Search

The object of this study is a clustering algorithm using various technologies. This paper compares clustering algorithms that are more commonly used to analyze urban road network trajectories, the growth curve model, with the elbow method and the x-means algorithm. Experiments were conducted with various volumes of big data to determine calculation time, accuracy, and ways to increase calculation time. These results can be used to manage traffic jams in congested areas and city streets. Considering the widespread use of clustering algorithms for solving various problems, this study proposes to introduce GCM, SPGK methods for monitoring and analyzing the state of congestion on city roads. The work was carried out in the following steps: research and selection of methods based on efficiency and time, implementation of parallel computing technologies to improve computation speed, demonstration of the selected method based on collected data from a real city with visualization of the results. The growth curve model algorithm has been proven to be almost 5 times more effective than the elbow method and the x-means algorithm. The time allocated for data processing has been calculated. An increase in the volume of processed data showed an almost stable execution time t=3 s for the GCM algorithm for data with a volume of up to almost 2,000 units. The effectiveness of SPGK-means was shown for different values of the number of points. Models of the Chengdu transport network obtained using a clustering algorithm with maximum grid density of neighborhoods are presented. There are some deviations between the grid and the road network due to the large grid size. This error is explained by an error of up to one between the points and the real grid. The results obtained clearly show how optimization of congested roads can be influenced. They provide information to obtain data on available routes, which allows you to analyze the road network individually and as a whole. [ABSTRACT FROM AUTHOR]

Smart cities can effectively improve urban life quality. But, with the rise in population size, there is an increase in the use of vehicles for transportation. In smart city development, traffic control and route planning are the primary tasks to address the problem of travel time estimation. It has become complex due to concerns of wrapped spatio-temporal data on dynamic real-time traffic conditions. The existing methods failed to estimate travel time efficiently due to not considering the spatio-temporal features and lack of computing resources. There is a need for a system like intelligent transportation system to monitor and accurately predict traffic flow to avoid traffic congestion and reduce the impact on the ecological system. This paper developed a novel hybrid deep learning model to estimate optimized travel time and possible trajectories. In this work, U-Net used to reduce the number of feature points for each temporal data and GNN is used to build the graph with connectivity between vehicular nodes. This hybrid model helps us predict traffic flow and aims to estimate accurate travel time from one location (node) to another. The model's performance is evaluated on the standard benchmark datasets Q-Traffic, TaxiBJ, and Chengdu. The experimental results show that the proposed framework significantly improves performance in extracting travel patterns and provides an optimal route with an estimated travel time than existing methods with RMSE 4%, MAE 20.49%, and MAPE 18%. The proposed model has accurately predicted the estimation of travel time of the vehicle for the given urban traffic data. [ABSTRACT FROM AUTHOR]

Traffic congestion and frequent traffic accidents have become the main problems affecting urban traffic. The effective location prediction of vehicle trajectory can help alleviate traffic congestion, reduce the occurrence of traffic accidents, and optimize the urban traffic system. Vehicle trajectory is closely related to the surrounding Point of Interest (POI). POI can be considered as the spatial feature and can be fused with trajectory points to improve prediction accuracy. A Local Dynamic Graph Spatiotemporal–Long Short-Term Memory (LDGST-LSTM) was proposed in this paper to extract and fuse the POI knowledge and realize next location prediction. POI semantic information was learned by constructing the traffic knowledge graph, and spatial and temporal features were extracted by combining the Graph Attention Network (GAT) and temporal attention mechanism. The effectiveness of LDGST-LSTM was verified on two datasets, including Chengdu taxi trajectory data in August 2014 and October 2018. The accuracy and robustness of the proposed model were significantly improved compared with the benchmark models. The effects of major components in the proposed model were also evaluated through an ablation experiment. Moreover, the weights of POI that influence location prediction were visualized to improve the interpretability of the proposed model. [ABSTRACT FROM AUTHOR]

