臺灣博碩士論文加值系統

藉由研究M/G/2/3服務系統的子密度分配(sub - density)，
$f_1(s)$、$f_2(s,t)$、$f_3(s,t)$，分別代表系統在穩定狀態時系統有1、2、3人並且服務員已服務時間為s、(s,t)、(s,t)的密度函數，能較有效的求系統的穩定機率及其他特殊值。在這篇研究中我們找到M/M/2/3密度函數的解析解，和M/G/2/3的數值解以及近似解，其中近似解可表現為三個已知函數的線性組合，並且有不錯的效率和近似。之後我們試著將演算法推廣至M/G/2/K，並討論M/G/C/K計算上的可能方法。這篇論文的架構如下，第一章回顧相似的文獻並介紹這篇研究所使用的方法，第二章中探討M/M/2/3的情況，以矩陣運算的方式得到系統的密度函數及機率，第三章中探討M/G/2/3的情況，並列出數值演算法和近似演算法，第四章中列出實驗結果，第五章將演算法推廣至M/G/2/K並討論M/G/C/K的情況，第六章是結論。

By studying the sub-density of the M/G/2/3 queuing system,$f_1(s)$、$f_2(s,t)$、$f_3(s,t)$,which respectively stand for the
density function of the system in a steady state when the system has 1,2,3 people and they are has been serving for s, (s ,t), (s, t) unit of time, we can find the density function of the system and other special values (e.g.stationary probability). In this study, we find the analytical solution of the M/M/2/3, the numerical solution and the approximate solution of M/G/2/3 where the approximate solution can be expressed as the linear combination of several known functions and have good efficiency and approximation. We then try to extend the algorithm to M/G/2 /K and discuss possible approaches to M/G/C/K calculations. The structure of this paper is as follows. In the first chapter, we review the similar literature and introduce the method used in this study. In chapter 2, we discuss the situation of M/M/2/3, and solve the density function and the stationary probability. The third chapter to explore the M/G/2/3 situation, and lists the numerical algorithm and approximate algorithm. The fourth chapter lists the experimental results. The fifth chapter will be extended to M/G/2/K and discuss the case of M/G/C/K. In the end, the chapter sixth is the conclusion.

目錄
摘要...ii
Abstract...iii
誌謝...iv
目錄...v
圖目錄...vi
表目錄...vii
一、緒論...1
1.1 研究動機...1
1.2 文獻回顧...1
1.3 研究方法...2
二、M/M/2/3...6
2.1 平衡方程式...6
2.2 穩定機率理論解...7
三、M/G/2/3...10
3.1 平衡方程式...10
3.2 數值演算法...13
3.3 近似演算法...14
四、數值實驗...16
4.1 穩定機率實驗...16
4.2 服務分配實驗...17
五、M/G/C/K...20
5.1 M/G/2/K...20
5.2 M/G/C/K...21
六、結論...24
參考文獻...25
附錄一Stationary probability of M/M/2/3...27

J.Sztrik.
Basic queueing theorem.[Online]
http://irh.inf.unideb.hu/~jsztrik/education/16/SOR_Main_Angol.pdf

Iversen, V. B. (1997). Queueing systems with constant service time and evaluation of M/D/1, k. COST257 TD (97), 38.

Tijms, H. (2006). New and old results for the M/D/c queue. AEU-International Journal of Electronics and Communications, 60(2), 125-130.

Neuts, M. F. (1981). Matrix-geometric solutions in stochastic models: an algorithmic approach. Courier Corporation.

Begin, T., & Brandwajn, A. (2013, July). A note on the accuracy of several existing approximations for M/Ph/m queues. In Computer Software and Applications Conference Workshops (COMPSACW), 2013 IEEE 37th Annual (pp. 730-735). IEEE.

De Smit, J. H. (1983). A numerical solution for the multi-server queue with hyper-exponential service times. Operations research letters, 2(5), 217-224.

Ramaswami, V., & Lucantoni, D. M. (1985). Algorithms for the multi-server queue with phase type service. Stochastic Models, 1(3), 393-417.

Baba, Y. (1983). Algorithmic solution to the M/PH/c queue with batch arrivals. J. OPER. RES. SOC. JAPAN., 26(1), 33-50.

Shin, Y. W., & Moon, D. H. (2009). Sensitivity and approximation of M/G/c queue: numerical experiments. In Proceedings of the 8th International Symposium Operations Research and its Application (ISORA’09) (pp. 140-147).

Gupta, V., Harchol-Balter, M., Dai, J. G., & Zwart, B. (2010). On the inapproximability of M/G/K: why two moments of job size distribution are not enough. Queueing Systems, 64(1), 5-48.


Sgouropoulos, N., Yao, Q., & Yastremiz, C. (2015). Matching a distribution by matching quantiles estimation. Journal of the American Statistical Association, 110(510), 742-759.

Tijms, H. C., Van Hoorn, M. H., & Federgruen, A. (1981). Approximations for the steady-state probabilities in the M/G/c queue. Advances in Applied Probability, 13(1), 186-206.

Xu, X., Wang, W., & Xu, S. (2008, October). Performance of a queuing model with Hyper-Erlang distribution service for wireless network nodes. In Wireless Communications, Networking and Mobile Computing, 2008. WiCOM'08. 4th International Conference on (pp. 1-4). IEEE.

Cossette, H., Landriault, D., Marceau, E., & Moutanabbir, K. Moment# Based Approximation with Mixed Erlang Distributions.

Psounis, K., Molinero-Fernández, P., Prabhakar, B., & Papadopoulos, F. (2005). Systems with multiple servers under heavy-tailed workloads. Performance Evaluation, 62(1), 456-474.

Brandwajn, A., & Begin, T. (2014). Reduced complexity in M/Ph/c/N queues. Performance Evaluation, 78, 42-54.