The Practical Effectiveness of Advanced Driver Assistance Systems at Different Roadway Facilities: System Limitation, Adoption, and Usage

Abstract

This research proposes a method to estimate the practical effectiveness of 6 advanced driver assistance systems (ADAS) in avoiding crashes on different roadway facilities (locations). The systems include blind spot warning/lane change warning (BSW/LCW), forward collision warning (FCW), intersection movement assist (IMA), pedestrian crash avoidance and mitigation (PCAM), lane departure warning (LDW) and left turn assist (LTA) systems. The method involves the use of meta-analysis and quasi-induced exposure methodologies to estimate the distribution of effectiveness factors across the driver population. These factors include system operating conditions, system adoption rates and system usage features. Then a Monte-Carlo procedure is applied to sample these factors for simulating the practical ADAS effectiveness. The research shows that the practical effectiveness could be as much as 48.25% lower than the ideal effectiveness proposed by previous research. It is also found that BSW/FCW, PCAM, FCW and LDW systems exhibited the highest effectiveness when operating on urban expressways/freeways. On rural multilane highways, such systems are demonstrated to have the least effectiveness. Furthermore, LTA systems are found to perform better at rural intersections than at urban intersections while the conclusion is contrary regarding IMA systems. Among the six systems, BSW/LCW has the most consistent performance among facilities, while LTA is most likely to be affected by facility type. The integrated system of six ADAS systems could prevent at most 29.50% of total crashes on urban multi-lane highways, while its lowest practical effectiveness is 13.72% on rural two-lane highways. Finally, technical and policy implications of the results are discussed.