期刊文献+

T-SPH排队队长平稳分布的尾部分析 被引量:1

Tail Analysis of the Stationary Queue Length Distributions for Two T-SPH Queues
下载PDF
导出
摘要 在复杂随机模型的研究中,经常会出现水平和位相都是无限的拟生灭过程,对这类过程,平稳分布的计算仍然是一个未很好解决的难题.然而对一类比较特殊三对角无限位相拟生灭过程,简记为T-QBD过程,文献[1]指出,在一定条件下可以估计其平稳分布的尾部特征.本文对文献[1]中提出的方法在某一环节上作了改进,使之更适合于实际计算,并用此方法分析了两个具有实际应用背景的排队模型,即T-SPH/M/1排队和M/T-SPH/1排队,分析结果表明,在一定条件下,这两类排队系统的队长分布的尾部都具有几何衰减的特性. In the research of complex stochastic models, Quasi-Birth-and-Death processes (QBDs) whose level and phase are all infinite will take place frequently. The computation of the stationary distributions for such processes is still an open problem. For some special T-QBD processes where the blocks of the density matrix have a tri-diagonal structure, Ref. [1] showed out that under certain conditions, the tail of the stationary distribution can be estimated. In this paper, we improve the method in Ref. [1] to make it more applicable in practice. We analyze the T-SPH/M/1 queue and the M/T-SPH/1 queue with this method and show that the tails of the stationary distributions of queue length for such two queues all have the characteristic of geometric decay.
机构地区 上海大学数学系
出处 《运筹学学报》 CSCD 北大核心 2008年第1期60-70,共11页 Operations Research Transactions
关键词 运筹学 无限位相 T-QBD过程 T-SPH分布 平稳分布 几何尾部 Operations research, infinite phase, T-QBD process, T-SPH distribution, stationary distribution, geometric tail
  • 相关文献

参考文献12

  • 1Hapue L., Zhao Y.Q. and Liu L.M. Sufficient conditions for a geometric tail in a QBD process with many countalbe levels and phases[J]. Stochastic Moddels, 2005, 21(1): 77-99.
  • 2Kijima M. Quasi-stationary distribution of single-sever phase type queues[J]. Mathematics of Operations Research, 1993, 18(2): 323-437.
  • 3Neuts M.F. Matrix-geometric solutions in stochastic models: An algorithmic Approach[R]. The Johns Hopkins University Press, Baltimore, 1981.
  • 4Neuts M.F. Structured stochasitc matrics of M/G/1 type and their applacations[M]. Marcel Decker Inc., New York, 1989.
  • 5Sengupta B. Phase type representation of matrix-geometric solution[J]. Stochasitc Models, 1991, 7(6): 163-167.
  • 6Seneta E. Non-negative matrices and Markov chains[M]. Springer verlag, New York, 1981.
  • 7Shi D.H., Guo J.L. and Liu L.M. SPH-distributions and the rectangle-iterative algorithm[R]. Matrix-analysis methods in stochastic models.. Chakravarthy S. R. and Alfa A. S. Eds., Marcel Decker, New York, 1997, 207-224.
  • 8Takahashi Y., Fujuimoto K. and Makimoto N. Geometric decay of the steady-state probabilities in a quasi-birth-and-death process with a countable number of phases[J]. Stochasitc Models, 2001, 17(1): 1-24.
  • 9Tweedie R.L. Operator-geometric stationary distributions of Markov chains with application to queueing models[J]. Adv. Appl. Prob., 1982, 14: 368-391.
  • 10Grassmann W.K. and Heyman D.P. Equilibrium distribution of block-structured Markov chains with repeating rows[J]. J. Appl. Prob., 1990, 27: 557-576.

同被引文献15

  • 1Neuts M F. Matrix-geometric solutions in stochastic models: An algorithmic Approach[M]. The Johns Hopkins University Press : Baltimore, 1981.
  • 2Tweedie R L. Operator-geometric stationary distributions of Markov chains with application to queueing models[J]. Adv Appl Prob,1982,14:368-391.
  • 3Miller D R. Computation of steady-state probabilities for M/M/1 priority queues[J]. Operations Research,1981, 29:945-958.
  • 4Takahashi Y, Fujuimoto K, Makimoto N. Geometric decay of the steady-state probabilities in a quasi-birth-and- death process with a countable number of phases[J]. Stochasitc Models,2001,17:1-24.
  • 5Hapue L, Zhao Y Q, Liu L M. Sufficient conditions for a geometric tail in a QBD process with many countable levels and phases[J]. Stochastic Models,2005,1 : 77-99.
  • 6Motyerm A, Taylor P G. Decay rates for quasi-birth-and-death processes with countably many phases and tridiagonal block generators [J]. Adv Appl Prob, 2006,38 : 522-544.
  • 7Liu L M, Miyazawa M, Zhao Y Q. Geometric decay in level-expanding QBD models[J]. Ann Oper Res,2008, 160:83-98.
  • 8Ramaswami V, Taylor P G. An operator-analytic approach to product-form networks[J]. Stochastic Models, 1996,12:121-142.
  • 9Latouche G, Taylor P G. Level-phase independence for GI/M/1-type Markov chains[J]. J Appl Prob,2000,37: 984-998.
  • 10Green D A. Departure Process from MAP/PH/1 Queues[M]. PhD thesis, University of Adelaide,1999.

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部