Ride-hailing services are becoming increasingly popular, and their influence on overall transportation network performance is consequential. Their convenience and flexibility attract many users, but their impact on traffic and their interference with public transit usage alarmed network regulators. Incentive-based measures, such as encouraging trip-pooling through ride-splitting platforms, potentially mitigate this impact. They pave the way for serving the same demand level with a smaller fleet size and lower overall Vehicle Kilometers Traveled. In this thesis, we aim to design and evaluate policies that incentivize trip-pooling to reduce multi-modal network delays. Moreover, we verify whether the trip-pooling advantages are reproducible for on-demand micro-transit services with relatively high vehicle capacity.
To ensure that the various transportation alternatives and the spatial network resources are used efficiently, we propose, in this work, an occupancy- and modal-dependent allocation policy where pool ride-hailing trips are allowed in originally dedicated and underutilized bus lanes. By resorting to Macroscopic Fundamental Diagrams for network supply modeling, we analyze the existence of the multi-modal user equilibrium. Additionally, we analyze the solo-pool demand split that minimizes the total Passenger Hours Traveled under a given network configuration to derive maximum efficiency and ensure minimum disturbances to bus operations. Numerical examples show that the optimal point that minimizes delays for multi-modal transport users occurs when only a fraction of the pooling vehicles use the dedicated bus lanes. Additionally, we define the ride-hailing user equilibrium under the proposed strategy and suggest a pricing scheme for mitigating the gap between total user delays of the system optimum and user equilibrium solutions. When the multi-modal demand is time-dependent, the optimal demand split between solo and pool trips that minimizes total network delays becomes dynamic under the proposed allocation policy. We develop dynamic feedback-based control frameworks that regulate pool vehicles' access to bus lanes by adjusting the fare gap between solo and pool trips. We precisely design various controller types to ensure minimal disturbances to bus operations, and to minimize the Total Time Spent for multi-modal travelers. The results mark the possibility of improving the overall network conditions by adjusting, through fare control, the number of pool vehicles in bus lanes. While the modeling, operation, and benefits of two-passenger ride-splitting services are well established, little is known about the macroscopic user performance and efficiency of high-capacity on-demand services and their potential to replace public transportation. Therefore, we focus in this work on developing a macroscopic comparison scheme between on-demand flexible-route and flexible-schedule micro-transit services and fixed-route and fixed-schedule public transit. By developing a simulation environment for each service, we aim to examine the replaceability potential between the two under different graph structures and various spatial demand distributions and intensities. By macroscopically examining the user and efficiency-related metrics, we infer for which network settings one service prevails over the other. The results show that the difference in aggregate waiting times is insignificant despite micro-transit trips yielding lower trip times.