A model for predicting Acceleration Severity Index in impacts with road safety barriers

Burbridge, Andrew; Troutbeck, Rod · 2019 · Crossref

DOI: 10.1080/13588265.2018.1474621

archive: archived pipeline: cataloged verified

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

Summary

This paper addresses the need for a predictive model for the Acceleration Severity Index (ASI), a non-dimensional metric used to assess vehicle occupant severity during impacts with road safety barriers. ASI is a required reporting metric under European (EN1317) and Australian/New Zealand standards, serving as a proxy for occupant injury risk. Previous literature suggests that ASI varies based on impact configuration (vehicle mass, speed, and angle) and barrier flexibility, but a comprehensive model linking these variables was lacking. The study aims to develop a theoretical model to predict ASI for any impact configuration given the barrier's flexibility, thereby aiding in the evaluation of barrier performance and occupant risk. The methodology involves analyzing data from 79 full-scale crash tests of longitudinal barriers on flat ground, excluding transitions, slopes, and temporary barriers to remove confounders. The authors build upon prior work establishing a linear relationship between barrier flexibility (dynamic deflection) and the reciprocal of ASI. Using least sum of squared differences regression, they determine the best-fit constants for a model where the slope term is a function of vehicle mass, speed, and angle. Constraints were applied to ensure physical rationality, specifically that the mass exponent is non-negative while speed and angle exponents are non-positive. The analysis utilized Microsoft Excel’s solver function to minimize residuals. The results indicate that a simplified model provides a statistically significant fit ($p<0.001$). The final model expresses ASI as a function of dynamic deflection, impact severity (derived from mass, speed, and angle), and vehicle mass. Specifically, the slope term was found to be proportional to vehicle mass ($a = 50 \times m$). The model predicts that ASI decreases with increasing vehicle mass and increasing barrier flexibility. Conversely, ASI increases with higher impact speeds and angles, particularly when barriers are rigid. The model accounted for 93.7% of results with an error magnitude of $\leq 0.3$, though outliers were observed in high-severity impacts involving lighter vehicles and stiff barriers. The authors note that variability in observed ASI increases with higher predicted values, potentially due to factors like vehicle-barrier friction and barrier face shape not explicitly included in the model. The significance of this work lies in demonstrating that ASI is predictable based on impact configuration and barrier characteristics. This challenges the current use of ASI as a rigid pass-fail threshold in standards like EN1317, suggesting it may instead represent a continuum of occupant severity. The model highlights that current test protocols may yield marginal pass/fail outcomes for rigid barriers under real-world conditions, such as higher speeds or angles. The authors conclude that while the model is limited by the standardized nature of the input data, it provides a valuable tool for predicting barrier performance and underscores the need for further research to definitively link ASI to specific injury outcomes.

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.

• incidence prevalence