RBG Lab, Department of Engineering Design, IIT Madras, Chennai, India.
Int J Inj Contr Saf Promot. 2020 Jun;27(2):144-155. doi: 10.1080/17457300.2019.1681005. Epub 2019 Nov 11.
Pedestrian safety is of growing concern with an increasing number of traffic accidents, especially in developing economies like India. In 2017, there were 20,457 pedestrian fatalities in India. Pedestrian crashes have also become a key concern in the state of Tamilnadu, India, due to the high percentage of deaths. If the available datasets are large and complex, identifying key factors is a challenging task. In this study, Multiple Correspondence Analysis (MCA), an exploratory data analysis technique was used to explore the roadway, traffic, crash, and pedestrian-related variables influencing pedestrian crashes. This study used the data from Government of Tamilnadu Road Accident Traffic Management System (RADMS) database, to analyse accident data of nine years (2009-2017) related to pedestrian crashes. The results of the study show that crashes occurring on the express highways on a multilane road are often associated with hit-and-run behaviour among drivers. Factors such as lighting conditions, location, pedestrian behaviour, crossings, and physical separation are also significantly contributing to pedestrian crashes. The key advantage of MCA is that it identifies a possible association between various contributing factors. The findings from this study will be useful for state transport authorities to improve countermeasures for mitigating pedestrian crashes and fatalities.
行人安全越来越受到关注，尤其是在印度等发展中经济体，与日俱增的交通事故数量更是雪上加霜。2017 年，印度有 20457 名行人死亡。在印度的泰米尔纳德邦，由于高死亡率，行人事故也成为一个主要关注点。如果可用的数据集庞大且复杂，那么确定关键因素就是一项艰巨的任务。在这项研究中，我们使用了一种探索性数据分析技术——多元对应分析（MCA），来探讨影响行人事故的道路、交通、事故和行人相关变量。本研究使用了来自印度泰米尔纳德邦政府道路事故交通管理系统（RADMS）数据库的数据，分析了 9 年来（2009-2017 年）与行人事故相关的数据。研究结果表明，在多车道的高速公路上发生的事故往往与司机的肇事逃逸行为有关。光照条件、位置、行人行为、交叉口和物理隔离等因素也对行人事故有重大影响。MCA 的主要优势在于它可以识别出各种因素之间可能存在的关联。这项研究的结果将有助于邦交通主管部门采取措施，减少行人事故和死亡人数。
