Traffic Flow Prediction Model Research Articles

Variable speed limit: A microscopic analysis in a connected vehicle environment

This paper presents a Variable Speed Limit (VSL) control algorithm for simultaneously maximizing the mobility, safety and environmental benefit in a Connected Vehicle environment. Development of Connected Vehicle (CV)/Autonomous Vehicle (AV) technology has the potential to provide essential data at the microscopic level to provide a better understanding of real-time driver behavior. This paper investigated a VSL control algorithm using a microscopic approach by focusing on individual driver’s behavior (e.g., acceleration and deceleration) through the use of Model Predictive Control (MPC) approach. A multi-objective optimization function was formulated with the aim of finding a balanced trade-off among mobility, safety and sustainability. A microscopic traffic flow prediction model was used to calculate Total Travel Time (TTT); a surrogate safety measure Time To Collision (TTC) was used to measure instantaneous safety; and, a microscopic fuel consumption model (VT-Micro) was used to measure the environmental impact. Real-time driver’s compliance to the posted speed limit was used to adjust the optimal speed limit values. A sensitivity analysis was conducted to compare the performance of the developed approach for different weights in the objective function and for two different percentages of CV. The results showed that with 100% penetration rate, the developed VSL approach outperformed the uncontrolled scenario consistently, resulting in up to 20% of total travel time reductions, 6–11% of safety improvements and 5–16% reduction in fuel consumptions. Our findings revealed that the scenario which optimized for safety alone, resulted in more optimum improvements as compared to the multi-criteria optimization. Thus, one can argue that in case of 100% penetration rates of CVs, optimizing for safety alone is enough to achieve simultaneous and optimum improvements in all measures. However, mixed results were obtained in case of lower % penetration rate which showed higher collision risk when optimizing for only mobility or fuel consumption. This indicates that with such % penetration rate, multi-criteria optimization is crucial to realize optimum and balanced benefits for the examined measures.

Robust causal dependence mining in big data network and its application to traffic flow predictions

In this paper, we focus on a special problem in transportation studies that concerns the so called “Big Data” challenge, which is: how to build concise yet accurate traffic flow prediction models based on the massive data collected by different sensors? The size of the data, the hidden causal dependence and the complexity of traffic time series are some of the obstacles that affect making reliable forecast at a reasonable cost, both time-wise and computation-wise. To better prepare the data for traffic modeling, we introduce a multiple-step strategy to process the raw “Big Data” into compact time series that are better suited for regression and causality analysis. First, we use the Granger causality to define and determine the potential dependence among data, and produce a much condensed set of times series who are also highly dependent. Next, we deploy a decomposition algorithm to separate daily-similar trend and nonstationary bursts components from the traffic flow time series yielded by the Granger test. The decomposition results are then treated by two rounds of Lasso regression: the standard Lasso method is first used to quickly filter out most of the irrelevant data, followed by a robust Lasso method to further remove the disturbance caused by bursts components and recover the strongest dependence among the remaining data. Test results show that the proposed method significantly reduces the costs of building prediction models. Moreover, the obtained causal dependence graph reveals the relationship between the structure of road networks and the correlations among traffic time series. All these findings are useful for building better traffic flow prediction models.

A Network Traffic Prediction Model Based on Quantum Inspired Pso and Wavelet Neural Network

Network traffic flow prediction model is fundamental to the network performance evaluation and the design of network control scheme which is crucial for the success of high-speed networks. Aiming at shortcoming of the conventional network traffic time series prediction model and the problem that BP training algorithms easily plunge into local solution, a network traffic prediction model based on wavelet neural network and PSO-QI is presented in the paper. Firstly, the quantum principle obtained from Quantum PSO(QPSO)has been combined with standard PSO to form a new hybrid algorithm called PSO with Quantum Infusion(PSO-QI). Then, the parameters of wavelet neural network were optimized with PSO-QI and the time series of network traffic data was modeled and predicted based on wavelet neural network and PSO-QI. Experiments showed that PSOQI-wavelet neural network has better precision and adaptability compared with the traditional neural network.

