Traffic violation is one of the leading causes traffic crashes. In context global aging, it important to study violations by elderly drivers for improving safety in preparation a worldwide aging population. this study, hybrid approach Latent Class Analysis (LCA) and XGBoost based SHAP proposed identify hidden clusters understand key contributing factors on severity drivers, police-reported dataset Beijing (China). First, LCA applied segment into several latent homogeneous clusters, then...

The uncontrolled growth in urbanization and motorization generally contributes to an urban land use transportation system that is socially, economically, environmentally unsustainable. This paper uses Karachi as a case study, which the largest economic centre of Pakistan, passing through phase rapid motorization. first reviews research related sustainable systems comprehend concept development transportation. then evaluates existing infrastructure system, national policies, projects...

The purpose of this paper is to investigate the existence stratification heterogeneity in traffic accidents Shenzhen, what factors influence casualties, and interaction those factors. Geographical detection methods are used for analysis Shenzhen. Results show that spatial does exist, influencing fatalities injuries different. accident causes types primary responsible party have a strong impact on injuries, followed by zones time interval. However, road factors, lighting, topography, etc.,...

This study explores risk factors influencing the at-fault party in traffic accidents and analyzes their impact on accident severity. Based data of Shenyang City, Liaoning Province, China, from 2018 to 2020, 19 attribute variables including road attributes, time environmental characteristics parties with either full responsibility, primary or equal responsibility were extracted analyzed conjunction built environment such as network density POI (points interest) at sites accidents. Using...

With the rapid development of urban informatization, era big data is coming. To satisfy demand traffic congestion early warning, this paper studies method real-time flow state identification and prediction based on data-driven theory. Traffic holds several characteristics, such as temporal correlation, spatial historical multistate. quantification, basis identification, achieved by a SAGA-FCM (simulated annealing genetic algorithm fuzzy c -means) clustering model. Considering simple...

Countdown signals can be found in many countries; however, there are only a few researches that study the impact of on driver behavior. In this study, we focus particularly behavior drivers reacting to countdown when signal shifted from red green. Although reduce startup delay, it also causes premature start. This paper determines factors influencing behaviors start, and headway due signals. The analysis results compared between Japan Turkey. From our findings, propose countermeasures...

Road crashes cause significant traffic delay and bring unnecessary financial losses. This study investigates the impact of contributing factors on level caused by (LDC) using Texas crash data. To capture unobserved heterogeneity, a latent class analysis (LCA) was first used to segment whole dataset into several homogeneous clusters. Then, XGBoost based SHAP developed each cluster identify main hidden in classes. The interaction effects between were subsequently analyzed, including high...

Traffic flow in a road network is mutually interactive and interdependent with each other. It challenging to describe the dynamics of traffic by using analytical methods. In this article, deep convolutional neural (DCNN) model employed address prediction. To improve parameter learning efficiency confronting big data, parallel training approach developed for DCNN prediction model. The theoretical foundation algorithm A master-slave computing solution implemented on Spark cloud. Real data are...

Bicycle–motor vehicle (BMV) crashes have been identified as a major type of traffic accident affecting transportation safety. In order to determine the characteristics BMV in cold regions, this study presents an analysis using police-reported data from 2015 2017 on Shenyang, China. A two-stage approach integrating latent class (LCA) and random parameter logit (RP-logit) model is proposed identify specific crash groups explore their contributing factors. First, LCA was used classify into...

When dealing with traffic big data under the background of Internet Things (IoT), control single-machine computing environment is difficult to adapt massive and rapid analysis decision-making. To tackle this problem, we propose a parallel approach network flow based on mechanism model predictive (MPC). A non-analytical rule-based developed forecast vehicle movements in prediction horizon according real-time feedback information flow, evaluate performances candidate strategies. Furthermore,...

Urban modelling is an important branch of land use science. It integrates geography, surveying and mapping, information science, system economics, sociology other disciplines to establish urban models, which have been used provide support for policymaking or analyses. models are understand, analyse, evaluate reproduce the process development, predict consequence planning scenarios. In this paper, we a systematic review including evaluation, classification, application timeline models'...

Travel by the elderly is generally characterized relatively short distances and low frequencies. However, with social development, travel patterns of are lacking in sustainability. In some European countries, increasingly dependent on cars while traveling. addition, influencing factors elderly’s behavior also changing. At present, most foreign domestic studies focus impact individual household socioeconomic attributes, policy frequency, mode choice, demand elderly, but they lack an analysis...

In an envisioned variable speed limit system, limits vary strategically and periodically in accordance with traffic weather conditions to achieve better safety, efficiency, environmental sustainability on a network. For such three-objective, bi-level programming model is proposed design optimal link-specific that minimize system travel time, number of expected accidents, exhaust emissions simultaneously for each specific strategic scenario. A solution framework developed solve the implements...

The theory of complex networks was used to classify public transport into transportation route networks, transfer and bus station networks. practical significance the network parameters then analyzed. in Langfang, Jining, Dalian were chosen as specific research cases. results show that have characteristics In addition, urban all significantly affect accessibility, convenience, terrorist security capability network. link findings with actual situations explore means solve system problems.