As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
Tunnels are an essential component of urban transportation. Compared to open roads, accidents inside tunnels tend to have longer durations, and the road closures resulting from accidents have a greater impact on traffic operations, particularly in the case of submarine tunnels. This study focuses on investigating the characteristics and trends of accident duration in submarine tunnels based on accident data from Qingdao Jiaozhou Bay in China from 2018 to 2020. Firstly, the study examines the influence of ten variables, including the number of vehicles involved, accident types, and weather conditions, on the duration of accidents. The data indicate that the manner and quantity of vehicles leaving the accident scene are critical factors affecting accident duration, while weather conditions have no significant impact. Furthermore, considering the correlations among the influencing factors and the high-dimensional sparsity of the data, a PCA-LGBM model for accident duration prediction is constructed. This model combines the dimensionality reduction capability of Principal Component Analysis (PCA) with the powerful prediction capability of LightGBM (LGBM). Finally, experimental results demonstrate that compared to other models such as MLR, BPNN, PCA-BPNN, and LGBM, the proposed model exhibits superior performance with a minute-level accuracy rate of 75%.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.