Development of an Inference System for Drivers’ Driving Style and Workload Sensitivity from their Demographic Characteristics

Abstract

This paper describes an investigation of the relationships among drivers’ demographic characteristics, their driving style, and their workload sensitivity using questionnaire surveys. The driver’s demographic characteristics included age, gender, driving experience, annual mileage, driving frequency, and region. The driving style was assessed using the driving style questionnaire (DSQ) that presents the driver’s personal features about driving attitudes We evaluated the drivers’ workload sensitivity using the workload sensitivity questionnaire (WSQ) that suggests what kinds of elements in driving contexts force the driver to increase the mental workload. 1616 drivers around Japan participated in the questionnaire surveys. Bayesian network modeling was applied to the responses obtained from the questionnaire surveys. The estimated Bayesian network models present that “gender” influences more factors of the DSQ and WSQ compared to the other demographic characteristics. The models indicate that no influences of the “driving frequency” were found in the DSQ and WSQ and no influences of the “driving experience” were found in the WSQ. We use the estimated Bayesian models to infer the distributions of the DSQ and WSQ factors from the driver’s demographic characteristics, and evaluate one driver’s score of the DSQ and WSQ compared to the other drivers with the same demographic characteristics. In addition, this inference system could be used to select the target drivers who have the focusing driving style or workload sensitivity.

Keywords: Driving Style. Workload Sensitivity, Questionnaire Survey, Bayesian Network Model

DOI: 10.54941/ahfe100737