Understanding the cycle traffic impacts of Cycle Superhighways in London

Abstract

Cycling as a clean, green, and environmentally friendly mode of transportation plays a crucial role in society by fostering physical activity and a healthy lifestyle, reducing traffic congestion, and improving mobility. To create more efficient strategies for promoting cycling, there is a need to gain a better understanding of the influential factors on cyclists’ route choice behaviour. Electric bikes (e-bikes) are an emerging technology that appeared to assist cycling by using battery-powered motors. Researchers consider e-bikes as an emerging technology with its most certain effect being easing up cycling. Hence, investigating individual route choice behaviour with respect to their bike type can unveil new insights for cycling promotion. To this end, we used data collected via a stated preference (SP) survey in Finland not only to investigate the factors affecting cyclists’ route choice behaviour but also to compare the behaviour of e-bikers with regular bike users (rbikers) in order to identify the changes that may happen by easing the pedalling fatigue due to the pedal-assist feature of e-bikes. Our results indicate that low interaction with traffic, fewer intersections, and separated bike facilities are the main factors unchanged to promote cycling among r-bikers and e-bikers. Furthermore, we compare the outputs of simple Logit models (SLMs) and random parameter Logit models (RPLMs) for r-bikers’ and e-bikers’ route choices to address the impact of error correlation among observations in SP data. Our findings imply that the SP data is well-designed to capture the preferences of the individuals accurately, so the observations are not severely correlated, i.e., the IID assumption is held. This suggests that using SLMs can lead to similar outputs with RPLMs, without increasing the complexity of the estimation process.