Response time in a tandem queue with blocking, Markovian arrivals and phase-type services

Size: px
Start display at page:

Download "Response time in a tandem queue with blocking, Markovian arrivals and phase-type services"

Transcription

1 Operatons Research Letters 33 (25) Operatons Research Letters wwwelsevercom/locate/dsw Response tme n a tandem queue wth blockng, Markovan arrvals and phase-type servces B Van Houdt a,,1, Attahru Sule Alfa b a Department of Mathematcs and Computer Scence, Unversty of Antwerp, Performance Analyss of Telecommuncaton Systems Research Group, Mddelhemlaan 1, B-22 Antwerp, Belgum b Department of Electrcal and Computer Engneerng, Unversty of Mantoba, Wnnpeg, Mantoba, R3T 5V6, Canada Receved 22 January 24; accepted 2 August 24 Avalable onlne 6 October 24 Abstract A novel approach for obtanng the response tme n a dscrete-tme tandem-queue wth blockng s presented The approach constructs a Markov chan based on the age of the leadng customer n the frst queue We also provde a stablty condton and carry out several numercal examples 24 Elsever BV All rghts reserved Keywords: Tandem queue; Blockng; Markovan arrvals; Phase-type servces; Response tme; Matrx analytc method 1 Introducton In most communcaton networks we are nterested n obtanng the response tme of jobs as well as the number of jobs n the system It has been common practce to frst obtan the queue length and then use that to obtan the response tme The two part procedure could be cumbersome n many stuatons, Correspondng author Tel: ; fax: E-mal address: bennyvanhoudt@uaacbe (B Van Houdt) 1 B Van Houdt s a post-doctoral Fellow of the FWO-Flanders Ths work was performed whle Houdt was vstng the Unversty of Mantoba and TR Labs n Wnnpeg, Mantoba We would lke to thank the Unversty and TR Labs for ther hosptalty especally when dealng wth tandem queues Ths s because n order to obtan the queue length we have to set up the assocated Markov chan, obtan ts statonary vector and then use that n a process that nvolves a consderable amount of effort to obtan the response tme In ths paper we present a dfferent and novel approach n whch we set up the Markov chan, based on the age of the leadng job n the frst queue and other auxlary varables The statonary dstrbuton of ths Markov chan wll lead us to the response tme easly and the queue length can also be computed from t usng a smple procedure We beleve that the effort requred to compute the two quanttes s less usng ths age process approach we are presentng, when compared to the tradtonal approach of usng the Markov chan of queue length More mportantly, snce the /$ - see front matter 24 Elsever BV All rghts reserved do:1116/jorl2484

