Effects of Personal Driving Styles on Preference for Fully Automated Driving and Their Heterogeneity by Travel Distance: A Hybrid Choice Model Approach with Multiple-Stated-Choice Experiments

Abstract

As individuals have different thoughts about automated driving (AD) vehicles, the gap between their manual driving (MD) styles and their expected AD styles would affect their willingness to ride AD vehicles. The objective of this study is to investigate the effects of personal driving styles on AD preference and the variations of these effects by travel distance. To identify latent driving styles and their effects simultaneously, we developed a hybrid choice model using sequential stated-choice data from an online survey administered to 511 respondents in South Korea in 2019. The latent driving styles were classified as impatient, over-nervous, and cautious driving styles. The results significantly show that individuals of each style prefer AD vehicles in different situations. Impatient drivers, who are concerned about travel delays, rarely worry about the driving mode if it drives them faster. Over-nervous drivers, who lack confidence in their driving skills, tend to trust AD technology, regardless of the travel distance. Cautious drivers, who are mindful of their surroundings, prefer AD when traveling relatively short distances. The elasticities of the travel time and cost show that the driving styles have impacts on the AD preference. Moreover, the results of value of time (VoT) for AD have been estimated to differ by travel distance. The results of this study have identified the effects of personal driving styles on AD preference, providing valuable insights that can be used to improve AD vehicle related policies and marketing strategies (such as implementing price discrimination by travel distance and lowering the overall price of AD vehicles), which may increase the market share of AD vehicles.