Índice de microacessibilidade em interseções semaforizadas : o caso do Hipercentro de Belo Horizonte/MG

Abstract

With the emergence and popularization of motor vehicles at the beginning of the 20th century, many cities began to suffer degradations and drops in the quality of life in their environments, due to the high circulation of cars. The neighborhoods where the almost exclusive circulation of automobiles is prioritized, become, in most cases, empty places, with little social or commercial activity. For this reason, it is essential that the circulation of pedestrians and cyclists is encouraged, so that cities can be more sustainable and democratic. Urban accessibility conditions are intrinsically related to local safety. Pedestrians are vulnerable road users. Therefore, in order to have more accessible cities, their safety must be taken into account. However, it is common for public authorities to neglect the development of safe areas for pedestrians in urban centers, as well as little attention to the topic in the literature. Intersections are dangerous places for pedestrians, as they are places where road users are more exposed to risky situations and there are difficulties for traffic organization. Thus, this dissertation proposes the development of an accessibility index with the point of view of pedestrian safety at intersections located in urban spaces, that can objectively assess the risks for pedestrians and return a score that exposes the level of safety, revealing the problems of such places. Indicators relevant to the topic were identified and selected to be part of the index and their ways of assessments were determined. The index, which was called IMS, was applied in the City Center of Belo Horizonte. The validation of the IMS was carried out, through statistical tests that could establish a relationship between the results of its application and accident statistics in the areas of each trafficked intersection. The validation obtained values with good relevance of the relationship between the two variables.