Discriminant analysis of transit operator distracted driving behaviors using naturalistic driving data

Bus transit driving is an occupation that requires high concentration in driving and is demanding due to work overload, time pressure, and responsibility for lives. This study is concerned with the effects of potential distractions on bus operator speed performances in their natural driving environment. Participating operators were recruited from Blacksburg Transit Agency; and a sample of 18 segments was extracted from 4-hour field observation data. Discriminant analysis was used to predict the classification of 6 potential distracted driving activities (exposures) with 17 predictors considered to construct the model. The final results produced a discriminant function of 4 distinguish exposures, of which 2 groups consist of 4 merged exposures that show less significant difference in operators' driving behaviors.