2 374 B Van Houdt, AS Alfa / Operatons Research Letters 33 (25) response tme s sometmes the only key measure of nterest, our approach s very favorable n such cases Tandem queues wth blockng have receved consderable attenton n the queueng, communcatons and manufacturng lterature because of ther pervasveness and sgnfcance n real lfe A number of survey papers have been publshed durng the last two decades [7,2,1] Some of the earler works on ths nclude those of Hunt [12], who frst studed the blockng effects n a sequence of watng lnes Later Av-Itzhak [1] studed the system wth arbtrary nput and regular servce tmes There has been a plethora of studes on ths subject and an addtonal lterature survey can be found n [21] A contnuous-tme tandem queue wth blockng, Markovan arrvals (MAP) and no ntermedate buffer was consdered by Gómez-Corral [6] Phasetype servce was assumed at the frst nfnte queue whereas the other queue s assumed to have a general servce tme Results for the jont queue length dstrbuton were provded and the stablty ssues were partally addressed Gómez-Corral used matrx analytcal methods (MAMs), as we do, to obtan hs results, the dfference n methodology beng that we propose a novel approach that keeps track of the age of the leadng customer n the frst queue as opposed to the number of customers, as ths more easly leads to the response-tme dstrbuton MAMs have been used on a varety of occasons when studyng tandem queues wth blockng, gong back to Latouche and Neuts [15] The bulk of the work done n ths area of tandem queues focussed on contnuous-tme models As ponted out by Daduna [4], the ntroducton of the asynchronous transfer mode as a multplexng technque for broadband ntegrated servces dgtal networks has ncreased the studes n the areas of dscrete-tme queueng models Even though carryng out dscrete-tme analyss of tandem queues may be done, to some extent, n a smlar manner as ther contnuous-tme counterparts ther analyss often ntroduce some addtonal challenges Settng up ther transton matrces s more complex, and obtanng the response tmes of such a system after that requres a more consderable effort Gün and Makowsky [8,9] consdered a dscrete-tme tandem queue wth blockng (and falures), Bernoull arrvals and phase-type servces They used the MAM approach also and assumed both watng rooms to be fnte Daduna [4] consdered the case wth an nfnte watng room for the frst queue, but restrcted hmself to Bernoull servce processes Desert and Daduna [5] focused on dscrete-tme tandem queues wth state-dependent Bernoull servce rates and a state-dependent Bernoull arrval stream at the frst node They obtaned the jont sojourn tme dstrbuton for a customer traversng the tandem system under consderaton In our current paper we consder an nfnte watng room n front of the frst server and allow Markovan arrvals (D-MAP), enablng us to model arrval processes that have some elements of correlatons, whch s more common n the telecommuncaton feld where arrvals are usually bursty Moreover, whle Gün and Makowsky focus on the jont queue length dstrbuton, we provde an algorthm for the total response tme of a customer and address the stablty ssues rased by the nfnte queue Other related works, n the sense that they focus on the response tme as opposed to the jont queue length, are those by van der Me et al [22] and Knessl and Ter [13], who studed the frst two moments of the response tme n an open two-node queueng network wth feedback for the case wth an exponental processor sharng (PS) node and a FIFO node (whle the arrvals at the PS node are Posson) Chao and Pnedo [3] consdered the case of two tandem queues wth batch Posson arrvals and no buffer space n the second queue They allowed the servce tmes to be general and obtaned the expected tme n system We start wth a descrpton of the model under consderaton n Secton 2, whle the GI/M/1-type Markov chan constructed to obtan the performance measures s gven n Secton 3 An effcent method to compute the response tme and the jont queue length dstrbuton from the steady-state vector s presented n Secton 4, whereas Secton 5 addresses the stablty ssues surroundng our model We end by demonstratng the strength of our model through a varety of numercal examples 2 Model descrpton Consder two queues n tandem, where the frst queue has an nfnte watng lne and the second has a fnte one wth capacty B Customers arrve (to the frst queue) accordng to a dscrete-tme Markovan arrval process (D-MAP), characterzed by the

3 B Van Houdt, AS Alfa / Operatons Research Letters 33 (25) l l sub-stochastc matrces D and D 1 The matrx D = D + D 1 s the stochastc matrx of the underlyng Markov chan that governs the arrval process The element (D k ),j, 1, j l, k=, 1, represents the probablty of makng a transton from state to j wth k arrvals Let γ be the statonary dstrbuton assocated wth D, then the arrval rate s gven by λ = γd 1 1 l, where 1 l s a l 1 column vector of ones For more detals on MAP see [2,16] The servce requred by a customer n the th queue s phase-type (PH) dstrbuted wth matrx representaton (m, α,t ), for = 1, 2 It s well known that PH dstrbutons are very good for representng most of the types of servces encountered n communcaton systems [14] The mean servce tme of a PH s gven as =α (I m T ) 1 1 m, where I x s an x-dmensonal dentty matrx The matrx T s sub-stochastc and s of order m, whle t s defned as 1 m T 1 m The elements (α ) s of the stochastc vector α represent the probablty that a customer starts hs servce μ 1 n phase s Let r (k) be the probablty that the servce tme at node lasts for k or more unts of tme, then r (k) = α T k 1 1 m,k 1 Notce, the mnmum servce tme at node s at least 1 and the probablty that the servce tme equals exactly k s found as α T k 1 t For more detals on the phase-type dstrbuton see [19] Whenever a customer fnshes servce n the frst queue t advances to the second queue (at no swtchng cost), unless the watng lne of the second queue s already fully occuped In ths case, the customer remans wthn the servce faclty of the frst queue untl there s a servce completon n the second queue Thereby, preventng any other customers watng n the watng lne of queue 1 from enterng the server (meanng, we adopt the blockng-after-servce mechansm, see [2, p 6]) Both queues serve ther customers n a FCFS order All events such as arrvals, transfers from a watng lne to the server and servce completons are assumed to occur at nstants mmedately after the dscrete tme epochs Ths mples, amongst others, that the age of a customer n servce at some tme epoch n s at least 1 3 The GI/M/1-type Markov Chan A Markov chan (MC) that allows us to effcently obtan the response-tme dstrbuton of an arbtrary customer, s constructed next The state space of ths MC wll be subdvded nto an nfnte number of groups, called levels Level zero wll contan all the states that correspond to a stuaton n whch the frst server s dle Whereas level, for >, reflects the fact that the frst server s occuped by a customer of age (ether because he s beng served or because he s blocked by the second queue) To be more specfc, the states of level are dvded nto 2 sets: BI ={j 1 j l}: The MC s sad to be n state j of the set BI at tme n, f both servers are dle and the arrval process s n state j at tme n FI ={(b, s 2,j) b B,1 s 2 m 2, 1 j l}: If the frst server s dle, the second server s occuped by a customer whose servce s n phase s 2 and b customers are watng to be served by the second server, whle the state of the D-MAP s j at tme n, then the MC s sad to be n state (b, s 2,j) of the set FI at tme n The states of level, for >, are further subdvded nto three sets: SI={(s 1,j) 1 s 1 m 1, 1 j l}: The MC s sad to be n state (s 1,j)of the set SI, at tme n, n case the second server s dle, an age customer s served by server 1, the phase of hs servce equalng s 1, whle the arrval process s, at tme n + 1, n state j BS = {(b, s 1,s 2,j) b B,1 s v m v,v = 1, 2; 1 j l}: The stuaton n whch, at tme n, a customer of age s beng served by server 1, there are b customers watng for servce n the second watng lne, the phase of servce n the vth server equals s v, for v = 1, 2, and the state of the D-MAP at tme n + 1 equals j, wll correspond to the state (b, s 1,s 2,j) of level BL ={(s 2,j) 1 s 2 m 2, 1 j l}: The scenaro where, at tme n, there s an age customer blocked n server 1, a customer s served by server 2, whose current phase s s 2, and the state of the D-MAP at tme n +1 equals j s represented by state (s 2,j) of the set BL Let S denote the number of elements n a set S Defne d t and d b as SI + BS + BL =(B+1)m 1 m 2 l+(m 1 + m 2 )l and BI + FI =(B + 1)m 2 l + l, respectvely As we shall explan later on, the transton matrx P

