Link Travel Time Prediction Based on O-D Matrix and Neural Networks
Name
Ayobami Adewale
Abstract
In public transportation system, commuters are often interested in getting accurate travel time information regarding trips in the future in order to plan their future schedules effectively. However, this information is often difficult to predict due to the irregularities in travel time which are caused by factors like future weather conditions, road accidents and fluctuations in traffic demand. With the introduction of Intelligent Transportation System into public transport system, it has been easy to collect data regarding bus trips such as travel times data. The data collected can be used to make predictions regarding trips in the future by applying scientific methods like Kalman filter, machine learning, and deep learning neural network.
The goal of this thesis is to develop a neural network model for predicting travel time information of a busy route using Origin-Destination matrix derived from a historical GPS dataset of the same route. The prediction accuracy of the NN model developed in this thesis was measured using Root Mean Square Error (RMSE). Analysis of the result showed that the model is sufficient for making predictions of travel time for trips in the future.
The goal of this thesis is to develop a neural network model for predicting travel time information of a busy route using Origin-Destination matrix derived from a historical GPS dataset of the same route. The prediction accuracy of the NN model developed in this thesis was measured using Root Mean Square Error (RMSE). Analysis of the result showed that the model is sufficient for making predictions of travel time for trips in the future.
Graduation Thesis language
English
Graduation Thesis type
Master - Computer Science
Supervisor(s)
Amnir Hadachi, PhD.
Defence year
2018