Online ride-splitting has rapidly evolved worldwide, providing new means to enhance the efficiency of urban travel and reduce the number of vehicles and traffic congestion. While ride-splitting is currently believed to improve vehicle occupancy rates and decrease total travel distances, passengers may perceive it as resulting in additional travel due to ride sharing. To gain a more intuitive understanding of the current operations of ride-splitting, this study delves into carpooling behavior and analyzes its relative impact using GPS trajectory data from DiDi Chuxing in Chengdu, China. The investigation focuses on the behavior of ride-splitting and examines its relative impact based on disparities in total vehicle miles traveled (VMT) and travel duration between ride-splitting traveling (RST) and corresponding separated traveling (ST), aiming to estimate the effect of RST on urban mobility. Following are the three main findings of this study: (a) The existing operation of RST induces 7%–10% extra VMT and 20%–35% extra duration for RST users; (b) The RST achieves 30%–35% VMT and 10%–20% duration saving for drivers who are willing to provide RST service; (c) The RST saves up to 10% VMT and duration varying by time in a day. This paper offers some insights to quantify the current impact of RST and evaluate the orientation of mobility service quality improvement. • Introducing a novel approach to analyzing ride-splitting behaviors. • By leveraging rich GPS data, the research provides a comprehensive analysis of ride-splitting behaviors. • Identifying key limitations, including low ride-splitting rates, prompting future research directions to address this issue. • Providing valuable suggestions and emphasizing the integration of psychological perception benefits. [ABSTRACT FROM AUTHOR]

Tree-based and deep learning methods can automatically generate useful features. Not only can it enhance the original feature representation, but it can also learn to generate new features. This paper develops a strategy based on Light Gradient Boosting Machine (LightGBM or LGB) and Gated Recurrent Unit (GRU) to generate features to improve the expression ability of limited features. Moreover, a SARIMA-GRU prediction model considering the weekly periodicity is introduced. First, LightGBM is used to learn features and enhance the original features representation; secondly, GRU neural network is used to generate features; finally, the result ensemble is used as the input for prediction. Moreover, the SARIMA-GRU model is constructed for predicting. The GRU prediction consequences are revised by the SARIMA model that a better prediction can be obtained. The experiment was carried out with the data collected by Ride-hailing in Chengdu, and four predicted indicators and two performance indexes are utilized to evaluate the model. The results validate that the model proposed has significant improvements in the accuracy and performance of each component. [ABSTRACT FROM AUTHOR]

As the metropolitan city in Western China, Chengdu has been suffered from serious traffic congestion. The strategy of urban public transport priority was put into agenda to relieve traffic congestion. But the public transport sharing rate is only 27% in Chengdu which is much lower than the developed country. Consequently, it is of great importance to study the measures to improve the service, and provide technical support to the policy-makers. This paper selected the traffic corridor between Southwest Jiaotong University district and downtown as the experiment subject. The orthogonal design was used to generate stated preference questionnaires in order to achieve the reliable parameter estimates. Some variables were used to define the utility of the three alternatives and construct the Logit model. Then, the relationships between the cost, time variable and the choice probability of the public transport were analyzed. According to the results, we found that the orthogonal design does improve the goodness-of-fit. The workability of Multinomial Logit Model was better than Nest Logit model. We also put forward some effective measures to improve the service level of public transit, including reducing the access time to Metro, limiting parking supply to control the car use. [ABSTRACT FROM AUTHOR]

