Balancing Supply with Demand in Ride-hailing Platforms: An Operational Approach

Abstract

Ride-hailing platforms such as Uber and Didi are concerned about supply-demand imbalance, as during peak periods, demand from passengers far exceeds supply from drivers, whereas during off-peak periods, the supply of drivers is plenty but the passenger demand can be weak. Although surge pricing can theoretically solve the issue, it is heavily criticized for exploiting passengers unfairly and hurting drivers' rating scores. We propose two novel operational schemes for shifting the supply during off-peak periods to meet the demand during peak periods. Specifically, under the qualification scheme, only drivers who work up to a peak-period target are eligible to serve off-peak demand, and under the prioritization scheme, drivers who work up to a peak-period target are given priority to serve off-peak passengers. Building on a model that captures the supply-demand imbalance and allows drivers' strategic work time allocation, we show that both schemes increase the transaction volume. In addition, if adopting an open system under which part-time drivers can provide service, a platform prefers the prioritization scheme. In contrast, for platforms that only allow full-time drivers to provide service, the qualification scheme is more desirable. We further endogenize a platform's decision on system adoption and show that the combination of an open system and a prioritization scheme is the most effective in solving the supply-demand imbalance. We also consider a varying commission rate policy in which the platform can adopt a lower commission rate during peak periods. Under certain conditions, the performance of operational incentive schemes outperforms the policy that involves the platform lowering the peak-period commission rate. Our results have general implications for on-demand platforms (e.g., food delivery platforms and freelance platforms) in resolving the challenge of supply-demand mismatch.