Effect of Curve Length on Operating Speed on Horizontal Curve of Multilane Divided Rural Roadway

S. Al-Sayyaf, Sarmad; A. Abdulmawjoud, Ayman · 2026 · Crossref

DOI: 10.33899/arej.v31i2.63528

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study addresses the need for accurate speed prediction models on horizontal curves of multilane divided rural highways, aiming to evaluate the specific impact of curve length on the operating speeds of passenger cars and heavy vehicles. The research is motivated by the safety challenges posed by horizontal curves and the limitations of traditional design speed approaches, which often rely on restrictive geometric features rather than actual driver behavior. By focusing on the 85th percentile operating speed ($V_{85th}$), the authors seek to bridge the gap between engineering design criteria and the speeds drivers actually select based on their perception of roadway geometry. The methodology involved field surveys conducted between November 2023 and March 2024 across 37 sites in northern Iraq, covering governorates such as Nineveh, Erbil, Duhok, and Sulaymaniyah. Data collection utilized radar speed guns to record spot speeds at the midpoint of horizontal curves and surveying equipment to measure curve lengths. The study focused exclusively on free-flowing vehicles, defined by a headway of at least five seconds, excluding emergency vehicles, tractors, and motorbikes. Statistical analysis was performed using Minitab 21 software, employing simple linear regression to develop predictive models. Normality of the speed data was confirmed using the Shapiro-Wilk test, ensuring the validity of the regression assumptions. The results demonstrate a strong positive correlation between curve length and operating speed for both vehicle types. Pearson correlation coefficients were 0.916 for passenger cars and 0.854 for heavy vehicles. Two distinct predictive models were developed: one for passenger cars with an adjusted $R^2$ of 0.874, and one for heavy vehicles with an adjusted $R^2$ of 0.797. The regression analysis revealed that the coefficient for curve length is approximately twice as large for passenger cars (0.2075) compared to heavy vehicles (0.1058). This indicates that passenger car speeds are significantly more responsive to changes in curve length than heavy vehicle speeds, which are constrained by physical dynamics and more cautious driving behaviors. The significance of this study lies in its contribution to data-driven decision-making for highway geometric design. The findings confirm that operating speed models must be customized for specific vehicle types to accurately reflect driver behavior. The higher explanatory power of the passenger car model suggests that curve length is a dominant factor for light vehicles, whereas heavy vehicles may be influenced by additional variables not captured in this study. The authors recommend further research to identify other geometric and behavioral factors affecting heavy vehicle speeds and suggest developing comprehensive models for various roadway types to enhance safety and design consistency.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• road geometry