How do truck platoons impact the behaviors of adjacent passenger cars? A random parameter logit with heterogeneity in means and variances approach

Abstract

Automated truck platoons are believed to be practice-ready given their great potential for fuel and emission reduction. However, safety concerns must be addressed before a large-scale deployment of automated truck platoons can occur, especially regarding their interaction with surrounding passenger cars. Despite its importance, research on how truck platoons affect adjacent drivers remains limited. Addressing this gap and given the lack of real-world automated truck platoon data, this study leverages high-resolution trajectory data on self-organized truck platoons and their adjacent passenger cars as a preliminary exploration of potential interactions involving automated truck platoons. Based on lateral movement, the safety states of adjacent passenger cars were clustered into four categories using the Entropy Weight Method (EWM) and the Gaussian Mixture Model (GMM). A random parameter logit model with heterogeneity in mean and variance was then developed to quantify the impact of contributory factors on the safety state of adjacent passenger cars. The results show that factors such as the lateral deviation of the truck platoon, the length of the truck platoon, and the complexity of the traffic environment increase the risk to adjacent vehicles. For instance, a one-unit increase in maximum lateral offset raises the probability of Offset Unstable State (OUS) by 3.99%, and adding another truck or increasing the average headway elevates the OUS probability by 0.34% and 3.35%, respectively. These findings highlight the influence of truck platoon configuration on adjacent-car safety and contribute to the development of safer, more robust strategies for integrating truck platoons into real-world traffic.