01154368

Component

For bus carriers, abnormal driving behaviors not only cause higher risks of accidents and traffic violations, but also cause vehicles to wear out and bus services to break down. Thus, monitoring and managing abnormal driving behaviors is an important issue to bus operators. With the increasing applications of Intelligent Transportation System (ITS), for decades many bus carriers have applied digital tachographs, instead of traditional mechanical recorders, to record bus driving details. The sample data from a digital tachograph database, including data from 61 buses and 210 drivers, as well as data related to bus accidents, traffic violations, fuel consumption, and maintenance costs, were collected and analyzed in this study. Data mining techniques, i.e. the extraction of implicit and potentially useful information from enormous databases, such as multiple regression analysis, cluster analysis, and determinant analysis, were used to find the relationship among abnormal driving behaviors and driving safety, fuel consumption, and maintenance costs. The variables and related threshold values of abnormal driving behaviors were defined. Multiple regression analysis was used to establish the empirical relationship among abnormal driving behaviors and driving safety, fuel consumption, and maintenance costs; cluster analysis was used to categorize the sample of drivers who have similar driving characteristics; and discriminant analysis was used to determine into which cluster a driver should be categorized. The findings of this study indicate that emergency deceleration and speeding are the key determinants of the frequency of accidents; speeding and abnormal operation of electromagnetic braking are the key determinants of the frequency of traffic violations; and emergency deceleration and acceleration, long idle time of engine operation, abnormal operation of electromagnetic braking, and abnormal engine rotation are the key determinants for fuel consumption and maintenance costs. Based on data mining results, integrated driver management strategies, including rewards, re-education, and punishments were proposed. It was shown that after applying driver management strategies, the frequency of accident decreased by 13, the frequency of traffic violations decreased by 2, and a cost of NT$745,480 for extra fuel consumption and maintenance cost was saved on a monthly basis. The above savings could be used either for employees¡¯ re-education and training or to equip the entire bus fleet with new digital tachographs.

01147878

10-1940

English

23p

2010

Transportation Research Board 89th Annual Meeting

DVD

Figures (1) ; References (17) ; Tables (10)

Behavior; Bus crashes; Bus drivers; Bus transit operations; Data mining; Human factors in crashes; Intelligent transportation systems; Public transit; Recording instruments; Tachographs

Public Transportation; Safety and Human Factors; I83: Accidents and the Human Factor

Transportation Research Board Annual Meeting 2010 Paper #10-1940

TRIS, TRB

Jan 25 2010 10:53AM