There have been several techniques for measuring bikeability; however, limited comprehensive research has been conducted focusing on travel distance as an important barrier for cyclists. Furthermore, existing measurements are mainly restricted by the availability of travel behaviour data. In this chapter, a new index for measuring bikeability in metropolitan areas is presented. The Cycling Accessibility Index (CAI) has been developed for computing cycling accessibility within Melbourne metropolitan, Australia. The CAI is defined consistent with gravity-based measures of accessibility. This index measures cycling accessibility levels considering mixed use developments as well as travel distance between origins and destinations. The Victorian Integrated Survey of Travel and Activity (VISTA) dataset was used to assess the proposed index and investigate the association between cycling accessibility levels and number of bicycle trips in local areas. Key findings indicate that there is a significant positive association between bicycle trips and the CAI.
Top2.1 Introduction
Promoting non-motorised accessibility has recently become an important objective for urban and transport planners (Iacono et al., 2010, Vale, 2013). Previous research on bicycle accessibility to destinations indicates that people commonly exclude potential destinations because of distance and travel time. Most studies consider travel distance between the origins and destinations as travel impedance. In these studies, accessibility reflects the attractiveness of facilities weighted by the travel time needed to reach those destinations (Sun et al., 2012, Hull et al., 2012, Silva and Pinho, 2010). However, travel distance has also been considered as travel impedance in some studies (Iacono et al., 2010, Lowry et al., 2012, Vasconcelos and Farias, 2012). Lowry et al. (2012) introduced a bikeability index which focused on bicycle trips. This study assessed the bikeability of the entire road network in terms of access to important destinations.
One practical reason of considering gravity-based or location-based accessibility measures for non-motorised modes of transport is their potential compatibility with regional travel forecasting models. Hence, they can easily extract travel times from one zone to another based on coded networks. In addition, a number of potential opportunities are available at the zone level (Iacono et al., 2010). However, one of the limitations of the use of these measures for non-motorised modes relates to the use of non-motorised modes in travel demand models. With respect to travel time, motorised modes are more sensitive to travel times and levels of network congestion than non-motorised modes of transport. Furthermore, non-motorised route choice tends to include factors that may be more qualitative, experiential or difficult to measure/quantify (Iacono et al., 2010, Tilahun et al., 2007, Hunt and Abraham, 2007).
Another limitation of existing approaches that measure cycling accessibility is that they are highly dependent on travel diary data. In addition, methods that have been applied to measuring cycling accessibility have not focused on the cycling availability of destinations in terms of service coverage areas. Some of the measures have focused on determining the level of service in terms of network infrastructure, such as the Bicycle Compatibility Index (BCI) or the Bicycle Level of Service (BLOS) for a bicycle network (Harkey et al., 1998a, Harkey et al., 1998b, Landis et al., 1997, Landis et al., 2003). These studies measure the performance of a bicycle network using various geometric measures, such as the width of the bicycle routes, pavement, route types, and connectivity. However, there are other methods that consider bikeability in terms of how accessible different destinations are for bicycles as a transport mode. Such methods measure the potential for cycling using travel behaviour data (Rybarczyk and Gallagher, 2014, Wahlgren and Schantz, 2012, Milakis et al., 2015, Espada and Luk, 2011).