Impacts of real-time feedback on driving behaviour: a case-study of bus passenger driver

Content Provider	Semantic Scholar
Author	Rolim, Catarina Baptista, Patrícia Campos Pavan Duarte, Gonçalo O. Farias, Tiago L.
Copyright Year	2015
Abstract	The transportation sector is one of the main contributors to enhance quality of life, by providing accessibility to people, places and goods. On the other hand, it also contributes to the degradation of the environment, presenting high levels of energy consumption and pollutants emissions. Over the years, rising attention has been given to behavior, both for light-duty and heavy-duty drivers, in particular to the effects of education, training, and feedback on driving performance on vehicle purchase, and mobility patterns. When developing programs for the adoption of more efficient driving techniques a distinction between professional and private drivers must be taken into consideration. Bus drivers are among the groups of professionals that are most stressed and negatively affected by their work demands and environment. The aim of this study is to assess the impacts over time of real-time feedback and training on efficient driving techniques towards driving behavior changes, considering drivers characteristics (age and time working at the company) and vehicle type and age. Over 200 bus drivers have been monitored for 4 years in a large bus fleet company in the Lisbon region. Rodoviaria de Lisboa S.A., a Portuguese bus passenger-transport company, uses since 2008 a data logger (VDO-FM2000 Plus) that is currently installed in 100 buses and drivers use these vehicles under regular daily operation, while driving in the Lisbon metropolitan area. The device is used to identify undesirable behaviors and to indicate them in real-time to the driver using a sound alert when an undesired event occurs. A managing software platform developed by the company, named GISFROT, collects and manages driving data including excess speeding, extreme brakes and accelerations, excess RPM and speed, among other indicators. Besides being monitored with the on-board device, drivers also participate in periodic training sessions on driving efficient techniques designed specifically for them taking into consideration their driving performance results. The impacts of real-time feedback (sound signals) on driving behavior were assessed considering three different monitoring periods. In the first monitoring period (Phase 1), from October 2010 to September 2011, the drivers received real-time sound feedback while driving the instrumented buses. In the second phase (Phase 2) drivers did not receive any sound feedback while driving, from October 2011 to September 2012. After this period with no sound feedback, drivers started again receiving real-time information regarding their driving performance (Phase 3 – from October 2012 to July 2013). A comparison between periods was performed regarding the percentage of time drivers spent in undesirable driving events such as hard brakes and hard accelerations, among others. This analysis also considered drivers characteristics such as age (Young adult, Adult, Senior) and working experience (Short-term, Medium-term and Long-term employee), vehicle type (mini, standard and articulated buses) and vehicle age (groups defined according to the Euro Standards). The driver characterization indicates that 37% of the bus drivers are in the group Adults, aged between 36 and 50 years old, and are long-term employees of the company. Also, 25% of the drivers belong to the group Seniors (over 50 years). Drivers participated in training sessions mainly in 2011, and no sessions were organized in 2012. Concerning drivers’ access to an online platform to check on their driving performance(data was only available for the year 2012), Young short-term employees check, on average, their performance more times than other groups. Resulted indicated that the influence of providing real-time feedback to the driver has proven to be effective. When drivers are no longer given feedback, their performance on driving indicators reveals a tendency to worsen, particularly in what concerns extreme brakes and accelerations, hard stops and excessive RPM, with an average increase of 24%, 42%, 27% and 41%, respectively. This is observed mainly when driving mini and articulated buses. On average, the groups of drivers that present higher increases are the short-term employee drivers and Senior drivers. From Phase 2 to Phase 3, when drivers are again given feedback, drivers decrease once again the incidence in the driving indicators analysed, with average reductions from 2 to 44% in the indicators mentioned previously. An analysis, with the application of statistical hypothesis tests, regression and discrete choice models, will be performed to better understand the combined impact of various variables on driving behavior and their interactions with the sound feedback. This analysis will help to quantify the real impact of the feedback on driving behavior, predict the influence of variables such as age and vehicle type on performance and, therefore, to identify the groups of drivers that need a more incisive training. Keywords: driving behavior, real-time feedback, large-scale monitoring, bus passenger driver
File Format	PDF HTM / HTML
Alternate Webpage(s)	https://d1rkab7tlqy5f1.cloudfront.net/TBM/Over%20faculteit/Afdelingen/Engineering%20Systems%20and%20Services/EJTIR/Back%20issues/17.3/2017_03_03.pdf
Language	English
Access Restriction	Open
Content Type	Text
Resource Type	Article