AI Case Study: Estimating CO2 Emissions in Non-Life Insurance Claim Management

Authors: Cerchiara Rocco Roberto (corresponding author), Abiuso Marco (Esplores), Kathib Angelo (Esplores), Capezzera Vito (Mediolanum Assicurazioni), Bisognano Rosangela (Mediolanum Assicurazioni), Nardi Lorenzo (Mediolanum Assicurazioni), Alessandro Ferro (All Consulting)

At COP 21 in Paris, on 12 December 2015, Parties to the United Nations Framework Convention on Climate Change (UNFCCC) reached a landmark agreement to combat climate change and to accelerate and intensify the actions and investments needed for a sustainable low carbon future. The Paris Agreement builds upon the Convention and – for the first time – brings all nations into a common cause to undertake ambitious efforts to combat climate change and adapt to its effects, with enhanced support to assist developing countries to do so. As such, it charts a new course in the global climate effort.

The transport sector is responsible for about a quarter of Europe's total CO2 emissions, 71.7% of which come from road transport, according to the European Environment Agency. In an effort to limit CO2 emissions, the EU has set a target of reducing transport emissions by 60% from 1990 levels by 2030. The EU aims to achieve a 90% reduction in greenhouse gas emissions from transport by 2050 compared to 1990 levels. This forms part of efforts to reduce CO2 emissions and achieve climate neutrality by 2050 under the European Green Deal roadmap.

The production of our food generates 37% of the total Greenhouse Gas Emissions (abbreviated as emitted on the planet, or 17.3 gigatonnes of CO2 equivalent per year, according to a Nature Food study published in 2021. In comparison, transport emit 13.4 gigatonnes of CO2 per year, or 28% of anthropogenic emissions.

There is great interest in how the growth of artificial intelligence and machine learning (ML) may affect global GHG emissions. However, such emissions impacts remain uncertain, owing in part to the diverse mechanisms through which they occur, posing difficulties for measurement and forecasting. Cerchiara R.R., Esplores and Mediolanum Assicurazioni developed a joint pilot study in order to estimate the CO2-emission due to the claim management of a Non-Life insurance company by using Artificial Intelligence. The data were provided by All Consulting in regards of the Italian market.In particular we introduce a framework for describing the effects of ML on GHG emissions and we identify priorities for impact assessment and scenario analysis, and suggest policy levers for better management all the activities about claim processes of a Non-Life insurance company.

References

Ahmeridad, B., Cattaneo, M., Luciano, E., Kenett, R. (2023). AI and Adversarial AI in insurance: from an example to some remedies. Risks, https://europepmc.org/article/ppr/ppr557169EIOPA (2019).

Big Data Analytics in motor and health insurance: a Thematic Review. EIOPA (2021). Artificial Intelligence Governance Principles: Towards Ethical and Trustworthy Artificial Intelligence in the European Insurance Sector.

Eling, M., Nuessle, D., and Staubli, J. (2021). The impact of artificial intelligence along the insurance value chain and on the insurability of risks. The Geneva Papers on Risk and Insurance - Issues and Practice. doi:https://doi.org/10.1057/s41288-020-00201-7

Kaack L. H., Donti P. L., Strubell E., Kamiya G., Creutzig F., Rolnick D. (2022). Aligning artificial intelligence with climate change mitigation. Nature Climate Change, volume 12, 518–527

United Nations Framework Convention on Climate Change (2015). The Paris Agreement. https://unfccc.int/most-requested/key-aspects-of-the-paris-agreement