Two essays on equilibrium analysis of customer's queueing behavior

Abstract

In the past decades, the development of information technology has reshaped how customers obtain product related information, and has revolutionized the ways organizations relate to the marketplace. Meanwhile, progress in behavioral research has deepened our understanding of customer behavior. It shows that customers are not passive receiving end but active decisions makers. Those changes in technology and knowledge have created a new world of opportunities and challenges for all aspects of the enterprise. In this thesis, we research into their impacts on the operations of service facilities. In the first topic, we consider the new challenges associated with the development of information technology. Social media and word-of-mouth forums are shown to have great influence on customers' purchase decisions, and have made the buyer-generated content non-negligible. We consider a typical situation where a service provider serves two types of customers, sophisticated and naive. Sophisticated customers are well-informed about service-related information and make their joining-or-balking decisions strategically, whereas naive customers do not have such information and rely on online rating information to make such decisions. We demonstrate that under certain conditions a service provider can increase its profitability by simply 'dancing' its price, that is, replacing the static pricing strategy with a cyclic high-low pricing strategy. The success of this strategy relies on two key conditions: the potential market size is large enough so that congestion is a key concern in the service system and the rating provides the average price and average utility information. We also show that when customers are loss averse and the rating serves as a reference point for them, the loss aversion behavior dilutes the effectiveness of cyclic pricing strategy. Finally we show that the cyclic pricing strategy is never socially optimal. In the second topic, we incorporate new understandings of customer behavior with their equilibrium queueing decisions in health care service context. We investigate the patient' doctor shopping behavior when they seek diagnostic service. When a patient' belief about her health status is inconsistent with a doctor's diagnosis, cognitive dissonance may arise. The patient may seek more doctoral opinions to mitigate such dissonance without referrals, that is, the patient is engaged in doctor shopping. We derive the patient' optimal "doctor shopping" stopping time by adopting the simple 'one-stage look-ahead' rule and Bayesian updating. We show that a patient stops the doctor shopping process whenever two successive diagnostic results are consistent with each other. Doctor shopping always results in a highly system congestion. Interestingly, we find that the patients' doctor shopping may not necessarily undermine the social welfare. Doctor shopping improves patients' psychological gains. Doctor shopping shall always be tolerated when the accuracy of the diagnosis is not very high. When the accuracy of the diagnosis is high, doctor shopping may be also tolerated if the policy maker cares about the patients' psychological gains.