Modelagem microscópica aplicada na avaliação do desempenho da segurança viária em interseções urbanas

• Ano de defesa: 2012
• Autor(a) principal: Sousa, Janailson Queiroz
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Transportes; Acidentes de trânsito; Segurança de trânsito
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/4885

The microscopic modeling of traffic flow in urban road infrastructure has proven to be a potential tool for analyzing the performance of road safety (DSV) compared to traditional methods of analysis based on observational studies. To consolidate this approach there are practical and methodological challenges that have been treated as objectives in this research, among which are: (i) the identification of efficient synthetic indicators for the analysis of safety in this environment, (ii) the analysis of the microscopic algorithms and their effectiveness in the representation of the driving processes in urban areas, (iii) the definition of a systematic procedure for estimating the DSV using a microscopic platform and (iv) concerning the validation of synthetic indicators used to measure the DSV. Given this context, this research intended to provide a general methodological contribution to the use of microscopic traffic simulators for the analysis of the DSV in tactical and operational transportation systems planning process with focus on urban intersections. The proposed methodological procedure considered the completion of five steps: 1) selection and site characterization study, 2) coding the site in microsimulator3) planning of the simulation, 4) calibration and validation of the model and 5) estimation of the indicators. The procedure considered three signalized intersections in the city of Fortaleza, using VISSIM © as simulation platform. The results indicated that it was possible to estimate the DSV with relative confidence level and a complementary procedure for validation of the indicators showed a consistent level of consistency between the estimated number of conflicts of longitudinal interactions and observed number of rear end collisions observed in the field.