Author(s)

K.A. D’SOUZA & S.K. MAHESHWARI

Abstract

The increase in bus transit ridership along with the proliferation of personal electronic control and communication gadgets is causing more distractions for the drivers. For transit vehicles, some distractions are caused by factors beyond the driver’s control such as operating additional equipment, attending to passengers, and communicating with the operations center. Several driver distraction studies have been conducted for personal vehicles and commercial vehicles. But bus transit driver distraction has received limited attention in the literature even though bus transit accidents may cause more injuries due to larger number of passengers. Hence, their distraction is not clearly understood; furthermore, no established methodology is available to conduct a detailed study at a transit agency because of inadequate research in the field.

The objective of this paper is to present a detailed modular research framework for studying bus transit driver distractions. The framework provides a transit agency with a set of standardized methodologies for studying distraction over a wide range of cost and time intervals. An agency may choose one or more modules to suit their study requirements. The modules for data collection, analysis, validation, and interpretation and usage of results are designed on the basis of in-depth studies and tests at transit agencies in the Commonwealth of Virginia. The paper provides a detailed process and a set of guidelines to study bus transit driver distraction which will make it easier for any transit agency to conduct such a study. The results of the bus transit driver distraction studies could be used for training bus drivers to mitigate distraction and assist state and city governments to formulate effective regulations to control distracted driving.

Keywords

Bus transit driver distraction, Distraction Risk Index, modular design for studying bus driver distraction, model validation, modeling and predicting driver distraction, Monte Carlo simulation, multinomial logistic regression, route observations