Agent-based modelling: A tool for addressing the complexity of environment and development policy issues

The document introduces agent-based modelling (ABM) as a bottom-up approach to study complex social and environmental systems by simulating the interactions of autonomous agents. It argues that ABM is particularly useful for problems involving decentralized interactions and non-linear dynamics where traditional top-down models fail. The authors highlight ABM's potential as a stakeholder engagement tool and its ability to be integrated with qualitative and quantitative methods like GIS and social network analysis. Four case studies are presented: disaster resilience in Turkey, agroforestry adaptation in Cameroon, biodiesel transitions in Malaysia, and artisanal fisheries management in Kenya.

• Agent-based modelling (ABM) addresses complexity by describing social systems at the micro-level of autonomous actors, using rules and internal data to simulate how individual interactions produce aggregate macro-level properties.
• ABM is most effective for problems characterized by dense, decentralized interactions, feedbacks, and unpredictable phenomena, though it is resource-intensive and can take a year or more to implement and analyze.
• Participatory ABM serves as a strong stakeholder engagement method because its logic is more intuitive for non-experts than mathematical models, facilitating the co-production of knowledge and social learning.
• ABM can be integrated with other research methods to provide a fuller context; it complements qualitative narrative data, quantitative parameter estimation, and relational data from GIS and social network analysis (SNA).
• In Cameroon, ABM was used to simulate farmer decision-making regarding cocoa production and agroforestry, identifying conditions that lead to successful adaptation versus maladaptation in the face of climate change and soil fertility decline.
• A system dynamics model for Malaysia's biodiesel industry explored the trade-offs between government incentives (subsidies and mandates) and risks from external shocks like fluctuating crude oil and palm oil prices.
• In Kenya, pilot ABMs of artisanal fisheries were used to highlight local perceptions of beach management units (BMUs) and facilitate knowledge transfer between local users and district-level managers.
• An ABM application in Turkey focused on disaster preparedness, suggesting that national insurance planning could positively influence preparedness intentions by limiting the 'hazard denial' coping mechanism.

APA

Taylor, R., Besa, M. C., & Forrester, J. (2016). Agent-based modelling: A tool for addressing the complexity of environment and development policy issues. Stockholm Environment Institute. https://www.sei.org/mediamanager/documents/Publications/SEI-WP-2016-12-Agent-based-modelling-MD-final.pdf

Chicago

Taylor, Richard, Mònica Coll Besa, and John Forrester. Agent-based modelling: A tool for addressing the complexity of environment and development policy issues. Stockholm Environment Institute, 2016. https://www.sei.org/mediamanager/documents/Publications/SEI-WP-2016-12-Agent-based-modelling-MD-final.pdf.

Wikipedia

{{cite report |last1=Taylor |first1=Richard |last2=Besa |first2=Mònica Coll |last3=Forrester |first3=John |title=Agent-based modelling: A tool for addressing the complexity of environment and development policy issues |publisher=Stockholm Environment Institute |date=November 2016 |url=https://www.sei.org/mediamanager/documents/Publications/SEI-WP-2016-12-Agent-based-modelling-MD-final.pdf |access-date=23 October 2026 |via=Climate Insights Directory}}

BibTeX

@techreport{taylor2016agentbased, author = {Taylor, Richard and Besa, Mònica Coll and Forrester, John}, title = {{Agent-based modelling: A tool for addressing the complexity of environment and development policy issues}}, institution = {Stockholm Environment Institute}, year = {2016}, month = nov, url = {https://www.sei.org/mediamanager/documents/Publications/SEI-WP-2016-12-Agent-based-modelling-MD-final.pdf}, urldate = {2026-10-23}, note = {Indexed by Climate Insights Directory} }