Modeling Operating Speed and Speed Differential on Two-Lane Rural Roads

TLDR

Highway geometry, especially on horizontal curves, influences collision risk, yet design consistency with driver expectations is lacking, leading drivers to select speeds based on their own perception rather than design speed. The study aims to evaluate operating speed consistency in Canada. The authors examined two sets of speed‑behavior models using traffic counter data from 20 curves on Ontario two‑lane rural highways. They found weak correlations for traditional operating speed but stronger links for the 85th‑percentile speed differential, and noted that nonintrusive data collection can uncover different speed patterns than radar measurements.

Abstract

The geometric features of a highway network play a significant role considering the fact that collisions occur disproportionately on horizontal curves. Based on extensive literature review, the problem mainly stems from the lack of geometric design consistency–conformance of highway geometric characteristics with drivers’ expectations. More specifically, drivers select their speeds according to their own perception of the road (referred to as the operating speed) rather than the designer’s perception (referred to as the design speed). To address operating speed consistency evaluation in Canada, two sets of models for speed behavior were examined based on speed data collected using traffic counters/classifiers on 20 curves on two-lane rural highways in Ontario. Relatively weak relationships were developed for the traditional operating speed on horizontal curves, while stronger relationships were found for the 85th percentile speed differential from a tangent to a curve. It was also shown that the nonintrusive approach for speed data collection might reveal different speed behavior than that observed using radar guns.