4 376 B Van Houdt, AS Alfa / Operatons Research Letters 33 (25) of ths MC has the followng form: B 1 B B 2 A 1 A P = B 3 A 2 A 1 A B 4 A 3 A 2 A 1 A, (31) In case (), the number of customers n the second queue wll ether reman the same or decrease by one, dependng on whether there s a servce completon n server 2 In cases () and (), the number of customers n the second watng lne has to reman equal to B, mplyng that there can be no servce completon As a result, we fnd A = K I l, where T 1 T 1 t 2 T 1 T 2 T 1 t 2 α 2 T 1 T 2 K =, (32) T1 T 2 T 1 t 2 α 2 T 1 T 2 t 1 T 2 T 2 where the matrces A k are d t d t matrces, B k, for k 2, s a d t d b matrx, B a d b d t and B 1 a square matrx of dmenson d b Notce, the matrx A k represents the transton probabltes of gong from level >tolevel k + 1, for k, whereas the matrces B k are related to transtons from and/or to level Let us now dscuss the matrces A k and B k n detal, the structure of P wll be apparent from ths dscusson Assume the MC s n some state of level > at tme n, meanng that an age customer, referred to as customer c, s occupyng server 1 In order to get a transton to level + 1, customer c has to reman n server 1 Ths s because customers are served n a FCFS order and there are no batch arrvals; hence, the age of the very next customer who arrves after c cannot be larger than at tme n + 1 There are three scenaros that would cause customer c to reman n server 1: () hs servce (n server 1) dd not fnsh at tme n, () hs servce fnshed, but he s blocked by the second queue, or () customer c remans blocked I k represents the dentty matrx of dmenson k, t = 1 m T 1 m, for = 1, 2; and 1 k sa1 k vector wth each entry set to 1 Notce, ths matrx does not depend on the age of customer c Next, we consder the transtons from level to k + 1, for k 1 Thus, as before we have a customer, called c, occupyng server 1 at tme n To get a transton to level 1 k + 1, customer c has to leave server 1 Moreover, a new customer, whose age should equal k + 1 at tme n + 1, has to enter the server at tme n, call hm customer c Meanng, the nterarrval tme between c and c has to equal k The fact that customer c leaves server 1 mples that the MC cannot make a transton to one of the states SI BL of level k + 1 Also, gven that the watng lne of server 2 was fully occuped at tme n, there should have been a servce completon n server 2 (otherwse c would become/reman blocked) Fnally, f there stll was a vacancy n the watng lne of the second queue at tme n, the number of watng customers there ether ncreases by one or remans the same, dependng on whether there s a servce completon n server 2 Hence, transtons from level to k + 1 are governed by the matrx A k = K 1 D k 1 D 1, where t 1 α 1 α 2 t 1 α 1 t 2 α 2 t 1 α 1 T 2 t 1 α 1 t 2 α 2 K 1 = (33) t1 α 1 t 2 α 2 t 1 α 1 T 2 t 1 α 1 t 2 α 2 α 1 t 2 α 2

