Globally, the use of electric vehicles, and in particular the use of electric buses, has been increasing. The city of Nanjing leads China in the adoption of electric buses, supported by city policies and infrastructure. To lower costs and provide a better service, vehicle selection is crucial, however, existing selection methods are limited. Accordingly, Nanjing Bus Company developed a test method based on road tests to select a bus. This paper presents a detailed description of the test method and a case study of its application. The method included an organization structure, selection of eight test vehicles (four 10 m length, four 8 m length) from four brands (a total of 32 test vehicles), selection of indicators and selection of routes. Data was collected from repeated drives by 65 drivers over an 8-week period. Indicators included power consumption, charging duration, failure duration and driving distance. It is concluded that the road test method designed and conducted by the Nanjing Bus Company provides a good framework for the selection of pure electric buses. Furthermore, subsequent experience with selected buses supports the validity and value of the model.

Road traffic accidents regarding commercial vehicles have been demonstrated as an important culprit restricting the steady development of the social economy, which are closely related to the distracted behavior of drivers. However, the existing driver’s distracted behavior surveillance systems for monitoring and preventing the distracted behavior of drivers still have some shortcomings such as fewer recognition objects and scenarios. This study aims to provide a more comprehensive methodological framework to demonstrate the significance of enlarging the recognition objects, scenarios and types of the existing driver’s distracted behavior recognition systems. The driver’s posture characteristics were primarily analyzed to provide the basis of the subsequent modeling. Five CNN sub-models were established for different posture categories and to improve the efficiency of recognition, accompanied by a holistic multi-cascaded CNN framework. To suggest the best model, image data sets of commercial vehicle driver postures including 117,410 daytime images and 60,480 night images were trained and tested. The findings demonstrate that compared to the non-cascaded models, both daytime and night cascaded models show better performance. Besides, the night models exhibit worse accuracy and better speed relative to their daytime model counterparts for both non-cascaded and cascaded models. This study could be used to develop countermeasures to improve driver safety and provide helpful information for the design of the driver’s real-time monitoring and warning system as well as the automatic driving system. Future research could be implemented to combine the vehicle state parameters with the driver’s microscopic behavior to establish a more comprehensive proactive surveillance system.