• Use ride-sharing vehicle trajectories data to statistically estimate real-world NDL. • Compute zone-to-zone travel time data from trajectory-level RV data to estimate NDL. • Provide a data sharing scheme that does not real individual or business information. • DNL-based user-optimal traffic routing reduces traffic congestion. • Though the NDL pattern varies from day to day, its average is stable and interpretable. Transportation systems are being reshaped by ride sourcing and shared mobility services in recent years. The transportation network companies (TNCs) have been collecting high-granular ride-sourcing vehicle (RV) trajectory data over the past decade, while it is still unclear how the RV data can improve current dynamic network modeling for network traffic management. This paper proposes to statistically estimate network disequilibrium level (NDL), namely to what extent the dynamic user equilibrium (DUE) conditions are deviated in real-world networks. Using the data based on RV trajectories, we present a novel method to estimate the real-world NDL measure. More importantly, we present a method to compute zone-to-zone travel time data from trajectory-level RV data. This would become a data sharing scheme for TNCs such that, while being used to effectively estimate and reduce NDL, the zone-to-zone data reveals neither personally identifiable information nor trip-level business information if shared with the public. In addition, we present an NDL based traffic management method to perform user optimal routing on a small fraction of vehicles in the network. The NDL measures and NDL-based routing are examined on two real-world large-scale networks: the City of Chengdu with trajectory-level RV data and the City of Pittsburgh with zone-to-zone travel time data. We found that, on weekdays in each city, NDLs are likely high when travel demand is high (thus when congestion is mild or heavy). Generally, a weekend midnight exhibits higher NDLs than a weekday midnight. Many NDL patterns are different between Chengdu and Pittsburgh, which are attributed to unique characteristics of both demand and supply in each city. For instance, NDL in Pittsburgh is much more stable from day to day and from hour to hour, comparing to Chengdu. In addition, we observe that origin-destination pairs with high NDLs are spatially and temporally sparse for both cities. For the Pittsburgh network, we evaluate the effectiveness of NDL-based traffic routing, which shows great potential to reduce total travel time with routing a small fraction of vehicles (1% in the experiments), even using dated NDL that is estimated in the prior hour. [ABSTRACT FROM AUTHOR]

Traffic prediction is a necessary function in intelligent transportation systems to alleviate traffic congestion. Graph learning methods mainly focus on the spatiotemporal dimension, but ignore the nonlinear movement of traffic prediction and the high-order relationships among various kinds of road segments. There exist two issues: 1) deep integration of the spatiotemporal information and 2) global spatial dependencies for structural properties. To address these issues, we propose a nonlinear spatiotemporal optimization method, which introduces hypergraph convolution networks (HGCN). The method utilizes the higher-order spatial features of the road network captured by HGCN, and dynamically integrates them with the historical data to weigh the influence of spatiotemporal dependencies. On this basis, an extended Kalman filter is used to improve the accuracy of traffic prediction. In this study, a set of experiments were conducted on the real-world dataset in Chengdu, China. The result showed that the proposed method is feasible and accurate by two different time steps. Especially at the 15-minute time step, compared with the second-best method, the proposed method achieved 3.0%, 11.7%, and 9.0% improvements in RMSE, MAE, and MAPE, respectively. [ABSTRACT FROM AUTHOR]

Rapid urbanization has brought along with it many environmental and social problems such as ecosystem damage and traffic congestion. Therefore, forecasting the trend of urban expansion and providing a reasonable urban planning basis for government departments have become the focus of researchers. An artificial neural network (ANN) can be used to consider spatial nonstationarity when obtaining the changing characteristics of urban land types. Therefore, in this study, we use cellular automata (CA) based on ANN (ANN-CA) to simulate and forecast urban expansion and discuss the parameter sensitivity of the model in detail. In addition, we propose a new Urban Expansion Early Warning Indicator system to warn about the deterioration of future land distribution patterns. Chengdu is selected as the study area, and the study period is from 2000 to 2020. The results showed the following: (1) The best accuracy was achieved when the neighborhood size is 7 × 7 and the number of model iterations is 250, and overall accuracy (OA), Kappa coefficient, and figure of merit (FOM) are 91.47%, 0.855, and 0.354, respectively; (2) ANN-CA is more suitable for predicting the urban expansion of Chengdu than CA based on logistic regression (LR-CA) and CA based on decision tree (DT-CA). Compared with the worst performance model, the score of OA increased by 6.23%, that of kappa increased by 0.062, and that of FOM increased by 0.056. (3) According to the current development trend, artificial built-up areas will increase substantially. The comprehensive evaluation results of the morphology effect, ecological effect, and intensity effect of urban expansion predict severe early warning for Jinniu District, Qingyang District, and Wuhou District by 2030. [ABSTRACT FROM AUTHOR]