5 B Van Houdt, AS Alfa / Operatons Research Letters 33 (25) The matrces B k, for k, descrbng the transtons to and from level can be obtaned usng smlar arguments (see [11] for detals) B k, for k>1, equals K1 c(α 1 1) D k 1, where K1 c(α 1 1) s obtaned from K 1 by replacng all vectors α 1 by the scalar 1 and removng the last m 2 columns The matrx B 1 can be wrtten as K r(t 1 α 1,t 1 ) D 1 To obtan K r(t 1 α 1,t 1 ) we replace all matrces T 1 appearng n the expresson for K nto the vector α 1, the vector t 1 nto the scalar and remove the last m 2 rows of K Fnally, the matrx B 1 obeys the followng equaton: 1 t 2 T 2 t B 1 = 2 α 2 T 2 D T2 t 2 α 2 T 2 (34) Let π =[π, π 1, π 2,] be the steady-state vector of P, where π sa1 d b vector and π, for >, a1 d t vector Snce, the MC characterzed by P s a GI/M/1-type MC, π exsts f and only f θβ > 1, where θ k A k =θ, θ1 dt =1 and β= k 1 ka k1 dt The stablty condton θβ > 1 s dscussed n Secton 5 The steady-state vector of a GI/M/1-type MC can be found by teratvely solvng for the mnmal nonnegatve R n the non-lnear equaton R = k Rk A k (see [19]) Havng solved ths equaton we fnd π as [π, π 1 ] =[π, π 1 ] [ B 1 B k 1 Rk 1 B k+1 k 1 Rk 1 A k ], (35) π = π 1 R, (36) where >1, π and π 1 are normalzed as π 1 db + π 1 (I R) 1 1 dt = 1 However, the computaton of π can be done much more effcently (both n terms of the tme and memory complexty) by constructng a quas-brth Death (QBD) Markov chan and applyng the cyclc reducton algorthm [17] to solve ths QBD We refer to [11] for detals on the QBD reducton 4 Performance measures In ths secton we demonstrate how to get the response-tme dstrbuton from the steady-state vector π We start by ntroducng the followng set of random varables: T W : The amount of tme a tagged customer has to wat n the th watng lne, for = 1, 2 T S : The servce tme duraton of a tagged customer n server, for = 1, 2 T B : The tme that elapses whle a tagged customer s blocked n server 1 Havng defned these varables the total response tme T R of a tagged customer s defned as T W1 + T S1 + T B + T W2 + T S2, whle the response tme n queue 1, denoted as T R1, equals T W1 + T S1 + T B We need two more varables before we can proceed: F N : The number of customers stll requrng full servce by server 2 before a tagged customer who just left server 1 can start hs servce F T : The remanng servce tme of the customer occupyng server 2 when a tagged customer leaves server 1 Wrte π as [π SI, π BS, π BL ] n accordance wth the three sets of states of level, then P [T R1 = r, F N = b, F T = h] = (t 1 ) s1 λ 1 {b=&h=} s {b<b h=} s2,j π BS r + 1 {b=b&h=} λ j r (s 1,j) π SI (b, s 1,s 2, j)(t2 h t 2) s2 π BL r (s 2, j)(t 2 ) s2, s2,j where (x) denotes the th component of the vector x From ths t s straghtforward to compute the probabltes P [T R1 + F T = r, F N = b] The total

