Modelo de suporte à tomada de decisão sobre de acidentes de trânsito com vítimas baseado em lógica fuzzy.

• Ano de defesa: 2013
• Autor(a) principal: Pereira, Ana Paula de Jesus Tomé
• Orientador(a): Moraes, Ronei Marcos de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraí­ba
• Programa de Pós-Graduação: Programa de Pós-Graduação em Modelos de Decisão e Saúde
• Departamento: Ciências Exatas e da Saúde
• País: BR
• Palavras-chave em Português: Saúde pública; Acidentes de trânsito; Lógica fuzzy
• Palavras-chave em Inglês: Traffic accident; Spatiotemporal cluster; Fuzzy logic
• Área do conhecimento CNPq: CIENCIAS DA SAUDE::SAUDE COLETIVA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/tede/6544

Traffic accidents represent, in Brazil, a serious economic and especially social, relevant for magnitude of the mortality and number of people suffering from sequelae arising, thus becoming a serious public health problem. This research aimed to develop a model to support decision making based on fuzzy logic, supported by analyzes spatial and spatio-temporal (Scan method) to categorize neighborhoods according to priority intervention for prevention and control of traffic accidents that produce victims. Secondary data were georeferenced and recorded by Mobile Emergency Care Service in João Pessoa, Paraíba, in the years 2010 and 2011. Throughout study period, João Pessoa was 10,070 traffic accidents with victims. Of this total, 17.8% had breath ethanol and 0.8% died at the scene. The majority of victims were male (74.5%), belonging to the age group 20-29 years (37.7%). The accidents occurred mainly on Sundays (19.2%), Saturdays (18.7%) and on Fridays (14.4%) as well as in the months of December (10%), October (9.8% ) and May (8.9%). Most of the vehicles involved was composed by motorcycles (68.1%) and cars (36.5%). The nature of accident, collision was more frequent (46.2%), followed by fall motorcycle (30.7%) and pedestrian injuries (11.1%). In analysis of the relative risk and spatial distribution of these events, it was found that neighborhoods with high relative risk and formed significant spatial clusters concentrated in the north, northwest and northeast of the municipality. We identified 15 clusters space-time, which concentrated mainly in the northern, northeastern and coastal strip of the municipality. It was observed that neighborhoods reported by Mobile Emergency Care Service were categorized as priority by model, Valentina and Mandacaru were categorized as with tendency to priority, and Mangabeira was categorized as non-priority. The proposed decision model showed good agreement when compared with Mobile Emergency Care Service, thus satisfying the identification and classification of neighborhoods as a priority, with tendency to priority, with tendency to non-priority and non-priority. The results may be of relevance to both Mobile Emergency Care Service as to other public officials linked to road traffic, traffic education and care for victims produced by road traffic in João Pessoa.