Combining the Statistical Model and Heuristic Model to Predict Flow Rate

Statistical and heuristic models have been proposed as applications that are well suited to short-term traffic flow prediction. However, traffic flow data often contain both linear and nonlinear patterns. Therefore, neither statistical nor heuristic models are adequate to model and predict traffic flow data. This paper discusses the relative merits of statistical and heuristic models for traffic flow prediction and summarizes the findings from a comparative study for their performances. Based on that, a hybrid support vector machine for regression (SVR) methodology that combines both statistical and heuristic models is proposed to take advantage of their unique strength in linear and nonlinear modeling. In addition, the dynamics of spatial-temporal patterns in traffic flow are considered in this study, and they are treated as part of the input data. The experiment results based on the real field data of a test region in Beijing suggest that the proposed method is able to provide accurate and reliable flow rate predictions under both low- and high-flow traffic conditions. The benefit from combining statistical and heuristic models as opposed to not combining [autoregressive integrated moving average (ARIMA) model or Elman neural network (NN)] is much more evident in all cases, and the accuracy can be improved by 9.04% on average. Regarding the incorporation of a combination of temporal and spatial characteristics, the use of the hybrid model is found helpful in a one-step-ahead flow rate prediction under high-flow traffic conditions, with a maximum 9.52% improvement on accuracy.

A Comparison of Detrending Models and Multi-Regime Models for Traffic Flow Prediction

Short-term traffic flow prediction received considerations from different fields, because of its essentialness in traffic engineering and its theoretical difficulties. In this paper, we studied two important approaches of traffic flow prediction: detrending methods and multi-regime methods. First, we compared their differences in modeling philosophy and compared their merits as well as shortcomings. Then, we tested several representative prediction models of these two approaches on the openly accessible PeMS traffic flow database to find their merits and shortcomings. The obtained results threw some interesting light on how to select the appropriate traffic prediction models in practices.

Significance of Sensor Location in Real-time Traffic State Estimation

Short-term congestion caused due to traffic incidents or other road environment factors significantly reduces traffic flow capacity of a link which forms a major part of travel time delays. Accurate and reliable estimate of real-time traffic state is essential for optimizing network performance during unpredictable events. Inaccurate estimate of current traffic state produces unreliable travel-time estimations which lead to ineffective traffic management strategies during traffic incident.This study highlights the accuracy and reliability of traffic state estimate when a traffic flow prediction model is not provided with information about duration and impact of the incident on traffic flow capacity of the link. Cell Transmission Model (CTM) is used for prediction of traffic state and measurements from the sensor are combined in Extended Kalman Filter (EKF) to minimize square of error between predicted and measured traffic state. A simple link is used to highlight the difference between actual traffic state and estimated traffic state using a naive prediction model for real-time traffic state estimation. Analysis of simulation results shows that estimate of traffic state is reliable and accurate for cells upstream of the measurement sensor when incident occurred downstream of measurement sensor. Whereas when incident location is upstream of measurement sensor, the estimated traffic state for downstream cells of measurement sensor is more close to actual traffic condition.

Traffic Flow Prediction Model for Large-Scale Road Network Based on Cloud Computing

To increase the efficiency and precision of large-scale road network traffic flow prediction, a genetic algorithm-support vector machine (GA-SVM) model based on cloud computing is proposed in this paper, which is based on the analysis of the characteristics and defects of genetic algorithm and support vector machine. In cloud computing environment, firstly, SVM parameters are optimized by the parallel genetic algorithm, and then this optimized parallel SVM model is used to predict traffic flow. On the basis of the traffic flow data of Haizhu District in Guangzhou City, the proposed model was verified and compared with the serial GA-SVM model and parallel GA-SVM model based on MPI (message passing interface). The results demonstrate that the parallel GA-SVM model based on cloud computing has higher prediction accuracy, shorter running time, and higher speedup.

