Resource description:
To foster sustainable land use and management, it is crucial – but challenging – to enhance our understanding of how policy interventions influence decision-making actors and how these interactions can be effectively modelled. Key challenges include endowing modelled actors with autonomy, accurately representing their relational network structures, and managing the often unstructured information exchange among them. Large language models (LLMs) offer new ways to address these challenges through the development of agents that are capable of mimicking reasoning, reflection, planning, and action. We present InsNet-CRAFTY (Institutional Network – Competition for Resources between Agent Functional Types) v1.0, a multi-LLM-agent model with a polycentric institutional framework coupled with an agent-based land system model. The institutional framework includes a high-level policymaking institution, two lobbyist organizations, two operational institutions, and two advisory agents. For exploratory purposes, illustrative numerical experiments simulating two competing policy priorities are conducted: increasing meat production versus expanding protected areas for nature conservation. We find that the high-level institution tends to avoid radical changes in budget allocations and adopts incremental policy goals for the operational institutions, but it leaves an unresolved budget deficit in one institution and a surplus in another. This is due to the competing influence of multiple stakeholders, which leads to the emergence of a path-dependent decision-making approach. Despite errors in information and behaviours by the LLM agents, the network maintains overall behavioural believability. The results highlight both the potential and the risks of using LLM agents to simulate policy decision-making. While LLM agents demonstrate high flexibility and autonomy in modelling human decision-making and institutional dynamics, their integration with existing land use models is complex, requiring careful workflow design to ensure reliability. These insights contribute to advancing land system modelling and the broader field of institutional analysis, providing new tools and methodologies for researchers and policymakers.
Author/Contact:
Yongchao Zeng, Calum Brown, Mohamed Byari, Joanna Raymond, Thomas Schmitt, Mark Rounsevell