Analysis of safety benefits and security concerns from the use of autonomous vehicles: A grouped random parameters bivariate probit approach with heterogeneity in means

Abstract

This paper investigates public perceptions towards potential safety benefits, and safety- and security-related concerns from the future use of autonomous vehicles by utilizing data collected from an online survey. The survey includes responses from 584 individuals from the United States, who responded to a varying range of questions related to autonomous vehicles and their usage. The subsequent exploratory statistical analysis is conducted by employing a novel method, namely the grouped random parameters bivariate probit model with heterogeneity in means. The proposed method accounts for the challenges stemming from the presence of multiple layers of unobserved heterogeneity in the data, and simultaneously offers more insightful results. From the analysis, several socio-demographic characteristics, and driving attitude related characteristics and opinions were found to affect the perceptions towards the safety and security related aspects of autonomous vehicles. The heterogeneity in means approach revealed distinct individual-specific characteristics that affect the peak of the distribution of the parameter density function of the random parameters, adding further clarity to the understanding of the factors affecting individuals’ perceptions towards autonomous vehicles. The findings from this study suggest the ongoing evaluation of public perceptions, and reinforce the requirement of analyzing temporal variations in public perceptions. This can, in turn, aid regulatory and governance entities and autonomous vehicle manufacturers to adapt their strategies and implementation plans accordingly.