Social Simulations: from Agent-Based Modeling to Digital Twins

2026-07-15

AI Summary

This study addresses the paradigm shift in social simulation—from exploring abstract mechanisms toward faithfully replicating high-fidelity real-world systems—by proposing “social digital twins” as a next-generation simulation framework. Integrating rule-based agent modeling, large language models, and data-driven approaches, this framework enables dynamic, fine-grained representations of specific socio-technical systems. The work systematically traces the three-stage evolution of social simulation methodologies, clarifying the applicability, strengths, and limitations of each approach. In doing so, it establishes a theoretical foundation and methodological guidance for constructing high-fidelity social digital twins that accurately mirror real-world complexities.

Abstract

This book chapter covers the evolution of social simulation from classical agent-based models, in which agents interact according to explicitly defined behavioral rules, to AI-enhanced simulations based on Large Language Models and, ultimately, Social Digital Twins: high-fidelity, data-driven representations of real-world socio-technical systems. Along this trajectory, we discuss the main methodological foundations, applications, advantages, and limitations of each paradigm, highlighting the progressive shift from abstract models designed to investigate general social mechanisms toward increasingly realistic computational representations of specific social systems.

Problem

• Social Simulation
• Agent-Based Modeling
• Digital Twins
• Large Language Models
• Socio-Technical Systems

Innovation

• Agent-Based Modeling
• Large Language Models
• Social Digital Twins
• AI-enhanced Simulation
• Socio-technical Systems