across domains, standardized methodologies—discrete-event, state-transition, and conceptual modeling—enabled cross-domain transfer and evaluation, while domain-specific pipelines for energy, climate, and environment highlighted end-to-end modeling chains under real-world constraints. Optimization- and meta-heuristics-driven approaches underpinned design and control within simulation frameworks, complemented by flow and turbulence modeling in urban environments using computational fluid dynamics to address wind and urban environmental studies.

• Formalization of the modeling lifecycle: from conceptualization and design to uncertainty quantification, validation, and transparent reporting, reflected in ISPOR-SMDM task forces and foundational modeling guidelines[1][3][6][10][12].
• Across domains, standardized modeling methodologies—discrete-event, state-transition, and conceptual modeling—are adopted to structure problems, enabling cross-domain methodological transfer and evaluation[1][2][7][10][20].
• Domain-specific modeling pipelines for energy, climate, and environment: from photovoltaic arrays to climate and weather systems, highlighting end-to-end modeling chains and domain constraints[4][8][13][14][15][17][18][20].
• Optimization- and meta-heuristics-driven modeling: emphasis on optimization-based design and control within simulation frameworks, abstracting from particular algorithms to underlying optimization paradigms[11][16][19].
• Flow and turbulence modeling in urban/architectural contexts using CFD and related methods, including guidelines and case studies for urban environments and wind studies[9][13][14][15].

Influential works

Breakthroughs in programmable cloud-oriented simulators and modular architecture, along with advances in structural equation modeling and high-throughput model testing, established reproducibility, scalability, and cross-disciplinary adoption. Tools like CloudSim and gem5 catalyzed experimental rigor in resource provisioning, architecture, and performance evaluation, while the photovoltaic array modeling framework unified cell-to-array predictions, accelerating energy-system optimization.

• CloudSim (2010) introduced a programmable toolkit for modeling and simulating cloud computing environments, enabling researchers to evaluate resource provisioning algorithms without real hardware[7].
• The gem5 simulator (2011) presented a modular, extensible computer architecture simulator with detailed, cycle-accurate models, fostering reproducible experiments across CPUs, memory, and accelerators[8].
• Principles and Practice of Structural Equation Modeling (2012) established SEM fundamentals—model specification, estimation, and fit assessment—spurring widespread adoption in social and behavioral sciences[3].
• jModelTest 2: more models, new heuristics and parallel computing (2012) expanded model-testing for phylogenetic analyses with new heuristics and parallel computing, accelerating model selection for substitution models and enabling scalable analyses[2].
• Comprehensive Approach to Modeling and Simulation of Photovoltaic Arrays (2009) integrated PV cell, module, and array models into a cohesive framework, advancing PV design, performance prediction, and system optimization[11].

2016–2024

Hybrid Data-Driven Modeling

Contemporary themes

During 2016–2024, Modeling and Simulation matured into a hybrid, data-informed discipline that blends physics-based simulation with data-driven learning and rigorous uncertainty quantification. Metaheuristic optimization emerged as a central framework for parameter estimation, forecasting, and control across engineering and energy systems, enabling robust performance under diverse objectives and constraints. Predictive maintenance and Remaining Useful Life forecasting increasingly combine model-based filters with data-driven forecasts, while climate, environmental, and autonomous-system modeling leveraged tuned atmospheric models, land-atmosphere coupling, hybrid time-series forecasting, and high-fidelity simulation environments to push simulation fidelity and decision-support.

• Metaheuristic optimization as a unifying modeling paradigm for parameter estimation, forecasting, and control across engineering and energy systems, enabling robust performance with diverse objective handling[1][2][3][7][12][18].
• Model-based predictive maintenance and Remaining Useful Life (RUL) forecasting combining data-driven techniques and model-based filters for machinery, batteries, and rolling elements[5][16][17].
• Climate and atmospheric modeling includes tuning philosophy, integrated atmospheric models, and land-atmosphere coupling to improve climate projections[9][13][19].
• Forecasting and time-series modeling for environmental/energy systems using hybrid/statistical ML approaches, combining decomposition, ensemble methods, and optimization for wind, solar, and load forecasts[3][6][7][12][14].
• Autonomous systems in simulation and control, including high-fidelity simulation environments and path-planning/navigation methods for vessels/vehicles, coupled with robust control strategies[4][11][18][20].

Influential works

The period solidified the shift toward integrated platforms that unify modeling, experimentation, and validation across domains. Foundational works such as the Scenario Model Intercomparison Project for CMIP6 established standardized multi-model experiments and data-sharing practices that accelerated cross-model validation and confidence in projections. Foundational computational tools—Simulation and the Monte Carlo Method, LAMMPS, the atomic simulation environment, and AirSim—defined scalable, accessible workflows for materials science, atomistic modeling, and autonomous-systems testing, shaping subsequent research and applications.

• The Scenario Model Intercomparison Project (ScenarioMIP) for CMIP6 (2016)[11] established standardized multi-model experiments and data sharing, enabling robust climate projections and cross-model validation that accelerated confidence and collaboration in climate modeling.
• Simulation and the Monte Carlo Method (2016)[5] offers foundational Monte Carlo frameworks and practical guidance for stochastic simulation, shaping uncertainty quantification and decision-support across Modeling and Simulation.
• LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales (2022)[2] introduced a scalable, modular molecular dynamics framework that became a standard workhorse for computational materials science.
• The atomic simulation environment—a Python library for working with atoms (2017)[6] provides a Pythonic API for constructing, simulating, and analyzing atomic systems, democratizing atomistic modeling and rapid prototyping.
• AirSim: High-Fidelity Visual and Physical Simulation for Autonomous Vehicles (2017)[7] delivers realistic sensor physics and environments, enabling end-to-end testing and reinforcement learning research in autonomous vehicle perception and control.