An Improved K-nearest Neighbor Model for Short-term Traffic Flow Prediction

In order to accurately predict the short-term traffic flow, this paper presents a k-nearest neighbor (KNN) model. Short-term urban expressway flow prediction system based on k-NN is established in three aspects: the historical database, the search mechanism and algorithm parameters, and the predication plan. At first, preprocess the original data and then standardized the effective data in order to avoid the magnitude difference of the sample data and improve the prediction accuracy. At last, a short-term traffic prediction based on k-NN nonparametric regression model is developed in the Matlab platform. Utilizing the Shanghai urban expressway section measured traffic flow data, the comparison of average and weighted k-NN nonparametric regression model is discussed and the reliability of the predicting result is analyzed. Results show that the accuracy of the proposed method is over 90 percent and it also rereads that the feasibility of the methods is used in short-term traffic flow prediction.

Dynamic Network Traffic Flow Prediction Model based on Modified Quantum-Behaved Particle Swarm Optimization

This paper aims at effectively predicting the dynamic network traffic flow based on quantum-behaved particle swarm optimization algorithm. Firstly, the dynamic network traffic flow prediction problem is analyzed through formal description. Secondly, the structure of the network traffic flow prediction model is given. In this structure, Users can used a computer to start the traffic flow prediction process, and data collecting module can collect and return the data through the destination device. Thirdly, the dynamic network traffic flow prediction model is implemented based on BP Neural Network. Particularly, in this paper, the BP Neural Network is trained by a modified quantum-behaved particle swarm optimization(QPSO). We modified the QPSO by utilizing chaos signals to implement typical logistic mapping and pursuing the fitness function of a particle by a set of optimal parameters. Afterwards, based on the above process, dynamic network traffic flow prediction model is illustrated. Finally, a series of experiments are conduct to make performance evaluation, and related analyses for experimental results are also given

Adaptive Correction Forecasting Approach for Urban Traffic Flow Based on Fuzzyc-Mean Clustering and Advanced Neural Network

Forecasting of urban traffic flow is important to intelligent transportation system (ITS) developments and implementations. The precise forecasting of traffic flow will be pretty helpful to relax road traffic congestion. The accuracy of traditional single model without correction mechanism is poor. Summarizing the existing prediction models and considering the characteristics of the traffic itself, a traffic flow prediction model based on fuzzyc-mean clustering method (FCM) and advanced neural network (NN) was proposed. FCM can improve the prediction accuracy and robustness of the model, while advanced NN can optimize the generalization ability of the model. Besides these, the output value of the model is calibrated by the correction mechanism. The experimental results show that the proposed method has better prediction accuracy and robustness than the other models.

Accurate Multisteps Traffic Flow Prediction Based on SVM

Accurate traffic flow prediction is prerequisite and important for realizing intelligent traffic control and guidance, and it is also the objective requirement for intelligent traffic management. Due to the strong nonlinear, stochastic, time-varying characteristics of urban transport system, artificial intelligence methods such as support vector machine (SVM) are now receiving more and more attentions in this research field. Compared with the traditional single-step prediction method, the multisteps prediction has the ability that can predict the traffic state trends over a certain period in the future. From the perspective of dynamic decision, it is far important than the current traffic condition obtained. Thus, in this paper, an accurate multi-steps traffic flow prediction model based on SVM was proposed. In which, the input vectors were comprised of actual traffic volume and four different types of input vectors were compared to verify their prediction performance with each other. Finally, the model was verified with actual data in the empirical analysis phase and the test results showed that the proposed SVM model had a good ability for traffic flow prediction and the SVM-HPT model outperformed the other three models for prediction.

