Travel time is an important measure for transportation system performance evaluation. In particular, it essential input to the advanced traveler information systems and route guidance applications that requires reliable traffic in real time. Therefore, numerous studies have been conducted predict segment corridor travel times on basis of data from loop detectors. Focusing freeway travel-time prediction incorporates effect propagation, this article presents a dynamic model using measurements...

Short‐term traffic flow forecasting has been regarded as essential for intelligent transportation systems, including both point prediction and interval prediction. Compared with prediction, of in the future will be critical managers to make reasonable decisions. This study applies fuzzy information granulation method obtain dispersion range collected time series, classical approaches K ‐nearest neighbours, back‐propagation neural network, support vector regression are applied on original...

Over the past decade, traffic heteroscedasticity has been investigated with primary purpose of generating prediction intervals around point forecasts constructed usually by short-term condition level forecasting models. However, despite considerable advancements, complete patterns, in particular seasonal effect, have not adequately handled. Recently, an offline adjustment factor plus GARCH model was proposed Shi <i xmlns:mml="http://www.w3.org/1998/Math/MathML"...

In this paper, we propose a load-aware and QoS- aware user association strategy that jointly considers the load of each BS user's achievable rate instead only utilizing latter, formulate it as network-wide weighted utility maximization problem to reveal how heterogeneous cellular network should self-organize. This is nonlinear mixed-integer optimization problem, its optimum solutions are very difficult be obtained when large scale one. To solve proposed relax indicator variables adopt...

Heteroscedasticity modeling in transportation engineering is primarily conducted short-term traffic condition forecasting to generate time varying prediction intervals around the point forecasts through quantitatively predicting conditional variance of series. Until recently, generalized autoregressive heteroscedasticity (GARCH) model and stochastic volatility have been two major approaches adopted from field financial series analysis for modeling. In this paper, recognizing pronounced...

It is critical to implement accurate short-term traffic forecasting in management and control applications. This paper proposes a hybrid method based on neural networks combined with the K-nearest neighbor (K-NN) for flow forecasting. The procedure of training network model using existing input-output data, i.e., indispensable fine-tuning prediction model. Based this point, K-NN was employed reconstruct data models while considering similarity patterns. done through collecting specific state...

Uncertainty quantification is important for making reliable transportation decisions. For grey-based uncertainty approaches, the data classification methods most models cannot yield real-time upper and lower limited sequence, limiting their application in dynamic systems. Therefore, this paper proposes an adaptive grey prediction interval model to quantify traffic condition uncertainty. To end, polynomial regression first used fit flow trend function real time, generating dynamically...

Because of the growing awareness importance traffic condition uncertainty-related studies, uncertainty modeling is gaining increasing attention from transportation research community. In this field, uncertainty, gauged mainly by conditional variance characteristics, has been investigated primarily with two major approaches, generalized autoregressive heteroscedasticity approach and stochastic volatility approach; however, both lack a thorough sound test on applicability these approaches. To...

Cyber-physical system (CPS) is a next-generation intelligent integrating computing, communication, and control. As unit of computing process physical process, CPS new research field, where cooperative adaptive cruise control (CACC) not only microcosm but also prerequisite for unmanned systems. CACC relies on the wireless communication network technology to achieve platooning through vehicle-to-vehicle collaborative methods. It can improve traffic efficiency ensure safe driving. Once...

Dynamic traffic demand is crucial for developing effective strategies and algorithms real‐time management control. The uncertainty of provides additional information while its prediction very complicated inadequately investigated in the existing literature. Recently, radio‐frequency identification (RFID) technology has been deployed to monitor condition traffic. In this study, authors propose a modelling system predict dynamic using RFID data link volume. includes an optimisation model...

Emerging technologies of connected and automated vehicles (CAVs) applied at intersections have great potential to improve traffic efficiency, driving safety, fuel economy. This paper proposes a virtual spring strategy for coordinating CAVs pass through signal-free intersections. A coordination system (VSCS) is established force the movements conflicting in terms characteristics. Inspired by properties springs with dampers, distributed control protocol designed regulate reach desired state...

Short-term traffic flow forecasting has an essential role in advanced traveler information systems, route guidance and proactive signal control systems. Numerous univariate multivariate models have been presented on both level variance forecasting. However, few studies incorporated the relationship between parameters (such as volume speed) into development of model. It is well known that there are inherent relationships heteroscedasticity exists series. On this basis a vector autoregressive...