6 378 B Van Houdt, AS Alfa / Operatons Research Letters 33 (25) response-tme dstrbuton s then found by P [T R = ]= P [T R1 + F T = r, F N = b] b r b P [S ( b+1) 2 = r], (47) where S ( x) 2 s the x-fold convoluton of the PH servce tme dstrbuton of server 2 characterzed by (m 2, α 2,T 2 ) The blockng probablty p BL can be computed as p BL = 1 π BS r (B, s 1,s 2, j)(t 1 ) λ s1 (T 2 ) s2, r> s 1,s 2,j (48) and the blockng tme dstrbuton as P [T B = t]= 1 π BS r (B, s 1,s 2, j)(t 1 ) λ s1 r> s 1,s 2,j (T t 2 t 2) s2, (49) for t> and P [T B = ]=1 p BL Fnally, notce that the probablty of havng a jont queue contents equal to (q 1,q 2 ), equals the probablty of havng a customer of age a n the frst servce center, q 2 customers n the ntermedate queue and havng q 1 arrvals durng a tme nterval of length a (that starts mmedately after the customer n the frst servce center arrved) Hence, a smple procedure can be devsed to obtan the jont queue contents dstrbuton from the steady state probablty vector π (see [11]) 5 Stablty condton The GI/M/1-type Markov chan ntroduced n Secton 3 s ergodc f and only f θβ > 1, where θ k A k = θ, θ1 dt = 1 and β = k 1 ka k1 dt The followng theorem summarzes the stablty results (see [11] for detals) Theorem 51 The MC developed n Secton 3 s stable f and only f (1) λ/k<1, where λ s the D-MAP mean arrval rate and k = μ 1 (1 κ BL 1 m2 ) = μ 2 (1 κ SI 1 m1 ), (51) wth μ the servce rate of the th servce center and κ=[κ SI, κ, κ 1,,κ B, κ BL ] s the nvarant probablty vector of K = K + K 1 Thus, the arrval process only affects the stablty through ts mean arrval rate λ (2) the queue contents process of the nfnte watng lne of queue 1 has a steady state Remark 1 Eq (51) suffces to prove that nterchangng both servce-tme dstrbutons does not affect the stablty of the system Intutvely, when studyng the stablty, we may assume that there are always customers ready to be served n the nfnte watng lne Now, usng an argument by Melamed [18], we can regard the empty places (holes) n the ntermedate buffer as dual customers For each regular customer that moves through the system, a dual customer recevng dentcal servce moves n the opposte drecton Hence, the maxmum stable throughput of the regular customers and the dual customers must be dentcal Clearly, the dual system s dentcal to the nterchanged system Ths type of nterchangeablty result was already establshed long ago for exponental servers (and Posson arrvals) [19, Secton 52] Remark 2 In the specal case where the sze of the ntermedate watng lne B equals, one can prove that 1/k s nothng but the mean of the maxmum of both PH servce tme dstrbutons 6 Numercal results In ths secton we present some numercal examples that provde nsght on the system behavor 61 Influence of the capacty Band the correlaton of the D-MAP arrval process Consder an nterrupted Bernoull process (IBP), where the mean sojourn tme n both states equals x c and an arrval occurs n the on-state wth probablty x p Thus, [ ] 1 1/xc 1/x D = c, D 1 = (1 x p )/x c (1 x p )(1 1/x c ) [ x p /x c x p (1 1/x c ) ] (611)

7 B Van Houdt, AS Alfa / Operatons Research Letters 33 (25) (a) (b) Fg 1 Response-tme dstrbuton for varous capactes B, x p = 1 16 and (a) x c = 4, (b) x c = 4 The servce tme dstrbuton n server 1 s hypergeometrc wth parameters: [ 45 ] α 1 =[ 1 3, 2 3 ], T 1 = (612) 19 2 Thus, wth probablty 1 3 and 2 3 the servce tme s geometrcally dstrbuted wth a mean of 5 and 2 tme unts, respectvely The mean of ths dstrbuton s 15 tme unts The servce-tme dstrbuton of the second server s characterzed by 3 1 α 2 =[ 16 1, , 16 ], T 2 = , (613) x l where x l = s chosen such that the mean servce tme equals 15 We have chosen the mean servce tme dentcal n both servers as ths ought to create a strong couplng between both queues Fg 1 depcts the response-tme dstrbuton for varous capactes B, for x p = 16 1 (meanng that the arrval rate λ = 32 1 ), and x c ether 4 or 4 Clearly, the larger x c the more correlated the arrval process Obvously, stronger correlated arrvals gve rse to slower response tmes, whle addng more capacty B between both servers reduces the response tme However, at some pont there s lttle use n further augmentng the capacty as the response tme seems to converge for B large Ths s easly understood as the blockng probablty tends to decrease to zero whle ncreasng B The fgure further llustrates that the rate of convergence s affected by the correlaton of the arrval process: stronger correlated arrval processes more easly justfy ncreasng the capacty B 62 The maxmum arrval rate λ and the varaton of the servce tmes We consder the same IBP process as n the prevous secton The servce-tme dstrbuton s ether geometrc (Geo), Erlang-5 (Er5) 1 or hypergeometrc (HypGeo) characterzed by [ 45 ] α 1 =[ 1 9, 1 1 ], T 1 = (614) The mean of each of these servce-tme dstrbutons s 15 tme unts The HypGeo dstrbuton s the most varable of the three, followed by Geo and Er5 Fg 2 shows the maxmum stable load ρ max, defned as E[S 1 ]λ max = 15λ max, as a functon of B for dfferent server confguratons λ max s the maxmum arrval rate λ for whch the system s stable Recall, λ max =μ 1 (1 κ BL 1 m2 )=μ 2 (1 κ SI 1 m1 ), see Secton 5 The notaton (X, Y ) ndcates that the servce tme n servers 1 and 2 s dstrbuted as X and Y, respectvely 1 That s, the sum of 5 ndependent and dentcally dstrbuted geometrc random varables