The average speed (AS) of a road segment is an important factor for predicting traffic congestion, because the accuracy of AS can directly affect the implementation of traffic management. The traffic environment, spatiotemporal information, and the dynamic interaction between these two factors impact the predictive accuracy of AS in the existing literature, and floating car data comprehensively reflect the operation of urban road vehicles. In this paper, we proposed a novel road segment AS predictive model, which is based on floating car data. First, the impact of historical AS, weather, and date attributes on AS prediction has been analyzed. Then, through spatiotemporal correlations calculation based on the data from Global Positioning System (GPS), the predictive method utilizes the recursive least squares method to fuse the historical AS with other factors (such as weather, date attributes, etc.) and adopts an extended Kalman filter algorithm to accurately predict the AS of the target segment. Finally, we applied our approach on the traffic congestion prediction on four road segments in Chengdu, China. The results showed that the proposed predictive model is highly feasible and accurate. [ABSTRACT FROM AUTHOR]

Sequential coupling of applications is the major cause of traffic congestion and low application quality. Modeling traffic congestion under sequential coupling applications is the primary means to explore congestion mechanism, which can further support application design and adjustment to improve application quality. Existing congestion models concentrate more on traffic flow modeling based on network components, with little consideration of the influence of network application and its coupling. This paper proposes a network-cell-based congestion model for describing dynamics of congestion propagation under sequential coupling multi-applications. First, the basic characteristics of network applications are analyzed, and network applications are abstracted into network cells as functional units, which is specifically described from three perspectives: Structure, function and connection. The dynamic process of network can be modeled as intracellular reaction (single application function behavior) and intercellular interaction (coupling behavior of multi-applications). Furthermore, a network-cell-based metric to evaluate application quality is presented, and the implementation algorithm of the model is also given. Numerical simulations are implemented with comparison of typical information flow model to verify the model. Moreover, the proposed model is applied to Chengdu Metro system case to show the applicability. Furthermore, the influence of sequential coupling on congestion is further investigated, and the results show that the larger coupling strength corresponds to less congestion in peak period. This study provides guidance for application design and management to improve application quality of transportation networks. • The impact of sequential coupling on traffic congestion is investigated. • Sequential coupling applications are modeled as network cells and their connections. • A network-cell-based metric to evaluate application quality is presented. • The proposed method is applied to modeling congestion of real metro systems. [ABSTRACT FROM AUTHOR]

The congestion propagation pattern in urban traffic networks has long been viewed as a complex question. Inspired by the fact that virus spread patterns are similar to traffic congestion propagation in some ways, this paper proposes a congestion propagation model (SIS-CP) to describe the congestion propagation patterns in large-scale traffic networks with very few parameters. In this SIS-CP model, the attribute of the real traffic network in the actual operation process is considered, a method based on Hilbert–Huang transform (HHT-based) is proposed to determine traffic states of links, and the congestion propagation patterns are analyzed. The real-world data of 'DiDi Chuxing GAIA Open Dataset Initiative of Chengdu city from March to June is used to calibrate and validate the proposed models. The results show that the basic and limit value of congestion links' ratio do exist in real traffic networks; the relationship between congestion links' ratio and its derivative is conformed to the parabolic function; the HHT-based method can determine traffic states of links accurately; and the SIS-CP model can describe the congestion propagation patterns correctly. [ABSTRACT FROM AUTHOR]

