Unsignalized Intersection Level of Service: A Bicyclist’s Perspective

Abstract

As the traffic conditions at unsignalized intersections are complex and chaotic, the perceived satisfaction levels of bicyclists are controlled by several factors. Although complex, the Bicycle Level of Service (BLOS) studies at this facility are highly important so as to improve its performance from the perspective of bicycling. This study is an initial attempt to identify and model the factors influencing BLOS at unsignalized intersections under mixed traffic conditions. For analysis purposes, extensive data sets on the geometric details and operational characteristics are collected from 70 unsignalized intersection approaches located in various parts of five Indian cities. Subsequently, eight variables (effective approach width, peak hour volume, conflicting traffic volume, and average bicycle delay, etc.) having significant influences on the perceived satisfaction levels of through bicyclists are identified. Of all these variables, the average bicycle delay is observed to have the highest influence on user-perceived BLOS. This suggests that the minimization of bicycle delay is the utmost important strategy to enhance perceived satisfaction levels of bicyclists. Further, a step-wise regression-based BLOS model has been developed for the service quality assessment at unsignalized intersection approaches. This model has shown a very good prediction ability in the present context with a coefficient of determination (R2) value of 0.83 with averaged observations. A service scale is also defined to convert the outcomes of this model to letter-graded service classes A–F (excellent–worst). The field application of these tools has shown that above 89% of the investigated sites are offering BLOS ‘C’ or inferior. This is a serious concern for the inhabitants in the long run. Hence, the modeled attributes should be prioritized (in their identified order of importance) in the transportation planning process to avail the better service classes effectively.