AI-LLM Empowering Supply Chain Resilience: An Analysis Based on a Manufacturer-Supplier-Consumer Tripartite Evolutionary Game Model
Description
This dataset comprises the numerical simulation results and associated code from the study entitled "AI-LLM Empowering Supply Chain Resilience: An Analysis Based on a Tripartite Evolutionary Game Model of Manufacturer-Supplier-Consumer." The study examines how artificial intelligence and large language models (AI-LLM) enhance supply chain resilience by influencing the strategic interactions among manufacturers, suppliers, and consumers. The core methodology employs a tripartite evolutionary game model, which simulates the dynamic strategic behaviors of the three stakeholders under various parameter configurations that reflect the impact of AI-LLM technology. Data used in this research are collected from prominent Chinese online platforms—including Toutiao, Sohu, and news forums—providing empirical support for parameter calibration in the model. Furthermore, the simulation codes for generating figures and solving equilibrium points were primarily implemented manually in MATLAB, encompassing time series data that depict the evolution of strategy adoption probabilities, along with the corresponding computational scripts necessary to reproduce the model outcomes. The dataset is systematically organized and comprehensive, thereby offering valuable reference material and practical insights for researchers working in the field of game theory.
Files
Steps to reproduce
This study employs a tripartite evolutionary game-theoretic approach to investigate how artificial intelligence and large language models (AI-LLM) enhance supply chain resilience through their influence on the strategic interactions among manufacturers, suppliers, and consumers. The empirical data utilized in this research are drawn from prominent Chinese online platforms—including Toutiao, Sohu, and news forums—as well as scholarly publications, which serve as the foundation for parameter calibration. The computational code, primarily developed manually using MATLAB, supports the generation of visualizations, the computation of equilibrium points, and the full replication of the game dynamics. This comprehensive dataset includes all derived data underlying the figures and equations, enabling direct reproducibility. As such, it provides valuable reference material and practical insights for researchers in the field of game theory.
Institutions
- Anhui University of Finance and Economics