IORA Seminar on "Coordinating Supply and Demand on an On-Demand Service Platform with Impatient Customers"

 

Dr. Tang is University Distinguished Professor and the holder of the Edward W. Carter Chair in Business Administration at the UCLA Anderson School. Chris has published 6 books and over 130 research articles in various leading academic journals, and written articles for Wall Street Journal, Financial Times, and The Guardian. He is the recipient of numerous teaching and research awards including UCLA (university‐wide) Distinguished Teaching Award in 2012. Lifetime fellows of Institute of Operations and Management Sciences (INFORMS), Production and Operations Management Society (POMS), and Manufacturing & Service Operations Management Society (M&SOM). He served as President of Production and Operations Management Society (POMS) in 2014, and now serves as Editor of Springer’s Supply Chain Management Series, and as Editor‐in‐Chief of Manufacturing & Service Operations Management (M&SOM). Chris received his B.Sc. (First class honours) from King’s College, University of London, M.A., M.Phil, and PhD from Yale University.

 

 

Coordinating Supply and Demand on an On-Demand Service Platform with Impatient Customers

Consider a situation when an on‐demand service platform uses earnings‐sensitive independent providers with heterogeneous reservation price (for work participation) to serve its waiting time and price sensitive customers with heterogeneous valuation of the service. As such, both the supply and demand are ``endogenously'' dependent on the price the platform charges its customers and the wage the platform pays its independent providers. In this paper, we present a queueing model with endogenous supply (number of participating agents) and endogenous demand (customer request rate) to model this on‐demand service platform. To coordinate endogenous demand with endogenous supply, we use the steady state performance in equilibrium to characterize the optimal price and wage rates that maximize the profit of the platform (as well as the total welfare).

We first analyze a base model that uses a fixed payout ratio (i.e., the ratio of wage over price). We then extend our model to allow the platform to adopt a dynamic payout ratio. Due to the fact that the exact analysis based on an M/M/k system is intractable, we develop an approximation scheme to generate some analytical results. In addition to the fact that our approximation scheme performs well, we find that it is optimal for the platform to charge a higher price, pay a higher wage, and offer a higher payout ratio when the potential customer demand increases. Furthermore, when customers become more time‐sensitive, the platform should also pay a higher wage and offer a higher payout ratio, but the price rate is not necessarily monotone. We used a set of actual data from a large on‐demand ride‐sharing to illustrate some of our main insights. This is a joint work with Jiaru Bai and Kut C. So of UC Irvine, Xiqun Chen of Zhejiang University, and Hai Wang of Singapore Management University.