Abstract:

Service providers often adopt customer segmentation strategies in the presence of heterogeneous customers （hold different waiting costs）. However， when the service operates as a queuing system， network externality among all the customers makes segmentation tricky： shorter wait time provided to one customer type must come at the expense of others， as segmentation redistributes waiting time from higher- to lower-priority customers. Moreover， when customers can compare wait time received by different customer types， customers receiving a worse wait time than expected， due to having a lower service priority， may experience a fairness perception effect that may cause customers to transfer， switch， or balk within the service system， then affect the service provider’s revenue. The first exact analysis of a service provider offering two classes of non-preemptive priority service is provided， which considers both customer perception of fairness and the heterogeneous of service values.First， customers’ fairness perception is modelled as a negative utility （i.e.，\alpha (W_{\mathrm{2}}-W_{\mathrm{1}})） on the regular customer proportional to the waiting time difference between the two queues （i.e.，W_{\mathrm{2}}-W_{\mathrm{1}}） and the parameter of fairness （i.e.，\alpha）. Then， the perspective of revenue and a non-monopoly system in which customers can leave the system freely are considered. The findings indicate that whether a service provider should use customer segmentation depends on the value of regular customers and customer’s fairness perception. At first， the service provider should adopt two queues and keep the priority queue and regular queue in relative proximity to present the advantages of shorter waiting times in priority services when the value of regular customers is small， and fairness is weak. With an increasing fairness preference， however， the service provider should maintain only priority service. Considering customer's fairness， it is found that if the value of regular customers is large， a profit-maximizing service provider should cancel the priority fee and induce all the customers to enter priority or keep a regular queue only. Finally， the properties of optimal results are analyzed with the existing results， and then management insight is proposed.The findings are verified by numerical simulation， and this research focuses on the scientific mechanism of classification service and the corresponding optimal service pricing， providing a more practical motivation. The findings confirm the extant research on customer segmentation and the benefits of offering differentiated service and pricing， while also challenging some commonly accepted notions and practices.

Key words: service operation, queuing, non-monopoly system, heterogeneity, fairness, revenue management