Ride-pooling demand prediction: A spatiotemporal assessment in Germany

Ride-pooling has attracted considerable attention from both academia and practitioners in recent years, promising to reduce traffic volumes its negative impacts urban areas. Simulation studies have shown that large-scale ride-pooling the potential increase vehicle utilization, thereby reducing kilometers traveled (VKT) required fleet sizes compared single-passenger mobility options. However, real world, services are rare not yet widely implemented, part due high operating costs expected decrease substantially with advent of automated vehicles. Two fleets operated by MOIA Hamburg Hanover, Germany, serve as testbeds for future pooled services. For this study, we analyze pre-pandemic demand data 2019 2020 perform spatial random forest regressions understand (spatial) characteristics trip origins cities. We then examine how well findings one study area (Hamburg our case) can be generalized transferred other cities (Hanover enable predictions beyond areas an existing service. The regression results similar show strongest impact on ridepooling variables capturing density workplaces, gastronomy culture. Given Hamburg, predict Hanover observe is overestimated all applied models. lag X (SLX) model showed most overall overestimation below 20%.