Abstract:Short-term traffic speed forecasting is an important component of Intelligent Transportation Systems (ITS). Multi-step-ahead prediction can provide more information and predict the longer trend of traffic speed than single-step-ahead prediction. This paper presents a multi-step-ahead traffic speed prediction approach by improving the gradient boosting regression tree (GBRT). The traditional multiple output strategies, e.g., the direct strategy and iterated strategy, share a common feature that they model the samples through multi-input single-output mapping rather than multi-input multi-output mapping. This paper proposes multivariate GBRT to realize simultaneous multiple outputs by considering correlations of the outputs which have not been fully considered in the existing strategies. For illustrative purposes, traffic detection data are extracted at the 5-min aggregation time interval from three loop detectors in US101-N freeway through the Performance Measurement System (PeMS). The support vector regression (SVR) is used as the benchmark. Assessments on the three models are based on the three criteria, i.e., prediction accuracy, prediction stability, and prediction time. The results indicate that (I) Multivariate GBRT and GBRT using the direct strategy have higher prediction accuracies compared with SVR; (II) GBRT using the iterated strategy has a good prediction accuracy in short-step-ahead prediction and the prediction accuracy decreases significantly in long-step-ahead prediction; (III) Multivariate GBRT has the best stability which means the higher reliability in multi-step-ahead prediction while iterated GBRT has the worst stability; and (IV) Multivariate GBRT has an enormous advantage in the prediction efficiency and this advantage will expand with the increasing prediction horizons.

Accelerating the gradient projection algorithm for solving the non-additive traffic equilibrium problem with the Barzilai-Borwein step size

A faster path-based algorithm with Barzilai-Borwein step size for solving stochastic traffic equilibrium models

A Distributed Computing Method Integrating Improved Gradient Projection for Solving Stochastic Traffic Equilibrium Problem

Freeway Network Simulation and Dynamic Traffic Assignment with Metanet Tools

AGNP: Network-wide Short-Term Probabilistic Traffic Speed Prediction and Imputation

Multi-step-ahead Traffic Speed Forecasting Using Multi-Output Gradient Boosting Regression Tree

Elastic-demand bi-criteria traffic assignment under the continuously distributed value of time: A two-stage gradient projection algorithm with graphical interpretations

Reduced gradient algorithm for user equilibrium traffic assignment problem

Fast Traffic Engineering by Gradient Descent with Learned Differentiable Routing

Accelerating Gradient Tracking with Periodic Global Averaging

Accelerating Strategies and Computational Studies of the Frank-Wolfe Algorithm for the Traffic Assignment Problem

Application of genetic algorithm in estimating origin-destination matrices of traffic network

Finding Equilibria in the Traffic Assignment Problem with Primal-Dual Gradient Methods for Stable Dynamics Model and Beckmann Model

On the acceleration of the Barzilai-Borwein method

Projected Gradient Method Combined with Homotopy Techniques for Volume-Measure-Preserving Optimal Mass Transportation Problems

An Inertial-type CG Projection Method with Restart for Pseudo-monotone Costs with Application to Traffic Assignment

GPTE - Efficient Graph Partitioning-based Traffic Engineering in Hybrid SDN/IP Networks.

User equilibrium traffic assignment: k paths subtracting-adding algorithm

Efficient Euclidean projections via Piecewise Root Finding and its application in gradient projection

Traffic State Estimation from Vehicle Trajectories with Anisotropic Gaussian Processes

Applications of Genetic Algorithms in BGP-Based Interdomain Traffic Engineering