Improved Intelligent Method for Traffic Flow Prediction Based on Artificial Neural Networks and Ant Colony Optimization

Real time traffic flow is often difficult to predict precisely because of the complexity, nonlinearity and uncertainty characteristics of the traffic flow data. Intelligent prediction methods such as artificial neural network (ANN), support vector machine (SVM), etc. have been proven effective to discover the nonlinear information hidden in the traffic flow data. Nevertheless, a single prediction model is difficult to ensure the prediction accuracy and efficiency. To overcome the lack of the single prediction method, this paper uses a prediction method that combining artificial neural networks (ANN) with ant colony optimization (ACO), called ANN-ACO, by exploiting complementary advantages of both approaches. Firstly, this method uses the ANN theory for data reduction pretreatment, and then constructs the traffic flow prediction model based on ACO according to the information structure. The analysis results show that the proposed method can extract the underlying rules of the testing data and decrease prediction error or better when compared with single ANN or ACO approach. Besides, the combined prediction model not only has fault tolerant and anti-jamming capability, but also can shorten the operation time and improve the speed of the system and also forecast accuracy. Hence, it can be used to forecast real-time traffic flow.

Short-term traffic flow prediction for multi traffic states on urban expressway network

Short-term traffic flow prediction for multi traffic states on urban expressway network is carried out in this paper. The model for multi traffic states is proposed by integrating the spatial and the temporal distribution characteristics of traffic flow parameters under free traffic, congested traffic and jam traffic respectively. Based on the classical traffic flow conservation equation, the ideology of spatial and temporal dispersions in partial differential equations is adopted to establish short-time traffic flow prediction model. Meanwhile, the impact factors, such as on and off ramp, lane change and road slope are considered, which convertes short-term traffic flow prediction model into short-time traffic flow prediction state space model. Finally, the short-term traffic flow prediction for multi traffic states on urban expressway network is realized. The empirical research shows that compared with the classic ARMA model, the proposed method can not only realize short-term traffic flow prediction for multi traffic states on urban expressway network but also achieve an accuracy of 90.23%. In the same condition, the accuracy of ARMA model is 81%.

Vehicle actuation based short-term traffic flow prediction model for signalized intersections

Traffic flow prediction is an important component for real-time traffic-adaptive signal control in urban arterial networks. By exploring available detector and signal controller information from neighboring intersections, a dynamic data-driven flow prediction model was developed. The model consists of two prediction components based on the signal states (red or green) for each movement at an upstream intersection. The characteristics of each signal state were carefully examined and the corresponding travel time from the upstream intersection to the approach in question at the downstream intersection was predicted. With an online turning proportion estimation method, along with the predicted travel times, the anticipated vehicle arrivals can be forecasted at the downstream intersection. The model performance was tested at a set of two signalized intersections located in the city of Gainesville, Florida, USA, using the CORSIM microscopic simulation package. Analysis results show that the model agrees well with empirical arrival data measured at 10 s intervals within an acceptable range of 10%–20%, and show a normal distribution. It is reasonably believed that the model has potential applicability for use in truly proactive real-time traffic adaptive signal control systems.

The construction of a route-diversion traffic flow prediction model and potential applications

This study aims to predict daily traffic volume during a roadwork period on a freeway or expressway network affected by the diversion behavior of road users by the construction of a reverse-S-Curve. This approach not only differs from the static methods in conventional transportation planning, which only obtains one equilibrium traffic volume at one point in time, it also avoids the daily learning and adaptive travel decision behavior theorem. Any variation in daily traffic volumes during the roadwork period are predicted by a simple and accurate method. Real-life traffic data on a roadwork section during long-term maintenance work on #3 National Freeway in Taiwan was provided by the National Taiwan-Area Freeway Bureau. After the data were selected and transformed, a reverse-S-Curve based on the logistic function was applied in the establishment and evaluation of the model. In addition to verification by statistical methods, this model demonstrates high-predictive accuracy based on the mean absolute percentage error (MAPE). This model can be used to predict gradually decreasing daily traffic flows towards another equilibrium level on roadwork sections undergoing long-term maintenance work in freeway or expressway networks. The shortcomings of past prediction methods are addressed with a view to effectively improving the construction of traffic volume forecasting methods. Finally, based on the characteristics of the reverse-S-Curve model, an innovative indicator-‘impact loss’ was developed as an evaluation criterion in roadwork plans. The model and the ‘impact loss’ features are considered in a concluding discussion of potential applications.

