A Study of Hybrid Neural Network Approaches and the Effects of Missing Data on Traffic Forecasting

Abstract

In this paper we present an application of hybrid neural network approaches and an assessment of the effects of missing data on motorway traffic flow forecasting. Two hybrid approaches are developed using a Self-Organising Map (SOM) to initially classify traffic into different states. The first hybrid approach includes four Auto-Regressive Integrated Moving Average (ARIMA) models, whilst the second uses two Multi-Layer Perception (MLP) models. It was found that the SOM/ARIMA hybrid approach out-performs all individual ARIMA models, whilst the SOM/MLP hybrid approach achieves superior forecasting performance to all models used in this study, including three naïve models. The effects of different proportions of missing data on Neural Network (NN) performance when forecasting traffic flow are assessed and several initial substitution options to replace missing data are discussed. Overall, it is shown that ARIMA models are more sensitive to the percentage of missing data than neural networks in this context.

How to Provide Accurate and Robust Traffic Forecasts Practically?

“…Over the past decade, another nonparametric technique, artificial neural networks (ANNs) have been applied in traffic forecasting because of their strong ability to capture the indeterministic and complex nonlinearity of time series (Smith & Demetsky, 1994Chang & Su, 1995;Dougherty & Cobbet, 1997;Lam & Xu, 2000;Park et al, 1999;Dharia & Adeli, 2003;Wei et al, 2009;Wei & Lee 2007;Lee, 2009). Motivated by the universal approximation property, neural network models ranging from purely static to highly dynamic structures include the multilayer perceptrons (MLPs) (Clark et al, 1993;Vythoulkas, 1993;Lee & Fambro, 1999;Gilmore & Abe, 1995;Ledoux, 1997;Innamaa, 2000;Florio & Mussone, 1996;Yun et al, 1998;Zhang, 2000;Chen et al, 2001), the radial basis function (RBF) ANNs (Lyons et al, 1996;Park & Rilett, 1998;Chen et al, 2001), the time-delayed ANNs (Lingras et al, 2000;Lingras & Mountford, 2001;Yun et al, 1998;Yasdi 1999;Abdulhai et al, 1999;Dia, 2001;Ishak & Alecsandru, 2003), the recurrent ANNs (Dia, 2001;Van Lint et al, 2002, and the hybrid ANNs (Abdulhai et al, 1999;Chen et al, 2001;Lingras & Mountford, 2001;Park, 2002;Yin et al, 2002;Vlahogianni ...…”

Forecasting urban traffic flow by SVR with continuous ACO

“…As mentioned above that rapid variational process changes underlying inter-urban traffic flow is complicated to be captured by a single linear statistical algorithm, the artificial neural networks (ANN) models, being able to approximate any degree of complexity and without prior knowledge of problem solving, have received much attention and been considered as alternatives for traffic flow forecasting models [14,[18][19][20][21][22][23]. ANN is based on emulating the processing of the human neurological system to determine related numbers of vehicle and temporal characteristics from the historical traffic flow patterns, especially for nonlinear and dynamic evolutions.…”

Application of seasonal SVR with chaotic immune algorithm in traffic flow forecasting

“…As mentioned above that the process underlying interurban traffic flow is complicated to be captured by a single linear statistical algorithm, the artificial neural networks (ANN) models, able to approximate any degree of complexity and without prior knowledge of problem solving, have received much attention and been considered as alternatives for traffic flow forecasting models [14,[18][19][20][21][22][23]. ANN is based on a model of emulating the processing of the human neurological system to determine related numbers of vehicle and temporal characteristics from the historical traffic flow patterns, especially for nonlinear and dynamic evolutions.…”