Search for: All records

This study employs a discrete event simulation (DES) model to understand the dynamic workload of remote truck operators managing partially-automated trucks. The DES model uses operator queues and event generators simulating automated truck events and leverages data from the California DMV’s disengagement database and driving simulation experiments. Disengagement data were partitioned into three groups by disengagement frequency: low, moderate, and high and separate arrival time distributions were developed for each group. Simulations from the model suggest that for companies with low disengagement rates, operator utilization will likely remain below minimal thresholds to prevent boredom. In contrast, companies with moderate or high disengagement rates both exceed operator utilization capacity and generate prolonged wait times as more trucks are controlled. These findings suggest that calibrating remote truck control to human capabilities will be challenging. A sensitivity analysis suggests that accurately estimating disengagement rates will be crucial for model accuracy and predictive performance. 

Driver reliance on automated vehicles (AV) is a critical component of safety particularly during high-risk traffic scenarios. Factors that influence reliance, including trust, situation awareness, fatigue, and demographics, have been independently explored; however, few analyses have investigated predicting AV reliance and compared factors comprehensively. The goals of this study were to develop a random forest (RF) model to predict reliance and to analyze the importance of factors for reliance decisions. We leveraged data from a driving simulation study where participants encountered four traffic events including responding to an illegal vehicle crossing, managing construction zones, stopping at a vandalized stop sign, and a pedestrian detection task. The dataset included reliance decisions and subjective assessments of dispositional trust, situational trust, fatigue, and workload. An RF model fit to the dataset using cross validation achieved an average AUC of 0.81 and accuracy of 0.77 and situational trust emerged as the most influential predictor.