Recurrence and decay properties of a star-typed queueing model with refusal

Abstract

We consider a multiclass service system with refusal and bulk-arrival. The properties regarding recurrence, ergodicity, and decay properties of such model are discussed. The explicit criteria regarding recurrence and ergodicity are obtained. The stationary distribution is given in the ergodic case. Then, the exact value of the decay parameter, denoted by λ_E, is obtained in the transient case. The criteria for the λ_E-recurrence are also obtained. Finally, the corresponding λ_E-invariant vector/measure is considered.

Keywords

Generation function / bulk-arrival queue / recurrence / ergodicity / decay parameter / invariant measure

References

[1] Anderson W. Continuous-Time Markov Chains: An Applications-Oriented Approach. New York: Springer-Verlag, 1991

[2] Chen Anyue, Li Junping, Hou Zhenting, Wang Ng Kai. Decay properties and quasistationary distributions for stopped Markovian bulk-arrival and bulk-service queues. Queueing Syst, 2010, 66: 275-311

[3] Chen Mufa. From Markov Chains to Non-Equilibrium Particle Systems. Singapore: World Scientific, 1992

[4] Darroch J N, Seneta E. On quasi-stationary distributions in absorbing continuous-time finite Markov chains. J Appl Prob, 1967, 245: 192-196

[5] Flaspohler D C. Quasi-stationary distributions for absorbing continuous-time denumerable Markov chains. Ann Inst Statist Math, 1974, 26: 351-356

[6] Kelly F P. Invariant measures and the generator. In: Kingman J F C, Reuter G E, eds. Probability, Statistics and Analysis. London Math Soc Lecture Note Ser, Vol 79. Cambridge: Cambridge Univ Press, 1983, 143-160

[7] Kijima M. Quasi-limiting distributions of Markov chains that are skip-free to the left in continuous-time. J Appl Probab, 1993, 30: 509-517

[8] Kingman J F C. The exponential decay of Markov transition probability. Proc London Math Soc, 1963, 13: 337-358

[9] Li Junping, Chen Anyue. Decay property of stopped Markovian bulk-arriving queues. Adv Appl Probab, 2008, 40(1): 95-121

[10] Li Junping, Chen Anyue. The decay parameter and invariant measures for Markovian bulk-arrival queues with control at idle time. Methodol Comput Appl Probab, 2011,

[11] Nair M G, Pollett P K. On the relationship between μ-invariant measures and quasistationary distributions for continuous-time Markov chains. Adv Appl Probab, 1993, 25: 82-102

[12] Pollett P K. Reversibility, invariance and mu-invariance. Adv Appl Probab, 1988, 20: 600-621

[13] Pollett P K. The determination of quasi-instationary distribution directly from the transition rates of an absorbing Markov chain. Math Comput Modelling, 1995, 22: 279-287

[14] Pollett P K. Quasi-stationary distributions for continuous time Markov chains when absorption is not certain. J Appl Probab, 1999, 36: 268-272

[15] Tweedie R L. Some ergodic properties of the Feller minimal process. Quart J Math Oxford, 1974, 25(2): 485-493

[16] Van Doorn E A. Conditions for exponential ergodicity and bounds for the decay parameter of a birth-death process. Adv Appl Probab, 1985, 17: 514-530

[17] Van Doorn E A. Quasi-stationary distributions and convergence to quasi-stationarity of birth-death processes. Adv Appl Probab, 1991, 23: 683-700

[18] Vere-Jones D. Geometric ergodicity in denumerable Markov chains. Quart J Math Oxford, 1962, 13(2): 7-28

[19] Yang Xiangqun. The Construction Theory of Denumerable Markov Processes. New York: Wiley, 1990

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg