Fragile, Resilient, Asymmetric: An Agent-Based Stress Test of European Cooperation under Geopolitical Shocks (2024-2040) (1.0.0)
An empirically calibrated agent-based model of cooperation among 14 EU member states. Adaptive state-agents update their cooperation propensity through behavioural inertia, influence along the observed intra-EU trade network (IMF bilateral flows), and repeated-game payoff indicators built from verified Eurostat, Eurobarometer and IMF data (2021-2024). An anchored logistic mapping makes the observed configuration stationary in the absence of shocks, so outcomes read as deviations from the empirical baseline. The model stress-tests European cooperation to 2040 under five scenarios of increasing severity, from a baseline to a Taiwan Strait crisis counterfactual, with 1,000 Monte Carlo replications and a full sensitivity suite (one-factor-at-a-time, joint parameter sampling, breaking-point analysis, alternative functional form). Documented with the ODD protocol; self-testing and fully reproducible under fixed seeds.
Release Notes
First public release (v1.0.0), companion to a manuscript submitted to the Journal of Artificial Societies and Social Simulation. Includes the full model core with built-in self-tests (modello_b.py), all experiment scripts (scenario runs, OFAT and joint-parameter sensitivity, breaking-point analysis, alternative functional form), verified input data with per-series source and confidence flags, provenance scripts for model verification and data extraction, the five paper figures, and the complete ODD description. All runs are reproducible under fixed seeds.
Associated Publications
Brugnoletti, A. (2026). Fragile, Resilient, Asymmetric: An Agent-Based Stress Test of European Cooperation under Geopolitical Shocks (2024-2040). Manuscript submitted to the Journal of Artificial Societies and Social Simulation.
Fragile, Resilient, Asymmetric: An Agent-Based Stress Test of European Cooperation under Geopolitical Shocks (2024-2040) 1.0.0
An empirically calibrated agent-based model of cooperation among 14 EU member states. Adaptive state-agents update their cooperation propensity through behavioural inertia, influence along the observed intra-EU trade network (IMF bilateral flows), and repeated-game payoff indicators built from verified Eurostat, Eurobarometer and IMF data (2021-2024). An anchored logistic mapping makes the observed configuration stationary in the absence of shocks, so outcomes read as deviations from the empirical baseline. The model stress-tests European cooperation to 2040 under five scenarios of increasing severity, from a baseline to a Taiwan Strait crisis counterfactual, with 1,000 Monte Carlo replications and a full sensitivity suite (one-factor-at-a-time, joint parameter sampling, breaking-point analysis, alternative functional form). Documented with the ODD protocol; self-testing and fully reproducible under fixed seeds.
Release Notes
First public release (v1.0.0), companion to a manuscript submitted to the Journal of Artificial Societies and Social Simulation. Includes the full model core with built-in self-tests (modello_b.py), all experiment scripts (scenario runs, OFAT and joint-parameter sensitivity, breaking-point analysis, alternative functional form), verified input data with per-series source and confidence flags, provenance scripts for model verification and data extraction, the five paper figures, and the complete ODD description. All runs are reproducible under fixed seeds.