An agent-based model of irregular warfare in which civilians adapt their alignment in response to local violence, security presence, and territorial control. The simulation explores how decentralized interactions generate spatial patterns of loyalty, conflict dynamics, and stabilization.

This model examines the potential impact of market collapse on the economy and demography of fishing households in the Logone Floodplain, Cameroon.

Flibs’NLogo is an agent-based simulation implemented in NetLogo that models the evolution of perfect predictors through a genetic algorithm. The agents, called flibs (finite living blobs), are finite‑state automata whose behaviour is encoded in circular chromosomes. They inhabit a “primordial computer soup” and are tasked with anticipating a user‑defined periodic binary sequence. Each generation consists of 100 evaluation cycles, during which a flib’s fitness is incremented each time its output correctly matches the next environmental signal.
Reproduction follows an elitist scheme: a donor (current fittest individual) replaces a randomly chosen recipient either by cloning (complete genome substitution) or by bacterial‑like conjugation (unidirectional horizontal transfer of a random chromosome segment). A stochastic mutagenesis operator introduces point mutations in genes, while the reproductive strategy gene can also switch under a mixed-reproduction regime. Population dynamics are monitored via genomic diversity indices (Shannon‑Wiener, Simpson), a phenotypic simpleness metric that distinguishes the low number of states actually used from the genomic potential.
The model serves as a digital evolutionary laboratory for exploring the interplay among bounded rationality, collective adaptation, and the emergence of anticipatory behaviour. By linking evolutionary computation with cognitive concepts, Flibs’NLogo investigates fundamental transitions from reactive to predictive systems and allows for testing whether populations evolve toward minimal necessary complexity or exhibit an intrinsic drift toward structural elaboration.

Country-by-Country Reporting and Automatic Exchange of Information have recently been implemented in European Union (EU) countries. These international tax reforms increase tax compliance in the short term. In the long run, however, taxpayers will continue looking abroad to avoid taxation and, countries, looking for additional revenues, will provide opportunities. As a result, tax competition intensifies and the initial increase in compliance could reverse. To avoid international tax reforms being counteracted by tax competition, this paper suggests bilateral responsive regulation to maximize compliance. This implies that countries would use different tax policy instruments toward other countries, including tax and secrecy havens.

To assess the effectiveness of fully or partially enforce tax policies, this agent based model has been ran many times under different enforcement rules, which influence the perceived enforced- and voluntary compliance, as the slippery-slope model prescribes. Based on the dynamics of this perception and the extent to which agents influence each other, the annual amounts of tax evasion, tax avoidance and taxes paid are calculated over longer periods of time.

The agent-based simulation finds that a differentiated policy response could increase tax compliance by 6.54 percent, which translates into an annual increase of €105 billion in EU tax revenues on income, profits, and capital gains. Corporate income tax revenues in France, Spain, and the UK alone would already account for €35 billion.

This model was utilized for the simulation in the paper titled Effect of Network Homophily and Partisanship on Social Media to “Oil Spill” Polarizations. It allows you to examine whether oil spill polarization occurs through people’s communication under various conditions.

・Choose the network construction conditions you’d like to examine from the “rewire-style” chooser box.
・Select the desired strength of partisanship from the “partisanlevel” chooser box. You can also set the strength manually in the code tab.
・You can set the number of dynamic topics using the “number-of-topics” slider.
・Use the “divers-of-opinion” slider to set the number of preference types for each dynamic topic.
…

MASTOC-LLM extends the classic Multi-Agent System Tragedy of the Commons (MASTOC) model by replacing hard-coded behavioral rules with autonomous decision-making powered by large language models (LLMs). Three heterogeneous agents manage herds of cows on a shared grassland commons. Each tick, an agent receives a structured prompt describing current resource levels, its own herd size, peer behavior, and — optionally — a rolling memory of recent rounds and messages from neighboring agents. The LLM returns a stocking decision (add, remove, or hold cows) together with a natural-language rationale and, when communication is enabled, a short message to broadcast to peers.

The model is designed to test whether LLM agents spontaneously develop Ostrom-style common-pool resource governance (mutual monitoring, graduated sanctions, graduated rule revision) or instead fall into identifiable failure modes. Preliminary experiments with Claude Haiku 4.5, GPT-5.4-mini, and DeepSeek R1:32b have revealed four recurring collapse patterns — Cooperative Paralysis, Defection Cascade, Overshoot-Panic, and Hybrid Architecture Failure — whose onset timing is sensitive to memory length, inter-agent communication, and the post-training alignment approach of the underlying model.

MASTOC-LLM is intended as a laboratory for generative agent-based modelling (GABM) methodology: it provides a clean, well-understood commons baseline against which LLM behavioral hypotheses can be systematically tested and compared across models, parameter sweeps, and alignment regimes.

Simulations based on the Axelrod model and extensions to inspect the volatility of the features over time (AXELROD MODEL & Agreement threshold & two model variations based on the Social identity approach)
The Axelrod model is used to predict the number of changes per feature in comparison to the datasets and is used to compare different model variations and their performance.

Input: Real data

…

What is it?

This model demonstrates a very simple bidding market where buyers try to acquire a desired item at the best price in a competitive environment

…

ReMoTe-S is an agent-based model of the residential mobility of Swiss tenants. Its goal is to foster a holistic understanding of the reciprocal influence between households and dwellings and thereby inform a sustainable management of the housing stock. The model is based on assumptions derived from empirical research conducted with three housing providers in Switzerland and can be used mainly for two purposes: (i) the exploration of what if scenarios that target a reduction of the housing footprint while accounting for households’ preferences and needs; (ii) knowledge production in the field of residential mobility and more specifically on the role of housing functions as orchestrators of the relocation process.

This code simulates the WiFi user tracking system described in: Thron et al., “Design and Simulation of Sensor Networks for Tracking Wifi Users in Outdoor Urban Environments”. Testbenches used to create the figures in the paper are included.