8 38 B Van Houdt, AS Alfa / Operatons Research Letters 33 (25) Fg 2 Maxmum stable load ρ max as a functon of the capacty B for dfferent server confguratons Fg 3 Response-tme dstrbuton for dfferent server confguratons for B = 1 and IBP arrvals (x c = 4,x p = 1 16 ) Fg 2 clearly demonstrates that more varable servce tmes, that s, HypGeo, gve rse to a lower maxmum stable nput rate λ max Ths result stems from the fact that a more varable servce tme, whether n the frst or second server, causes a hgher degree of blockng n comparson wth a more determnstc servce tme dstrbuton As proved n Secton 5, nterchangng both servce tme dstrbutons does not alter the system stablty Notce, the actual nature of the D-MAP arrval process s rrelevant as the stablty s only affected by the arrval process through ts mean The response-tme dstrbuton for dfferent server confguratons s depcted n Fg 3 We assume that arrvals occur accordng to a IBP wth x c = 4 and x p = 16 1, see Secton 61 The capacty of the ntermedate watng lne B s assumed to be 1 Fg 3 confrms that the response of the system slows down as the servce tmes become more varable It further demonstrates that nterchangng the servce tmes generally causes a (lmted) change n the response tme (as opposed to the stablty) Placng the more varable server frst seems to result n a somewhat slower response Ths mght be explaned by notcng that the output process of server 1 s more bursty n such case, causng a hgher blockng probablty On the other hand, less varablty n server 2 decreases the blockng probablty, so when we nterchange both servce tme dstrbutons both these effects nfluence the degree of blockng n the system Varous numercal exper- ments, ncludng Fg 3, seem to ndcate that reducng the varaton of server 1 should be slghtly favored References [1] B Av-Itzhak, A sequence of servce statons wth arbtrary nput and regular servce tmes, Manage Sc 11 (5) (1965) [2] C Blonda, A dscrete-tme batch markovan arrval process as B-ISDN traffc model, Belg J Oper Res Statst Comput Sc 32 (3,4) (1993) [3] X Chao, M Pnedo, Batch arrvals to a tandem queue wthout an ntermedate buffer, Stochastc Models 6 (4) (199) [4] H Daduna, Queueng Networks wth Dscrete Tme Scale, Sprnger, Berln, 21 [5] B Desert, H Daduna, Dscrete tme tandem networks of queues: effects of dfferent regulaton schemes for smultaneous events, Performance Evaluaton 47 (22) [6] A Gómez-Corral, A tandem queue wth blockng and markovan arrval process, Queueng Systems 41 (22) [7] L Gün, Annotated bblography of blockng systems Techncal Report, Insttute for Systems Research ISR , 1987 [8] L Gün, M Makowsk, Matrx geometrc soluton for fnte capacty queues wth phase-type dstrbutons, n: Proceedngs of Performance 87, Brussels, 1987, (pp ) [9] L Gün, M Makowsk, Matrx geometrc soluton for two node tandem queueng systems wth phase-type servers subject to blockng and falures Techncal Report, Insttute for Systems Research, ISR 87-21, 1987

