Analysis Of Customer Balking Strategies For Queues Under Different Information Levels

Traditionally, the analysis of queueing system is based on stochastic process. We often derive some performance measures such as queue length, sojourn time distribu-tion and the cycle of busy period. But in the last few decades, there is a growing popularity that researchers investigate queueing systems from an economic viewpoint. In such models, customers decide whether or not to join the queue based on whether they can maximize their benifit, according to a "reward-cost" structure. Unavoidably, every customer’s decision may be affected by others, so it may eventually lead to an equilibrium among customers. In this paper, our key problem is to find customer’s equilibrium balking strategy and socially optimal balking strategy.Firstly, we introduce the acceptance sets, in-cooperative game, game theory in queueing systems and the concept of Nash Equilibrium.Then, we have done some concrete works. First, we considered a queue model with setup and vacation times, and obtained the customer’s equilibrium balking strategy and socially optimal balking strategy in the observable and unobservable cases respectively. Second, we studied a queue model with vacation times and N policy, specially. We gave the optimal control of the system size related to N. Third, we investigated an N policy and stochastic clearing queue, and derived customer’s balking strategy under the observable case.Finally, we presented some numerical examples.