Network Connectivity for Low-Stress Bicycling

When streets with high traffic stress—on which the mainstream population is unwilling to ride a bike—are removed, the remaining network of streets and paths can be fragmented and poorly connected. This paper describes the development of methods to visualize and to analyze the lack of connectivity in a low-stress bicycling network. A proposed measure to evaluate bicycling networks is the fraction of origin–destination pairs, which are connected without the use of high stress, without excessive detour, and with the origin–destination pairs weighted by travel demand. A new method is proposed to classify segments and crossings into four levels of traffic stress (LTS) on the basis of Roger Geller’s classification of the cyclist population and Dutch design standards, which are known to attract the mainstream population. As a case study, every street in San Jose, California, was classified by LTS value. Maps that showed only lower stress links revealed a city divided into islands within which low-stress bicycling was possible, but these islands were separated from one another by barriers that could be crossed only with the use of high-stress links. The fraction was 4.7% of home-to-work trips up to 6 mi long that were connected at a low LTS value. The figure would almost triple if a modest slate of improvements were implemented to connect low-stress streets and paths with each other.