• A spatial-temporal density-based clustering algorithm is proposed for trip end detection. • Three scenario optimization models are applied to improve the method capability. • A smartphone application is developed for GNSS data collection. • The potential of the method is demonstrated by comparing results with travel log. As a widespread portable probe, smartphone equipped with Global Navigation Satellite System (GNSS) can continuously track individual's travel trajectory, it is potential for trip end information identification. Existing studies mostly use rule-based methods with specific dwelling time and distance thresholds for trip end detection, these methods are highly dependent on expert experience and lack universality across different traffic environments. Some density-based spatial clustering methods also have issues in the case of GNSS signal missing, traffic congestion, short-time stays and repeat visits to the same place etc. Therefore, this paper proposes a two-step method for trip end identification by using smartphone GNSS positioning data. First, a spatial-temporal density-based clustering algorithm (ST-DBCA) is proposed for trip end identification, the method considers both spatial and temporal travel trajectory point density at the same time, and performs much better than traditional clustering methods. Second, three optimization models are further proposed to optimize the identification results, including 1) a short time stay optimization model, 2) a redundant stay optimization model, and 3) a traffic congestion stay optimization model. Field tests in Chengdu China are conducted to verify the feasibility and effectiveness of the proposed methods. Results show that the average trip end identification accuracy under different trip purposes reaches 92.8%, and the average errors of arrival and departure time are smaller than 150 s. [ABSTRACT FROM AUTHOR]

Nowadays, traffic situation prediction is a major concern. Travel time index prediction is one of the representative tasks in traffic situation prediction. Most research of the current prediction focuses on road-level speed, but the travel time index prediction of complex road network at the regional level has not been well solved. In this paper, we proposed a novel model with convolutional neural network called Seq2Img-DRNET. The sequence data with 10 min interval are converted into image data as the input of our model. The model is constructed by fusing dense connection network and residual network to solve the travel time index prediction of complex road network at the regional level. Taking two actual traffic networks of commercial district and scenic district in Chengdu as the example, we compared our model with common sequential model and convolution neural network model, and the results show that our model can effectively capture the spatial–temporal relationship characteristics of complex road network at the regional level and make accurate prediction. • Our method transforms time series data into image data for spatiotemporal information fusion. • A convolutional neural network model is constructed to accurately predict the travel time index. • Aiming at urban business districts and scenic spots prone to traffic congestion. • The prediction accuracy is significantly higher than the existing baseline model. [ABSTRACT FROM AUTHOR]

Reducing car use is commonly considered as a potential strategy to reduce transport-related problems such as traffic congestion and air pollution. The increasing use of online shopping may potentially replace shopping trips, thus possibly reducing car use. However, car owners – compared to non-car owners – can more easily visit physical stores and transport goods. Therefore, it can be assumed that online shopping is less likely to reduce shopping trips for car owners. Using 653 structured face-to-face interviews in Chengdu (China) in 2016, an empirical study is conducted. The results show that 44.0% of respondents indicated a decrease in shopping trip frequency after they started to purchase online, while only 14.4% indicated an increase in the frequency. This confirms that online shopping tends to be a substitute for shopping trips. Applying a propensity score matching approach, this paper further compares the likelihoods of changes in shopping trips caused by online buying between car owners and non-car owners, while considering sociodemographic factors, internet experiences, spatial attributes, and online shopping attitudes as covariates. The results indicate that – due to online buying – shopping trip frequency is less likely to decrease for car owners compared to non-car owners, while there is no significant difference in the likelihood of increasing shopping trip frequency between owners and non-owners. These findings imply that online shopping may not effectively reduce driving, thus unlikely being a valid solution for transportation problems resulting from the increasing use of cars. • Propensity score matching is applied to infer causality; • Shopping trips can be partly replaced by e-shopping; • E-shopping is less likely to replace shopping trips for car owners; • E-shopping may not effectively reduce driving. [ABSTRACT FROM AUTHOR]

