Mobility On Demand: What About the Weekend?

Abstract

Mobility on demand (MoD) services like ride-hailing, ride-sharing, and car-sharing are changing travel behavior by providing increased options and flexibility. These services can be best understood and planned for through the use of detailed computer simulations. However, existing simulations predominantly focus on modeling average working days, characterized by high and predictable travel demand. This approach overlooks the distinct travel patterns observed during weekends. Unlike weekdays, which feature pronounced peak hours, weekend travel is distributed more evenly throughout the day, particularly on Saturdays. This study compares the differences in travel demand patterns between weekends and weekdays and their possible impact on policies drawn from MoD simulations. Using an agent-based simulation framework MATSim, we simulate the introduction of an autonomous mobility on demand (aMoD) service to Zurich, Switzerland, and its environs. We then compare weekday and weekend travel patterns highlighting unique aspects of weekend travel and their implications for MoD service operations. The findings suggest that transport policies should account for the unique characteristics of weekend travel. The results provide insights into modal shifts, showing how more public transport and private vehicle trips could be replaced by MoD services during weekends, especially for long-distance travel. Furthermore, results show that optimal fleet sizes vary between weekdays and weekends owing to differences in demand. While weekends see higher MoD demand, wait times don't necessarily increase. However, longer detours for pickups may extend travel times. Accounting for weekend travel in simulations helps ensure policy planning supports reliable service while balancing wait times, travel times, occupancy, and operational costs.