Short-term Hourly Traffic Forecasts using Hong Kong Annual Traffic Census

Abstract

The need for acquiring the current-year traffic data is a problem for transport planners since such data may not be available for on-going transport studies. A method is proposed in this paper to predict hourly traffic flows up to and into the near future, using historical data collected from the Hong Kong Annual Traffic Census (ATC). Two parametric and two non-parametric models have been employed and evaluated in this study. The results show that the non-parametric models (Non-Parametric Regression (NPR) and Gaussian Maximum Likelihood (GML)) were more promising for predicting hourly traffic flows at the selected ATC station. Further analysis encompassing 87 ATC stations revealed that the NPR is likely to react to unexpected changes more effectively than the GML method, while the GML model performs better under steady traffic flows. Taking into consideration the dynamic nature of the common traffic patterns in Hong Kong and the advantages/disadvantages of the various models, the NPR model is recommended for predicting the hourly traffic flows in that region. Copyright Springer 2006Annual Traffic Census, Auto-Regressive Integrated Moving Average, Gaussian Maximum Likelihood, Neural Network, Non-Parametric Regression,

Triple seasonal methods for short-term electricity demand forecasting

“…The empirical results provide insight for the number of seasonal cycles to include in a method, for the relative importance of the different types of seasonal cycles, and for the choice between the two exponential smoothing methods and ARMA modelling. The study also evaluates the benefit in incorporating a residual autocorrelation term in the exponential smoothing method of Gould et al Research in this area would seem to be timely, given the increasing availability of intraday data in a variety of other applications, such as traffic management and call centre staff scheduling (see, for example, Lam et al, 2006;. Another potential application area for the methods in this paper is the modelling of intraday electricity prices, which is of primary importance for trading electricity.…”

Traffic forecasting using least squares support vector machines

“…The results are obtained from the Transportmetrica 197 average of the historical traffic data at the same time of day and day of the week. Since non-parametric regression generally performs better than its parametric counterparts (Smith and Demetsky 1997, Lam and Xu 2000, Ding et al 2002, Smith et al 2002, Yin et al 2002, Tang et al 2003, Vanajakshi and Rilett 2004, Wu et al 2004, Liu et al 2005, Luo et al 2005, Hong et al 2006, Lam et al 2006a,b, Lee et al 2006, Zhang and Xie 2007, Lam and Toan 2008, Li et al 2008, Manoel et al 2009a, the comparison of different non-parametric forecasting results is investigated in detail after specifying the variable d. In the study, the mean absolute percentage error …”

A novel three-step procedure to forecast the inspection volume

“…Specifically, Artificial Neural Networks (ANNs), an extremely popular class of AI models, have constantly been applied to forecasting and planning tasks in transportation and traffic flows (Smith and Demetsky, 1994;Dougherty, 1995;Amin et al, 1998;Park and Rilett, 1998;Abdelwahab and Sayed, 1999;Faghri et al, 1999;Sayed and Razavi, 2000;Dharia and Adeli, 2003;Sarvareddy et al, 2005;Vlahogianni et al, 2005a;Lam et al, 2006;Celikoglu and Cigizoglu, 2007;Tsai et al, 2009;Gosasang et al, 2011) due to its relatively easy way to approximate a nonlinear mapping with any degree of complexity, overcoming the problem of nonlinearity (Hornik et al, 1989).…”