KF-MA Model for Short-Term Traffic Flow Prediction Based on SCATS Data

Aimed at the typical problems of lower accuracy and efficiency by using traditional Kalman Filtering (KF) to model the SCATS data, a KF model based on Moving Average (KF-MA) is put forward, that is, to find out the Markov properties of the SCATS data by Moving Average model at first then model the data by Kalman Filtering. Taking the SCATS data in Hang-Zhou as an example, the KF-MA model is compared with the Time Serial (TS) model, the Time Serial model based on Moving Average (TS-MA) and the KF model. The result shows that KF-MA model can maximum elevate the computational efficiency by reducing parameters need to solved in the traditional KF models. Moreover, compared with other models, KF-MA model has better predicted accuracy.

Traffic Flow Prediction with Genetic Network Programming (GNP)

In this paper, a method for traffic flow prediction has been proposed to obtain prediction rules from the past traffic data using Genetic Network Programming (GNP). GNP is an evolutionary approach which can evolve itself and find the optimal solutions. It has been clarified that GNP works well especially in dynamic environments since GNP is consisted of directed graph structures, creates quite compact programs and has an implicit memory function. In this paper, GNP is applied to create a traffic flow prediction model. And we proposed the spatial adjacency model for the prediction and two kinds of models forN-step prediction. Additionally, the adaptive penalty functions are adopted for the fitness function in order to alleviate the infeasible solutions containing loops in the training process. Furthermore, the sharing function is also used to avoid the premature convergence.

Dynamic Equilibrium Model of Expressway Toll Collector Based on Traffic Flow Prediction

Based on the excellent characteristics of neural network such as favorable adaptability to incomplete information and of genetic algorithm with relatively strong and comprehensive search ability, these two objectives are combined to predict the short-term traffic flow of expressway. In this paper, the methods of elite's choices, fitness assignment by proportion and by sort are combined, as well as self-adaptive crossing and mutation probability are used to improve genetic algorithm. To enhance the comprehensive optimal search speed and develop the prediction model for expressway short-term traffic flow, self-adaptive learning rates are introduced to improve BP algorithm and new combing methods are presented to get new species. Meanwhile, queuing theory is adopted to simulate toll process of expressways, and then queuing model of expressways' toll stations is formulated. By combining short-term traffic flow prediction model and toll station queuing model, the demands of toll collectors are predicted according to the allocation of lanes and toll collectors, which realizes the dynamic optimal equilibrium of toll collectors. The validity of the model is finally testified by field tests.

Short-Time Traffic Flow Prediction Based on Chaos Time Series Theory

Traffic flow prediction has become a kernel study in intelligent transportation system. A prediction model of short-time traffic flow is presented based on chaotic time series analysis method. After the phase space reconstruction using traffic flow data, a two-step optimized selection method is proposed, which considers Euclidean distance and equal coefficient between neighboring point and predicted point. Then the prediction model is educed with the local polynomial method to approximate the neighboring points. The model proposed in this paper is applied to predict the real traffic flow in Dongjiang road, Dongguan city in Guangdong province, China. Comparing the predicted traffic flow value with the flow measured in reality, the results show that the maximal relative error is 0.445%, whereas, the minimal one is 0.038%. Moreover, the single-step forward prediction only requires 38.52 seconds. As a result, it is proved that the method can significantly improve the prediction accuracy and meet the requirement of the real-time prediction.