An empirical reappraisal of the level of traffic stress framework for segments

Abstract

The Level of Traffic Stress (LTS) framework is widely used to assess the suitability of roadway environments for cycling. Its main strength is the identification of infrastructure characteristics that meet the needs of different cyclist groups. The four levels of infrastructure ratings, LTS 1 to LTS 4, roughly map to the cyclist types defined by the Four Types of Cyclist typology. Despite its popularity, the LTS framework has several limitations, including reliance on a cyclist typology that was developed subjectively, and a lack of empirical evidence to define thresholds between levels. This work builds on our previous empirically-based findings that cyclists form three groups rather than four: Uncomfortable or Uninterested, Cautious Majority, and Very Comfortable Cyclists. We use survey data from Edmonton, Canada, to update the LTS framework such that levels match the three types of cyclists. Direct infrastructure ratings, binary logistic regression, and route choice data provide the empirical foundation to determine infrastructure characteristics that are suitable for the three types of cyclists. This adjusted framework is called Level of Cycling Comfort (LCC). We apply the framework to Edmonton and compare connectivity outcomes using both LTS and LCC frameworks. Overall, the LCC framework yields more conservative estimates of connectivity.