This report presents the results of a study to develop a model for predicting the speeds of free vehicles in New Zealand (NZ) by developing a model which could evaluate traffic performance on rural two-lane highways. There are a variety of physical factors which influence speeds on two-lane highways. Of these, gradient and curvature have the greatest impact and so received the most attention in this project. A computerised data logger was acquired which recorded the data from up to 16 detectors at once. This made it possible to establish speed profiles of the same vehicle at different points along a section. The speeds of over 300.000 vehicles were measured at 58 sites. The speeds were found to be normally (Gaussian) distributed. After reviewing the characteristics of the NZ vehicle fleet, 15 representative vehicles were selected for use in the modelling. The data were analysed to develop a measure for differentiating between free and bunched vehicles. The results were used to form a model for predicting speeds on two-lane highways. Called SPEEDSIM, the model uses Monte Carlo simulation principles to simulate the travel of a vehicle along a section of highway. A dissertation submitted in partial fulfilment of the requirements for the degree of PhD in Engineering.
Abstract