Comparação das estimativas de atraso veicular médio por diferentes métodos

Abstract

The growth of the vehicle fleet in the country's large and medium-sized cities and the greater need for people and goods movements have been causing increasing congestion, thus requiring more knowledge and analysis regarding transport phenomena. Traffic engineering uses the most varied tools in the evaluation of traffic, which allow a greater precision for the analyst make the decision. Good, well-researched traffic analysis and analysis rely on data that can be very difficult to obtain and can be very costly to collect, thus being a barrier to traffic research. Delay is one of the main variables present in traffic engineering studies, being the parameter most often used to determine the service level of a traffic signal intersection, corresponding to the time lost by a vehicle caused by the interventions inherent to the traffic. By having some derivations and various estimation methods, the delay variable can be considered relatively complex. This work aims to compare the delay estimated by six methodologies, to evaluate the relationship between these data and to consider the possible applications of this information. In order to achieve this objective, six methodologies of collection were used, estimation and measurement of average vehicle delay were identified and characterized: SCOOT, HCM 2010 calculation model, HCM 2010 collection methodology, GPS collection and modeling in the VISSIM and TransModeler microsimulators. Subsequently, the variables were collected following the procedure indicated by each method. Finally, using comparison and statistical analysis, it was investigated how the variable average vehicle delay is measured or modeled by the different processes. The results showed that the six methodologies approximate considerably the values obtained of delay, besides considering what type of delay each of them obtains and the main differences between them. This research contributes to elucidate the methodology of estimation of the delay provided by the methods, besides indicating their possible uses and relations between them.