This study proposes a simulation-based mode choice method for ride sharing services. Conventional mode choice for ride sharing services is a feedback loop based process where multiple simulation runs are expected to achieve an equilibrium state between assumed and simulated service attributes. The achieved equilibrium state is viable for the initial set of supply parameters. This thesis contributes in proposing the use of an analytical ride sharing market equilibrium model to perform the mode choice for simulating ride sharing services which is viable against any change in the supply, reducing the otherwise required regress computational time for running multiple simulations. The proposed methodology calibrates the market equilibrium model parameters using a set of observed service (attribute) data. The observed service data contains the service attributes for a range of fleet sizes serving a range of ride sharing demand. Whereas the analytical market equilibrium model proposed to be used in this study also outputs ride sharing demand and the network attributes detour and waiting time. This is taken as an optimization problem to be solved to calibrate the market equilibrium model parameters so that it outputs the observed service attributes for a given range of fleet sizes. A number of different goodness of fit errors are also utilized within the optimization problem. Two different case studies are used for the setup, first synthetic and then Munich network to calibrate the model parameters. The converged error results from both case study network shows that the analytical model fits to most of the extent with the observed data. Computational time of calibrated model is negligible compared to computationally expensive and regress process of simulation based equilibrium making the calibrated model as an effective alternate to perform mode choice of ride sharing services.     «

This study proposes a simulation-based mode choice method for ride sharing services. Conventional mode choice for ride sharing services is a feedback loop based process where multiple simulation runs are expected to achieve an equilibrium state between assumed and simulated service attributes. The achieved equilibrium state is viable for the initial set of supply parameters. This thesis contributes in proposing the use of an analytical ride sharing market equilibrium model to perform the mode ch...     »