Platform-as-a-Bank vs. Bank-as-a-Platform : ce que les plateformes nous enseignent en matière d’analyse stratégique
Published in Mind Fintech.fr
Transferts de risques d’impayés au cœur de l’actualité
Published in Economie Matin
Causal inference and improvement of customer knowledge models (Uplift)
Our work demonstrates the limitations of Big Data in the field of customer knowledge. It has paved the way for new research avenues on Machine Learning, especially Preference Learning. Applied to target marketing, this analytical technique makes it possible to characterize profit-generating customers and can be extended to other fields (health, etc.).
Digital twins of warehouses – intralogistics
A digital twin is a type of simulation that relies on the principle of a virtual clone of a physical system or process. While digital twins are broadly used in industrial and scientific settings to support product-associated decision-making, little research work has looked into the use of a digital twin to support process-associated decision-making. Management of the carbon footprint and greenhouse gas emissions in intralogistics is a process that would benefit from the digital twins we are currently developing.
Modelling of the impact of physical and transition risks on the solvency of insurance companies
We work on improving the integration of the non-stationary and multivariate nature of climate variables in the statistics field. Starting from the newly established mathematical framework, we allow non-life insurers to evaluate their risks pertaining to climate change by considering the inter-dependencies of risks.