Optimized Structure of the Traffic Flow Forecasting Model With a Deep Learning Approach

TLDR

Forecasting accuracy is critical for intelligent traffic management, and the advent of big data offers opportunities to improve it, making this the first study to present an optimized deep‑learning traffic flow forecasting model. The study proposes a stacked autoencoder Levenberg‑Marquardt deep‑learning model to enhance traffic flow forecasting accuracy. The model employs a Taguchi‑optimized stacked autoencoder with greedy layer‑wise unsupervised learning, trained on M6 freeway data and benchmarked against three existing predictors. Evaluation shows the optimized model outperforms existing predictors in traffic flow forecasting.

Abstract

Forecasting accuracy is an important issue for successful intelligent traffic management, especially in the domain of traffic efficiency and congestion reduction. The dawning of the big data era brings opportunities to greatly improve prediction accuracy. In this paper, we propose a novel model, stacked autoencoder Levenberg-Marquardt model, which is a type of deep architecture of neural network approach aiming to improve forecasting accuracy. The proposed model is designed using the Taguchi method to develop an optimized structure and to learn traffic flow features through layer-by-layer feature granulation with a greedy layerwise unsupervised learning algorithm. It is applied to real-world data collected from the M6 freeway in the U.K. and is compared with three existing traffic predictors. To the best of our knowledge, this is the first time that an optimized structure of the traffic flow forecasting model with a deep learning approach is presented. The evaluation results demonstrate that the proposed model with an optimized structure has superior performance in traffic flow forecasting.