The book adopts a "problem-driven, multi-disciplinary integration, practice-oriented" approach. It starts with real social issues (cross-cultural crowd safety, supply chain risks), integrates AI, complex system science, and behavioral economics, and follows a "theory modeling-algorithm development-platform construction-scenario verification" closed loop to ensure theoretical rigor and practical applicability.
It breaks three traditional limitations: integrates agent value functions with causal mechanisms to solve static modeling issues; develops a parallel "sub-regional evolution algorithm" to balance large-scale simulation and interpretability; embeds cultural dimensions into modeling to enhance cross-cultural adaptability, avoiding single-scenario bias. It covers five modules: heterogeneous agent modeling (theoretical foundation), interpretable behavior pattern modeling (core technology), social system simulation platform construction (tool support), scenario applications (crowd simulation, supply chain risk management), and guarantee systems (international cooperation, risk control).
The target audience may include academic researchers (graduates, scholars), industry practitioners (smart city, emergency management staff), and policy makers. Content level: progressive, with basic theories for beginners, advanced technologies for developers, and application cases for practitioners.
The book adopts a "problem-driven, multi-disciplinary integration, practice-oriented" approach. It starts with real social issues (cross-cultural crowd safety, supply chain risks), integrates AI, complex system science, and behavioral economics, and follows a "theory modeling-algorithm development-platform construction-scenario verification" closed loop to ensure theoretical rigor and practical applicability.
It breaks three traditional limitations: integrates agent value functions with causal mechanisms to solve static modeling issues; develops a parallel "sub-regional evolution algorithm" to balance large-scale simulation and interpretability; embeds cultural dimensions into modeling to enhance cross-cultural adaptability, avoiding single-scenario bias. It covers five modules: heterogeneous agent modeling (theoretical foundation), interpretable behavior pattern modeling (core technology), social system simulation platform construction (tool support), scenario applications (crowd simulation, supply chain risk management), and guarantee systems (international cooperation, risk control).
The target audience may include academic researchers (graduates, scholars), industry practitioners (smart city, emergency management staff), and policy makers. Content level: progressive, with basic theories for beginners, advanced technologies for developers, and application cases for practitioners.
Desheng Wu
Multi-agent Technology Social Complex System Heterogeneous Agent Modeling Global Supply Chain Disturbance Simulation Cross-cultural Crowd Behavior Simulation High-concurrency Simulation Platform Policy Intervention Optimization System Modelling Computational Social Science Agent Behavior Interpretability