9 B Van Houdt, AS Alfa / Operatons Research Letters 33 (25) [1] NG Hall, C Srskandarajah, A survey of machne schedulng problems wth blockng and no-wat n process, Operatons Research 44 (1996) [11] B van Houdt, AS Alfa, Response tme n a tandem queue wth blockng, markovan arrvals and phase-type servces: extended verson Techncal Report, TR4/1, Unversty of Antwerp, 24 [12] GC Hunt, Sequental arrays of watng lnes, Operatons Research 4 (6) (1956) [13] C Knessl, C Ter, Approxmaton to the moments of the sojourn tme n a tandem queue wth overtakng, Stochastc Models 6 (3) (199) [14] A Lang, JL Arthur, Parameter approxmaton for phase-type dstrbutons, n: SR Chakravarthy, AS Alfa (Eds), Matrx- Analytc Methods n Stochastc Models, Marcel-Dekker, New York, 1996, pp [15] G Latouche, MF Neuts, Effcent algorthmc solutons to exponental tandem queues wth blockng, SIAM J Algebrac and Dscrete Meth 1 (1) (198) [16] DM Lucanton, New results on the sngle server queue wth a batch markovan arrval process, Stochastc Models 7 (1) (1991) 1 46 [17] B Men, Solvng QBD problems: the cyclc reducton algorthm versus the nvarant subspace method, Advances n Performance Analyss 1 (1998) [18] B Melamed, A note on the reversblty and dualty of some tandem blockng queueng systems, Management Scence 32 (12) (1986) [19] MF Neuts, Matrx-Geometrc Solutons n Stochastc Models, An Algorthmc Approach, John Hopkns Unversty Press, MD, 1981 [2] H Perros, Queueng Networks wth Blockng, Oxford Unversty Press, New York, 1994 [21] M Schamber, Decomposton methods for fnte queue networks wth a non-renewal arrval process n dscrete tme, thess, Unversty of Mantoba, 1997 [22] RD van der Me, BMM Gjsen, N n t Veld, JL van den Berg, Response tmes n a two-node queueng network wth feedback, Performance Evaluaton 49 (22) 99 11

Analysis of Discrete Time Queues (Section 4.6)

Analysis of Discrete Time Queues (Section 4.6) Analyss of Dscrete Tme Queues (Secton 4.6) Copyrght 2002, Sanjay K. Bose Tme axs dvded nto slots slot slot boundares Arrvals can only occur at slot boundares Servce to a job can only start at a slot boundary

More information

Queueing Networks II Network Performance

Queueing Networks II Network Performance Queueng Networks II Network Performance Davd Tpper Assocate Professor Graduate Telecommuncatons and Networkng Program Unversty of Pttsburgh Sldes 6 Networks of Queues Many communcaton systems must be modeled

More information

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS Avalable onlne at http://sck.org J. Math. Comput. Sc. 3 (3), No., 6-3 ISSN: 97-537 COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

More information

Structure and Drive Paul A. Jensen Copyright July 20, 2003

Structure and Drive Paul A. Jensen Copyright July 20, 2003 Structure and Drve Paul A. Jensen Copyrght July 20, 2003 A system s made up of several operatons wth flow passng between them. The structure of the system descrbes the flow paths from nputs to outputs.

More information

Introduction to Continuous-Time Markov Chains and Queueing Theory

Introduction to Continuous-Time Markov Chains and Queueing Theory Introducton to Contnuous-Tme Markov Chans and Queueng Theory From DTMC to CTMC p p 1 p 12 1 2 k-1 k p k-1,k p k-1,k k+1 p 1 p 21 p k,k-1 p k,k-1 DTMC 1. Transtons at dscrete tme steps n=,1,2, 2. Past doesn

More information

NUMERICAL DIFFERENTIATION

NUMERICAL DIFFERENTIATION NUMERICAL DIFFERENTIATION 1 Introducton Dfferentaton s a method to compute the rate at whch a dependent output y changes wth respect to the change n the ndependent nput x. Ths rate of change s called the

More information

College of Computer & Information Science Fall 2009 Northeastern University 20 October 2009

College of Computer & Information Science Fall 2009 Northeastern University 20 October 2009 College of Computer & Informaton Scence Fall 2009 Northeastern Unversty 20 October 2009 CS7880: Algorthmc Power Tools Scrbe: Jan Wen and Laura Poplawsk Lecture Outlne: Prmal-dual schema Network Desgn:

More information

Applied Stochastic Processes

Applied Stochastic Processes STAT455/855 Fall 23 Appled Stochastc Processes Fnal Exam, Bref Solutons 1. (15 marks) (a) (7 marks) The dstrbuton of Y s gven by ( ) ( ) y 2 1 5 P (Y y) for y 2, 3,... The above follows because each of

More information

Lecture 12: Discrete Laplacian

Lecture 12: Discrete Laplacian Lecture 12: Dscrete Laplacan Scrbe: Tanye Lu Our goal s to come up wth a dscrete verson of Laplacan operator for trangulated surfaces, so that we can use t n practce to solve related problems We are mostly

