國立虎尾科技大學 |

Performance Approximations and Controls to Queueing Networks.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Performance Approximations and Controls to Queueing Networks./
Author:	Yu, Yao.
Description:	1 online resource (167 pages)
Notes:	Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.
Contained By:	Dissertation Abstracts International78-10B(E).
Subject:	Industrial engineering. -
Online resource:	click for full text (PQDT)
ISBN:	9781369857344

Performance Approximations and Controls to Queueing Networks.
Yu, Yao.

Performance Approximations and Controls to Queueing Networks. - 1 online resource (167 pages)

Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.

Thesis (Ph.D.)

Includes bibliographical references

This item is not available from ProQuest Dissertations & Theses.

This dissertation provides closed-form formulas to approximate the performance of general queueing models and effective control policies to reduce congestion in queueing networks. Queueing models are widely used in the analysis of service systems such as call centers and healthcare systems. For the smaller-scale system having exponential distributions, it is possible to obtain exact performance solutions. But when the scale is large or the system is endowed with more realistic stochastic features, it can be extremely difficult to develop closed-form performance formulas. Computer simulations can be used for studying system performance, but they are time-consuming and lack of structural insights. Hence, heavy-traffic limits, obtained from asymptotic analysis on queueing models, are widely used as efficient approximations.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9781369857344Subjects--Topical Terms:

679492
Industrial engineering.
Index Terms--Genre/Form:

554714
Electronic books.

Performance Approximations and Controls to Queueing Networks.
LDR:04117ntm a2200373Ki 4500 001 910807
005 20180517112611.5
006 m o u
007 cr mn||||a|a||
008 190606s2017 xx obm 000 0 eng d
020 $a 9781369857344
035 $a (MiAaPQ)AAI10610851
035 $a AAI10610851
040 $a MiAaPQ $b eng $c MiAaPQ
099 $a TUL $f hyy $c available through World Wide Web
100 1 $a Yu, Yao. $3 1182272
245 1 0 $a Performance Approximations and Controls to Queueing Networks.
264 0 $c 2017
300 $a 1 online resource (167 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.
500 $a Adviser: Yunan Liu.
502 $a Thesis (Ph.D.) $c North Carolina State University $d 2017.
504 $a Includes bibliographical references
506 $a This item is not available from ProQuest Dissertations & Theses.
520 $a This dissertation provides closed-form formulas to approximate the performance of general queueing models and effective control policies to reduce congestion in queueing networks. Queueing models are widely used in the analysis of service systems such as call centers and healthcare systems. For the smaller-scale system having exponential distributions, it is possible to obtain exact performance solutions. But when the scale is large or the system is endowed with more realistic stochastic features, it can be extremely difficult to develop closed-form performance formulas. Computer simulations can be used for studying system performance, but they are time-consuming and lack of structural insights. Hence, heavy-traffic limits, obtained from asymptotic analysis on queueing models, are widely used as efficient approximations.
520 $a To gain heavy-traffic limits, we need to justify the existence and uniqueness of heavy-traffic limits. Then, by appropriately scaling heavy-traffic limits, we can use to develop approximations for realistic queueing models. However, if approximations sometimes do not perform good, we refine approximations following some engineering principles to enhance the approximation accuracy.
520 $a In this dissertation, we study the heavy-traffic limits for various queueing models (networks) in Chapters 2-5. We start from establishing heavy-traffic limits for loss model (GI/GI/n/0) in Chapter 2. Due to the difficulty, we prove the existence of convergent sequences for GI/GI/ n/0 in the asymptotic analysis but left the uniqueness and the analytical form of heavy-traffic limits unsolved. In Chapters 3 and 4, based on the well established heavy-traffic limits for time-varying queues with customer abandonments (Gt/M/st + GI), we find approximations sometimes become ineffective to approximate system performance. Therefore, we develop simple engineering approximation formulas for the overloaded steady-state multi-server queues (GI/GI/n + GI) and time-varying queues with customer abandonments (Gt/GI/st + GI). We then conduct extensive numerical experiments to test the effectiveness of new approximations across a wide range of model parameters. In Chapter 5, we observe that, in real service systems with dedicated servers, customers usually experience pre-arrival delays before actually arrive at the dedicated server. When the pre-arrival delay is close to zero, the well-studied joining-the-shortest queue (JSQ) policy has been proven effective. However, JSQ starts to degrade significantly as this delay grows. We propose new routing policies which make the queue asymptotically achieve resource pooling and develop the corresponding heavy-traffic limits. We also conduct computer simulations to verify the efficiency of the new policy.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Industrial engineering. $3 679492
650 4 $a Operations research. $3 573517
655 7 $a Electronic books. $2 local $3 554714
690 $a 0546
690 $a 0796
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a North Carolina State University. $b Industrial Engineering. $3 1182273
773 0 $t Dissertation Abstracts International $g 78-10B(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10610851 $z click for full text (PQDT)