Gestão de Risco sob a Ótica Fuzzy em Modelo Multivariado

• Autor(a) principal: Sá, Felipe de Oliveira
• Data de Publicação: 2021
• Outros Autores: [e-mail]
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/20116
• topic: Fuzzy logic; Machine learning; Auto insurance; Risk; Lógica difusa; Inteligência computacional; Seguros de automóveis; Lógica Fuzzy; Aprendizado de máquina; Seguro de automóveis; Risco; CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::SISTEMAS DE COMPUTACAO

The insurance industry is a very important sector for the national economy and plays an important role in the development and generation of income in Brazil. The domestic insurance market keeps expanding and currently accounts for a significant portion of the national Gross domestic product (GDP). With high growth expectations it has the potential to account for an even larger part of the GDP just like in countries with stronger, more developed economies. Due to theft and property damage the automobile sector is the most important part of the domestic insurance market which forces the insurance companies to continuously develop their pricing strategies. Based in a highly competitive and agile automobile market, this paper proposes a Fuzzy Multivariate Logistic Model for risk management that offers categorization into five risk groups: very low risk, low risk, medium risk, high risk and very high risk. This enables an efficient and manageable pricing strategy that considers the uncertainties and subjectivities of the real world. The model is flexible and efficient thus able to handle policy changes as well as process large amount of data while taking into account the limitations, subjectivities and uncertainties associated with the automobile sector. It provides quick, real-time risk classification and thus differentiates from the static classifications offered by the standard models within the insurance sector.