Safety Evaluation of Bus Running State Based on Multi-Source Data

The studying of bus running state, especially risk evaluation, is vital to understanding the risk, and optimizing bus early warning systems to improve bus operation safety. In this study, the bus running state in Zhenjiang city, China, was integrated and mined by multi-source data such as radar and video warning data, vehicle driving characteristics, bus road network, and meteorological information, and then a risk evaluation model of bus running state was established. Using ArcGIS 10.2 software and SPSS 22 software, the risk distribution characteristics of bus running state were analyzed from the aspects of weather, time, and space, respectively. Finally, quantitative indicators for the risk evaluation of bus running state were determined, and the risk evaluation models of bus running state were established. The results showed that the models had both high accuracy and good applicability. This study will provide theoretical and algorithmic references for the optimization of bus early warning system and bus operation management.