20191008新加坡管理大學王宇博士講座

發布時間:2019-09-29 


主題:   Strategic Switchers, the Unique Feature of Carpooling

主講人:  Rowan Wang 王宇博士

地點:  旭日樓306

時間:   2019年10月8日上午10:30


報告人簡介:

Rowan Wang is an Assistant Professor of Operations Management in the Lee Kong Chain School of Business at the Singapore Management University (SMU). Rowan's research interests include service and queueing systems; supply chain and inventory management; sharing economy; and healthcare operations. His work has been published in journals such as Manufacturing & Service Operations Management.


報告摘要:

This paper studies the operation of a carpooling system with the unique feature of having the strategic switchers, who can either be riders and take seats; or be drivers and take seats, depending on the real-time imbalance of riders and drivers in the system. Different from most of existing economic systems where the pool of demand and the pool of supply are separated, for the carpooling system that we analyze, the pool of demand and the pool of supply are joint, as the strategic switchers belong to both the pool of demand and the pool of supply simultaneously. When the system is lack of demand, more switchers will join the demand side; while when the system is lack of supply, more switchers will join the supply side. Using an equilibrium model, we show that, the fact of having strategic switchers brings robustness to the system. For example, sudden changes in the arrival rates of riders (e.g., during peak period) or drivers, have less impact on system congestion level, for the system with strategic switchers than for the system without them. This result explains the interesting phenomenon observed in practice that, the time it takes to wait for a ride may not increase significantly even at peak period. We also show that dynamic pricing mechanisms, such as surge pricing, could be less effective, with the existence of strategic switchers. This result supports the static pricing policy that is currently used by leading carpooling platforms in practice.


Baidu
sogou