Ride-hailing services like Designated Driver Service (DDS) offer reservation options, allowing users to book rides at their desired times. To improve the convenience for reservation service customers and ensure the highest number of completed rides, it’s essential to maximize the number of available drivers for assignments. To reach this objective, we predict the conversion probability - the likelihood of customers proceeding after seeing the price - and the assignment probability, which is the likelihood of drivers accepting the reservation request. Based on these two probabilities, we aim to identify a pricing strategy that maximizes both probabilities and propose this price within the service. To accomplish this, we leverage the conversion and assignment probabilities to establish the expected actual ride count as the objective function. We optimize the pricing in the direction that maximizes this anticipated ride count, and the resultant optimal price is integrated into the reservation service. This study presents an innovative method employing Bayesian Optimization(BO) to establish a mutually advantageous pricing scheme for both customers and drivers. This contribution enhances the investigation into effective pricing strategies within the context of reservation-based vehicle dispatch services.

The main contribution of this paper is to present a straightforward yet effective historical input selection method for real-time traffic forecasting. Historical inputs are crucial elements in traffic forecasting because they help models recognize historical traffic patterns. However, selecting appropriate historical inputs becomes challenging, especially on irregular holidays with abnormal traffic patterns.As a solution, we suggest a method with three clear strengths. First, it is simple enough to employ in real-time computation. Second, our method provides more sophisticated features than existing studies which roughly categorize and match similar days. Finally, our method outperforms existing methods remarkably, given the unusual traffic patterns on special holidays.Our experiments verified the outperformance through an industrial prediction model and traffic dataset from a leading mobility service provider in South Korea.