AGENTSIM: Agent-Based Modeling and Simulation Platform
SHORT SUMMARY
The Agent-Based Modeling and Simulation Platform provides a dedicated computational environment for designing, executing, and analysing large-scale agent-based models (ABMs) across a broad range of complex systems. Built on NVIDIA RTX 6000 ADA and H100 NVL GPUs with high-capacity storage and multi-core CPU infrastructure, AGENTSIM is configured to support the parallel execution demands of simulations involving large agent populations, extended time horizons, and high-dimensional parameter spaces. AGENTSIM is specifically oriented toward the study of emergent phenomena arising from the interactions of autonomous agents operating under defined behavioural rules, making it suited to domains where system-level behaviour cannot be derived analytically. The platform supports research and applied use cases across social and economic systems, urban and logistics planning, epidemiological modelling, and public policy simulation. This testbed contributes to the CITADELS Framework by providing accessible infrastructure for computationally intensive simulation research that bridges academic modelling and real-world decision support applications.
HOSTING INSTITUTION AND PI INFO
| Name of Host Organization | NOVA Information Management School (NOVA IMS), Universidade Nova de Lisboa |
| Department or Lab | MagIC (Information Management Research Center) – the NOVA Information Management School research and development center |
| Name of Building | Manuel Vilares Building |
| Physical Address | Campus de Campolide, 1070-312 Lisboa |
| Website Links | https://www.novaims.unl.pt/ |
| Institutional contact name | Cristina Oliveira |
| Institutional contact email | magic@novaims.unl.pt |
APPLICATION CASES
| Application case: | Short description: |
| Urban Mobility and Logistics Modelling | Simulate the movement of individuals, vehicles, and goods within urban environments to study congestion, last-mile delivery efficiency, and the impact of infrastructure or policy changes on system-level mobility outcomes, supporting evidence-based urban planning and logistics optimisation. |
| Social and Economic Systems Simulation | Model the emergence of collective economic behaviour, market dynamics, or social norm diffusion from agent-level interactions, enabling the study of phenomena such as inequality, adoption of innovations, and the effects of regulatory interventions in complex adaptive systems. |
| Policy and Scenario Analysis for Decision Support | Use AGENTSIM as a virtual laboratory for testing the likely outcomes of competing policy options across domains such as energy transition, public health, or urban development, providing stakeholders with simulation-based evidence to support informed decision-making under uncertainty. |
| Sensitivity Analysis and Parameter Space Exploration | Leverage the high-throughput compute infrastructure to conduct systematic sensitivity analyses and large-scale parameter sweeps across ABM configurations, enabling rigorous uncertainty quantification and model validation in support of reproducible simulation research. |
| MSc and PhD Research in Complex Systems | The platform supports graduate students conducting dissertation and research projects in computational social science, complex systems, and simulation modelling, providing the CPU and GPU capacity required to run and replicate experiments at scales beyond standard desktop computing. |
POTENTIAL STAKEHOLDERS
Non-academic stakeholders
Industrial Partners, Startups, Professional Associations, SMEs, Government Bodies
Academic stakeholders
PhD students, MSc students, Researchers