Paccar Hall, Room 393

Seminar Speaker: Yong-Pin Zhou
Affiliation: University of Washington
Area: Operations Management

Name of Presentation: Social Loafing and Queue Driven Speedup: Evidence from a Supermarket

Abstract:
We study factors affecting server's service time using data from a supermarket. We first develop an analytical model to compare the server's optimal service rates in comparable single-server and two-server configurations, and derive a theoretical prediction that pooling has a negative impact on cashiers' service rate due to the social loafing effect. We also theoretically show that the queue pressure cost leads the average service time to be a convex decreasing function of the queue length. Using the supermarket's unique checkout layout that acts as a natural experiment, and a unique set of checkout transaction details, as well as video recordings of the queue information, collected from the supermarket, we test our theoretical predictions and find empirical evidence supporting our hypotheses built on theory. Specifically, servers in single-server queues are estimated to be 5.29% faster than those in two-server queues. In addition, the queue speedup effect is estimated to be about 5.07% on average for each additional waiting customer. And the marginal speedup effect is diminishing as the queue length increases. These findings have important implications on managerial decisions such as queue configurations and staffing.