Learning with Projection Effects in Service Operations Systems
Qiao-Chu He, Ying-Ju Chen, Rhonda Righter
Production and Operations Management · 2020 · Volume 29 · Issue 1
Abstract
In this paper, we study customers’ learning behaviors in service operations systems, when the customers hold incorrect beliefs about the population distribution (“projection effects”). We propose a basic model wherein the customers are heterogeneous in both delay sensitivity and awareness of service quality, which explains “rational hesitations”: The uninformed and patient customers obfuscate the quality signaled by the queue length for the uninformed and impatient customers, who wait and join when the queue becomes longer. Ironically, with such bounded rationality, the customers who are more averse to waiting will react more sensitively to the observed long queue, which leads to an overestimation of the service quality and waiting in the long queue. Such bounded rationality also impedes effective learning by inducing decision errors, which, among other consequences, reduces the social welfare due to a blind “buying frenzy” even if the service quality is low.
Keywords
service operations, learning, projection effects, system performance
Enable JavaScript for citation tools and additional paper interactions.