Sousa Daiane Castro Bittencourt de, Pitombo Cira Souza, Rocha Samille Santos, Salgueiro Ana Rita, Delgado Juan Pedro Moreno
Universidade Federal da Bahia. Escola Politécnica. Departamento de Engenharia de Transportes e Geodésia. Salvador, BA, Brasil.
Universidade de São Paulo. Escola de Engenharia de São Carlos. Departamento de Engenharia de Transportes. São Carlos, SP, Brasil.
Rev Saude Publica. 2017 Dec 11;51:127. doi: 10.11606/S1518-8787.2017051007085.
To carry out a spatial analysis of the occurrence of acts of violence (specifically robberies) in public transportation, identifying the regions of greater incidence, using geostatistics, and possible causes with the aid of a multicriteria analysis in the Geographic Information System.
The unit of analysis is the traffic analysis zone of the survey named Origem-Destino, carried out in Salvador, state of Bahia, in 2013. The robberies recorded by the Department of Public Security of Bahia in 2013 were located and made compatible with the limits of the traffic analysis zones and, later, associated with the respective centroids. After determining the regions with the highest probability of robbery, we carried out a geographic analysis of the possible causes in the region with the highest robbery potential, considering the factors analyzed using a multicriteria analysis in a Geographic Information System environment.
The execution of the two steps of this study allowed us to identify areas corresponding to the greater probability of occurrence of robberies in public transportation. In addition, the three most vulnerable road sections (Estrada da Liberdade, Rua Pero Vaz, and Avenida General San Martin) were identified in these areas. In these sections, the factors that most contribute with the potential for robbery in buses are: F1 - proximity to places that facilitate escape, F3 - great movement of persons, and F2 - absence of policing, respectively.
Indicator Kriging (geostatistical estimation) can be used to construct a spatial probability surface, which can be a useful tool for the implementation of public policies. The multicriteria analysis in the Geographic Information System environment allowed us to understand the spatial factors related to the phenomenon under analysis.
运用地统计学方法，对公共交通中暴力行为（特别是抢劫）的发生情况进行空间分析，确定高发区域，并借助地理信息系统中的多标准分析找出可能原因。
分析单位是2013年在巴伊亚州萨尔瓦多市进行的名为“Origem-Destino”调查的交通分析区。确定2013年巴伊亚州公共安全部记录的抢劫案位置，并使其与交通分析区边界相匹配，随后与各自的质心关联。在确定抢劫概率最高的区域后，在抢劫可能性最高的区域对可能原因进行地理分析，考虑在地理信息系统环境中使用多标准分析所分析的因素。
本研究的两个步骤实施后，我们确定了公共交通中抢劫案发生概率较高的区域。此外，在这些区域确定了三个最易发生抢劫的路段（自由大道、佩罗·瓦斯街和圣马丁将军大道）。在这些路段，公交车上抢劫可能性最大的因素分别是：F1 - 靠近便于逃脱的地点、F3 - 人员流动量大、F2 - 缺乏治安。
指示克里金法（地统计估计）可用于构建空间概率面，这可能是实施公共政策的有用工具。地理信息系统环境中的多标准分析使我们能够了解与所分析现象相关的空间因素。
