Andre Orelien Chuisseu Tchuisseu

Doctorat en actuariat,
Université Laval

Direction de recherche:

  • Etienne Marceau, professeur à l’École d’actuariat de l’Université Laval
  • Hélène Cossette, professeure à l’École d’actuariat de l’Université Laval
  • Jean-Philippe Boucher, professeur au Département de mathématiques de l’Université du Québec à Montréal

Publications

  • A.O.Chuisseu Tchuisseu, J.-P. Boucher, H.Cossette & E.Marceau (2026), Large-Scale Actuarial Fire-Spreading Models on Tree-Structured Graphical Networks, [Projet en cours]

    This paper extends a previously developed actuarial fire-spreading framework—initially limited to small tree structures—to buildings composed of an arbitrary number of interconnected units. Fire propagation is modeled using a probabilistic tree-based structure, where damage spreads recursively across connected nodes according to edge-specific propagation probabilities and node-level characteristics. The proposed approach derives the full distribution of aggregate fire losses using probability generating functions combined with Fast Fourier Transform techniques, allowing for exact and scalable computation without reliance on simulation. The framework further incorporates heterogeneous loss distributions, ignition probabilities, and flashover effects across units, providing a more realistic representation of building risk. Numerical experiments highlight the critical role of network topology, showing that highly connected (hub-like) structures generate larger and more variable losses than linear configurations, while sensitivity analyses identify structurally critical connections driving systemic risk.

  • A.O.Chuisseu Tchuisseu, J.-P. Boucher, H.Cossette & E.Marceau (2026), Application of an Actuarial fire-spreading model for assessing fire risk in farms, [Projet en cours]

    This paper extends an actuarial fire-spreading model (AFSM) to the assessment of fire risk in farm environments by integrating contagion-based propagation mechanisms derived from distance-dependent models. Farms are represented as spatially connected structures whose configuration and ignition location determine fire spread dynamics. By embedding these mechanisms into a tree-based analytical framework and conditioning on the origin of the fire, the approach enables the exact computation of aggregate loss distributions, risk measures, and insurance premiums without reliance on simulation. Beyond applying the model to farms, the paper also investigates what minimal structural information a farmer must provide about the site—such as the number, arrangement, and relative distances of buildings—to allow a pricing model to remain as accurate and fair as possible. A numerical application highlights the impact of structural configurations on loss variability and tail risk, and shows how the proposed framework complements and generalizes existing contagion models in insurance applications.

  • A.O.Chuisseu Tchuisseu, J.-P. Boucher, H.Cossette & E.Marceau (2026), Actuarial Fire-Spreading Model Based on Tree-Structured Graphical Models, Soumis pour publication

    In this paper, we propose a general framework to model fire propagation in spatially connected environments, such as adjacent buildings or interconnected rooms within a single dwelling. Each such environment is represented as a tree-like structure, where units (e.g., rooms or substructures) are connected by edges that reflect potential paths for fire spread. Our analysis considers tree-structured sites with up to five units. Since the cases with one, two, and three units are straightforward, we focus on the more intricate configurations with four and five units. For each configuration, we provide a stochastic representation of the total fire losses and analyze the effect of probabilities of propagation using stochastic orders. The investigation compares various tree structures by employing the increasing convex order to highlight differences in risk profiles between them. The model is further examined through a numerical application, offering insights into practical implications for insurance, particularly in calculating premiums and risk measures. Our approach provides a comprehensive tool for assessing fire risk in complex structural layouts, with analytical advantages over simulation-based methods.

Présentations scientifiques

  • Actuarial fire-spreading model based on tree-structured probabilistic graphical model, with insurance applications, Congrès annuel de la Société statistique du Canada, Saskatoon, Canada (SK), 29 mai 2025.
  • Actuarial fire-spreading model based on tree-structured Markov Random Fields, with insurance applications, Waterloo Student Conference in Statistics, Actuarial Science and Finance, University of Waterloo, Canada (ON), 18 octobre 2024.
  • Assessing Fire-Induced Losses through Markovian Random Fields on Trees, Actuarial Research Conference (ARC), Murfreesboro, USA (TN), 17 juillet 2024.

Implications

Articles de vulgarisation

Présentations locales

  • Modélisation de la propagation des flammes dans une maison par l’utilisation des graphes aléatoires pondérés, Bureau de direction de la Chaire Industrielle de Recherche sur la Construction Ecoresponsable en Bois (CIRCERB), Université Laval, Québec, Canada (QC), 25 octobre 2023.
  • Modélisation de la propagation des flammes dans une maison par l’utilisation des graphes aléatoires pondérés, Séminaire de la Chaire Co-operators en analyse des risques actuariels, UQAM, Montréal, Canada (QC), 9 août 2023.
  • Contribution actuarielle à la compréhension des risques assurables liés aux constructions en hauteur de bâtiments non-résidentiels en CLT, École d’été de la Chaire Industrielle de Recherche sur la Construction Ecoresponsable en Bois (CIRCERB), Université Laval, Québec, Canada (QC), 14 juin 2023.

Autres

  • Bénévole résident-référent, service des résidences, Université Laval (2021, 2022, 2023)
  • Bénévole en cas d’incendie résidences, service des résidences, Université Laval (2022, 2023)
  • Bénévole programme de jumelage étudiants étrangers du Bureau de la vie étudiante, Université Laval (2023)