• Identify three key factors in ride-pooling service. • Investigate the relationship between average passengers' detour distance and demand. • Investigate the relationship between average vehicle routing distance and demand. • Extend the law of pool-matching probability. • Verify our findings with real data from three representative cities. By serving two or more passenger requests in each ride in ride-sourcing markets, ride-pooling service is now becoming an important component of shared smart mobility. It is generally expected to improve vehicle utilization rate, and therefore alleviate traffic congestion and reduce carbon dioxide emissions. A few recent theoretical studies are conducted, mainly focusing on the equilibrium analysis of the ride-sourcing markets with ride-pooling services and the impacts of ride-pooling services on transit ridership and traffic congestion. In these studies, there are three key measures that distinguish ride-pooling service analysis from the non-pooling ride-sourcing market analysis. The first is the proportion of passengers who are pool-matched(referred to as pool-matching probability), the second is passengers' average detour distance, and the third is average vehicle routing distance to pick up and drop off all passengers with different origins and destinations in one specific ride. These three measures are determined by passenger demand for ride-pooling and matching strategies. However, due to the complex nature of ride-resourcing market, it is difficult to analytically determine the relationships between these measures and passenger demand. To fill this research gap, this paper attempts to empirically ascertain these relationships through extensive experiments based on the actual on-demand mobility data obtained from Chengdu, Haikou, and Manhattan. We are surprised to find that the relationships between the three measures (pool-matching probability, passengers' average detour distance, average vehicle routing distance) and number of passengers in the matching pool (which reflects passenger demand) can be fitted by some simple curves (with fairly high goodness-of-fit) or there exist elegant empirical laws on these relationships. Our findings are insightful and useful to theoretical modeling and applications in ride-resourcing markets, such as evaluation of the impacts of ride-pooling on transit usage and traffic congestion. [ABSTRACT FROM AUTHOR]

With the continuous growth of traffic demand and the mismatch of urban transportation facilities, urban traffic congestion has been caused, leading to various related problems, such as environmental pollution, traffic accidents, and slow economic development. Many cities have implemented relevant measures to improve traffic congestion, but fewer are ideal. This study used the hidden Markov model combined with the dissipative structure theory and entropy theory to predict the congestion more accurately. The temporal and spatial distributions of the online ride-hailing Didi data in Chengdu were analyzed. There are morning peaks, noon peaks, and evening peaks during workdays. During the noon peak and evening peak, travel demand in the city's central area is relatively stable. It is found that the prediction model has a higher accuracy after combining the dissipative structure theory and entropy theory, which could be used to propose methods to prevent congestion. [ABSTRACT FROM AUTHOR]

Ridesourcing services play a crucial role in metropolitan transportation systems and aggravate urban traffic congestion and air pollution. Ridesplitting is one possible way to reduce these adverse effects and improve the transport efficiency, especially during rush hours. This paper aims to explore the potential of ridesplitting during peak hours using empirical ridesourcing data provided by DiDi Chuxing, which contains complete datasets of ridesourcing orders in the city of Chengdu, China. A ridesplitting trip identification algorithm based on a shareability network is developed to quantify the potential of ridesplitting. Then, we evaluate the gap between the potential and actual scales of ridesplitting. The results show that the percentage of potential cost savings can reach 18.47% with an average delay of 4.76 min, whereas the actual percentage is 1.22% with an average delay of 9.86 min. The percentage of shared trips can be increased from 7.85% to 90.69%, and the percentage of time savings can reach 25.75% from 2.38%. This is the first investigation of the gap between the actual scale and the potential of ridesplitting on a city scale. The proposed ridesplitting algorithm can not only bring benefits on a city level but also take passenger delays into consideration. The quantitative benefits could encourage transportation management agencies and transportation network companies to develop sensible policies to improve the existing ridesplitting services. • The percentage of cost savings can be increased from 1.22% to 18.47%. • The percentage of shared trips can be increased from 7.85% to 90.69%. • The percentage of time savings can reach 25.75% from 2.38%. • The average delay can be reduced from 9.86 min to 4.76 min. • Passengers wait for less time on average. [ABSTRACT FROM AUTHOR]