Combinação de técnicas da inteligência artificial para previsão do comportamento do tráfego veicular urbano na cidade de São Paulo

• Ano de defesa: 2011
• Autor(a) principal: Ferreira, Ricardo Pinto
• Orientador(a): Sassi, Renato José
• Banca de defesa: Marte, Claudio Luiz , Santana, José Carlos Curvelo
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Nove de Julho
• Programa de Pós-Graduação: Programa de Pós-Graduação de Mestrado e Doutorado em Engenharia de Produção
• Departamento: Engenharia
• País: BR
• Palavras-chave em Português: rede neuro fuzzy; roteirização de veículos; roteirização dinâmica de veículos; inteligência artificial; tráfego veicular urbano
• Palavras-chave em Inglês: neuro fuzzy network; vehicle routing; dynamic vehicle routing; artificial intelligence; urban vehicular traffic
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA DE PRODUCAO
• Link de acesso: http://bibliotecatede.uninove.br/tede/handle/tede/155

In recent years, the increase in consumption of Brazilian families, result of economic the stability experienced in the country, has resulted in a high volume of items that should be collected or distributed daily in São Paulo city. This scenario has caused profound changes in the market of distribution and collected orders, making the distribution highly complex and directly affecting the efficiency of this service. The conditions of traffic flow and safety, of the city of São Paulo, depends directly on some issues such as: broken trucks, manifestation on roads, lack of energy, tree falling, accidents with or without victims and other occurrences. There are three levels of routing to be examined: the Operational Level, at this level we consider the methods for vehicle routing; Tactical Level, this level is the Dynamic Vehicle Routing, that offers real-time alternatives to reduce unproductive time in stretches, with slow or interrupted, due to some accentuated remarkable occurrence and the Strategic Level, this level is the prediction of the behavior of urban vehicular traffic at the beginning of the script. Many techniques and software are used to predict the behavior of vehicular traffic in the São Paulo city, including techniques based on Artificial Intelligence. Thus, this work was applied to predict the behavior of traffic, two Artificial Intelligence techniques combined: Fuzzy Logic or Diffuse Logic and Artificial Neural Networks, which together form a network called Neuro Fuzzy. This paper aims to predict the behavior of city vehicular traffic in the city of São Paulo using a Neuro Fuzzy Network. The results indicate a positive impact to application of Neuro Fuzzy Network for predicting the behavior of urban vehicular traffic in São Paulo. It can be stated that the dynamic routing of vehicles combined with the prediction of traffic behavior possible to increase the efficacy of the routing in a city like São Paulo.