Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Industrial and Systems Engineering | en_US |
| dc.contributor.advisor | Xu, Min (ISE) | en_US |
| dc.contributor.advisor | Chung, Sai Ho Nick (ISE) | en_US |
| dc.creator | Ren, Qiaoqiao | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/14533 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | en_US |
| dc.rights | All rights reserved | en_US |
| dc.title | Post-crash injury severity analysis for conventional and autonomous vehicles | en_US |
| dcterms.abstract | Road traffic crashes remain a significant concern, consistently ranking as one of the leading causes of death and disability worldwide. Analysing and reducing the crash injury and damage severities is essential for ensuring public safety and promoting sustainable economic development. A critical challenge in this issue lies in the complex and heterogeneous nature of crash causation and outcomes, which stem from the dynamic interplay of human, vehicle, road, and environmental factors. This complexity is further amplified by the emergence of autonomous vehicles (AVs), whose distinct crash patterns are not yet fully understood. To advance road safety, this thesis systematically analyses post-crash injury and damage severities for both conventional human-driven vehicles (HDVs) and AVs, thus forging evidence-based strategies that can contribute to a safer road traffic environment. | en_US |
| dcterms.abstract | Firstly, to investigate the heterogeneous impact of factors determining different crash injury severity levels under adverse road conditions for conventional single-vehicle rollover crashes, the random parameters logit model with unobserved heterogeneity in means and variances was employed. A series of likelihood ratio tests and marginal effects were used to assess the temporal instability or impact location non-transferability of the explanatory variables. The results indicate an overall temporal and locational instability of model estimates while several determinants are identified to have consistent effects on injury severity outcomes, such as animal-involved collisions, old drivers, safety restraint usage, female drivers, and physically impaired drivers. | en_US |
| dcterms.abstract | Secondly, to explore the safety interaction between conventional vehicles and railways and investigate crash injury severities at undivided two-way highway-railway grade crossings (HRGCs), the XGBoost model was applied to identify important features for various injury severity outcomes. The results indicate that some significant predictors overlap between the killed and injured models, such as old driver, driver was in vehicle, main track, went around the gate, adverse crossing surface, and truck, while the other significant factors reveal that the XGBoost model can distinguish between different severity levels. Moreover, to further compare factors affecting post-crash injury severities for conventional vehicles before and during the COVID-19 pandemic, a multinomial logit model was developed for 2018-2019 HRGC crashes, and a random parameters logit model was developed for the 2020-2021 HRGC crashes. The results indicate that the old driver and illuminated crossing were identified as random parameters with heterogeneity in means during the pandemic. As for the pre-pandemic period, no significant random parameters were captured. Significant variables also differ markedly between these two periods, with no significant driver-related or vehicle-related indicators identified for 2018-2019. | en_US |
| dcterms.abstract | Thirdly, to understand the heterogeneity and identify key factors influencing the damage severity in AV crashes, a methodology integrating the latent class analysis and multinomial logit model was employed to explore the heterogeneity in AV crash patterns. Two heterogeneous clusters were determined based on skewed distributions of vehicle status and driving mode. The model estimation results indicate that partitioning the AV crash dataset into heterogeneous subsets facilitates the identification of critical factors that remain obscured when the dataset is analysed as a whole, such as the evening indicator. | en_US |
| dcterms.abstract | Fourthly, to disentangle the direct and mediation effects of key factors on AV crash damage severities, the structural equation modelling was employed with additional point-of-interest characteristics. The results indicate direct pathways between AV crash outcomes and multiple risk factors, such as dark with streetlights and bus stop count. Results also reveal the impacts of two mediator variables, including zebra crossing and vulnerable road user. | en_US |
| dcterms.abstract | Fifthly, to uncover the combinations of contributing factors that frequently co-occur in AV rear-end crashes, the association rule mining was employed with supplementary built environment data. Besides, HDV rear-end crashes were also collected to conduct the comparative analysis between AV and HDV crashes. The results confirm the existence of distinct patterns, individual risk factors, and co-occurrence mechanisms that are unique to AV systems in real-world contexts. Critical factors such as roadside parking, two-way without medians, multiple vehicles, traffic signals, intersections, and land types play pivotal roles in AV rear-end crashes, while HDV crashes are more frequently influenced by behavioural context and land use type. | en_US |
| dcterms.abstract | The findings of this thesis highlight the importance of accounting for unobserved heterogeneity, temporal instability, mediation effects, and built environment factors when analysing HDV and AV crash severity outcomes. These enhanced understandings of crash patterns can inform the development of evidence-based safety strategies. | en_US |
| dcterms.extent | xviii, 213 pages : color illustrations | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2026 | en_US |
| dcterms.educationalLevel | Ph.D. | en_US |
| dcterms.educationalLevel | All Doctorate | en_US |
| dcterms.accessRights | open access | en_US |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/14533