No description available.

Providing transportation system operators and travelers with accurate travel time information allows them to make more informed decisions, yielding benefits for individual travelers and for the entire transportation system. Most existing advanced traveler information systems (ATIS) and advanced traffic management systems (ATMS) use instantaneous travel time values estimated based on the current measurements, assuming that traffic conditions remain constant in the near future. For more effective applications, it has been proposed that ATIS and ATMS should use travel times predicted for short-term future conditions rather than instantaneous travel times measured or estimated for current conditions. This dissertation research investigates short-term freeway travel time prediction using Dynamic Neural Networks (DNN) based on traffic detector data collected by radar traffic detectors installed along a freeway corridor. DNN comprises a class of neural networks that are particularly suitable for predicting variables like travel time, but has not been adequately investigated for this purpose. Before this investigation, it was necessary to identifying methods for data imputation to account for missing data usually encountered when collecting data using traffic detectors. It was also necessary to identify a method to estimate the travel time on the freeway corridor based on data collected using point traffic detectors. A new travel time estimation method referred to as the Piecewise Constant Acceleration Based (PCAB) method was developed and compared with other methods reported in the literatures. The results show that one of the simple travel time estimation methods (the average speed method) can work as well as the PCAB method, and both of them out-perform other methods. This study also compared the travel time prediction performance of three different DNN topologies with different memory setups. The results show that one DNN topology (the time-delay neural networks) out-performs the other two DNN topologies for the investigated prediction problem. This topology also performs slightly better than the simple multilayer perceptron (MLP) neural network topology that has been used in a number of previous studies for travel time prediction.

This thesis enhances rare literature focusing on modeling and forecasting of term structure of crude oil markets. Using dynamic Nelson-Siegel model, crude oil term structure is decomposed to three latent factors, which are further forecasted using both parametric and dynamic neural network approaches. In-sample fit using Nelson-Siegel model brings encouraging results and proves its applicability on crude oil futures prices. Forecasts obtained by focused time-delay neural network are in general more accurate than other benchmark models. Moreover, forecast error is decreasing with increasing time to maturity.

Neste trabalho, redes neurais dinÃmicas sÃo avaliadas como modelos nÃo-lineares eficientes para prediÃÃo de sÃries temporais complexas. Entre as arquiteturas avaliadas estÃo as redes FTDNN, Elman e NARX. A capacidade preditiva destas redes sÃo testadas em tarefas de prediÃÃo de um-passo-adiante e mÃltiplos-passos-adiante. Para este fim, sÃo usadas as seguintes sÃries temporais: sÃrie laser caÃtico, sÃrie caÃtica Mackey-Glass, alÃm de sÃries de trÃfego de rede de computadores com caracterÃsticas auto-similares. O uso da rede NARX em prediÃÃo de sÃries temporais à uma contribuiÃÃo desta dissertaÃÃo. Esta rede possui uma arquitetura neural recorrente usada originalmente para identificaÃÃo entrada-saÃda de sistemas nÃo-lineares. A entrada da rede NARX à formada por duas janelas deslizantes (sliding time window), uma que desliza sobre o sinal de entrada e outra que desliza sobre sinal de saÃda. Quando aplicada para prediÃÃo caÃtica de sÃries temporais, a rede NARX à projetada geralmente como um modelo autoregressivo nÃolinear (NAR), eliminando a janela de atraso da saÃda. Neste trabalho, à proposta uma estratÃgia simples, porÃm eficiente, para permitir que a rede NARX explore inteiramente as janelas de tempo da entrada e da saÃda, a fim de melhorar sua capacidade preditiva. Os resultados obtidos mostram que a abordagem proposta tem desempenho superior ao desempenho apresentado por preditores baseados nas redes FTDNN e Elman.