Shared Autonomous Vehicles: Travel Decision-Making Behavior under Comprehensive Risk

The risks associated with shared autonomous vehicles (SAVs) have not been eradicated with technological advancements; instead, they have significantly impacted travelers’ travel decision-making, which in turn affects the utilization rate of SAVs and hinders the development of SAVs. Thus, it is necessary to study the travel decision-making behavior (TDMB) of travelers under comprehensive risk. In this paper, first, the concept of comprehensive risk is introduced, along with a detailed categorization of comprehensive risk associated with SAVs. Then the TDMB model under comprehensive risk is built based on the structural equation model (SEM)-Logit model. Finally, elasticity analysis is applied to regulate the intervention factors that affect the travel decision-making probability of choosing SAVs. The results show that travelers’ decision-making is mainly influenced by the safety of SAVs. There is a negative correlation between the comprehensive risk and the travel decision-making probability of choosing SAVs. The greater the comprehensive risk, the lower the probability that travelers will choose to travel by SAVs. Meanwhile, the willingness of travelers to choose SAVs can be enhanced by optimizing and improving the performance of the autonomous driving system. For example, when the vehicle speed is greater than or equal to 25 km/h, it can effectively increase the probability of choosing SAVs. It is recommended that managers set up SAV lanes or AV lanes to ensure that the average speed is greater than or equal to 25 km/h through right-of-way priority and signal coordination, so as to expand the user base by using the speed threshold effect. The study can provide theoretical support for the technological development of SAVs and the formulation of related policies.