Shared-ride operations with multiple service types: Premium vs economy, standard vs large robotaxis

Ride-hailing apps offer services with different vehicle sizes and service quality. This paper investigates the impacts of such differentiated supply-side offerings on the operations of a robotaxi/shared autonomous vehicle (SAV) fleet. A nested service-choice model was implemented across three fleet size scenarios using the POLARIS agent-based travel demand model for Bloomington, Illinois. The study compares fleet performance for a scenario with diverse service offerings (standard 4-seaters and larger 6-seaters, in both economy and premium styles) and one with just 4-seater economy service. The demand for 4-seater economy service peaks during commute hours (6-9 a.m.), while premium services, especially 6-seater premium SAVs, show negligible demand due to their higher fares and niche market. Comparisons across three fleet sizes indicate that diverse SAV services lead to operational inefficiencies, including higher wait times, lower profits, and a 13-percentage-point increase in empty vehicle-miles traveled (from 24.2% to 37.4%). Premium and 6-seater SAVs struggled to achieve profitability due to low demand and higher operating costs. Across all three fleet sizes, multi-service fleets incurred higher costs and earned substantially lower profits per SAV, with profit per SAV falling by over 80%, on average, when diverse services are offered instead of a single, standard service.