More information

Queuing system theory

Queuing system theory Elements of queung system: Queung system theory Every queung system conssts of three elements: An arrval process: s characterzed by the dstrbuton of tme between the arrval of successve customers, the mean

More information

COMPLETE BUFFER SHARING IN ATM NETWORKS UNDER BURSTY ARRIVALS

COMPLETE BUFFER SHARING IN ATM NETWORKS UNDER BURSTY ARRIVALS COMPLETE BUFFER SHARING WITH PUSHOUT THRESHOLDS IN ATM NETWORKS UNDER BURSTY ARRIVALS Ozgur Aras and Tugrul Dayar Abstract. Broadband Integrated Servces Dgtal Networks (B{ISDNs) are to support multple

More information

TCOM 501: Networking Theory & Fundamentals. Lecture 7 February 25, 2003 Prof. Yannis A. Korilis

TCOM 501: Networking Theory & Fundamentals. Lecture 7 February 25, 2003 Prof. Yannis A. Korilis TCOM 501: Networkng Theory & Fundamentals Lecture 7 February 25, 2003 Prof. Yanns A. Korls 1 7-2 Topcs Open Jackson Networks Network Flows State-Dependent Servce Rates Networks of Transmsson Lnes Klenrock

More information

Application of Queuing Theory to Waiting Time of Out-Patients in Hospitals.

Application of Queuing Theory to Waiting Time of Out-Patients in Hospitals. Applcaton of Queung Theory to Watng Tme of Out-Patents n Hosptals. R.A. Adeleke *, O.D. Ogunwale, and O.Y. Hald. Department of Mathematcal Scences, Unversty of Ado-Ekt, Ado-Ekt, Ekt State, Ngera. E-mal:

More information

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems Numercal Analyss by Dr. Anta Pal Assstant Professor Department of Mathematcs Natonal Insttute of Technology Durgapur Durgapur-713209 emal: anta.bue@gmal.com 1 . Chapter 5 Soluton of System of Lnear Equatons

More information

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module 3 LOSSY IMAGE COMPRESSION SYSTEMS Verson ECE IIT, Kharagpur Lesson 6 Theory of Quantzaton Verson ECE IIT, Kharagpur Instructonal Objectves At the end of ths lesson, the students should be able to:

More information

Numerical Heat and Mass Transfer

Numerical Heat and Mass Transfer Master degree n Mechancal Engneerng Numercal Heat and Mass Transfer 06-Fnte-Dfference Method (One-dmensonal, steady state heat conducton) Fausto Arpno f.arpno@uncas.t Introducton Why we use models and

More information

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4)

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4) I. Classcal Assumptons Econ7 Appled Econometrcs Topc 3: Classcal Model (Studenmund, Chapter 4) We have defned OLS and studed some algebrac propertes of OLS. In ths topc we wll study statstcal propertes

More information

Markov chains. Definition of a CTMC: [2, page 381] is a continuous time, discrete value random process such that for an infinitesimal

Markov chains. Definition of a CTMC: [2, page 381] is a continuous time, discrete value random process such that for an infinitesimal Markov chans M. Veeraraghavan; March 17, 2004 [Tp: Study the MC, QT, and Lttle s law lectures together: CTMC (MC lecture), M/M/1 queue (QT lecture), Lttle s law lecture (when dervng the mean response tme

More information

The optimal delay of the second test is therefore approximately 210 hours earlier than =2.

The optimal delay of the second test is therefore approximately 210 hours earlier than =2. THE IEC 61508 FORMULAS 223 The optmal delay of the second test s therefore approxmately 210 hours earler than =2. 8.4 The IEC 61508 Formulas IEC 61508-6 provdes approxmaton formulas for the PF for smple

More information

Continuous Time Markov Chains

Continuous Time Markov Chains Contnuous Tme Markov Chans Brth and Death Processes,Transton Probablty Functon, Kolmogorov Equatons, Lmtng Probabltes, Unformzaton Chapter 6 1 Markovan Processes State Space Parameter Space (Tme) Dscrete

More information

Temperature. Chapter Heat Engine

Temperature. Chapter Heat Engine Chapter 3 Temperature In prevous chapters of these notes we ntroduced the Prncple of Maxmum ntropy as a technque for estmatng probablty dstrbutons consstent wth constrants. In Chapter 9 we dscussed the

More information

Equilibrium Analysis of the M/G/1 Queue

Equilibrium Analysis of the M/G/1 Queue Eulbrum nalyss of the M/G/ Queue Copyrght, Sanay K. ose. Mean nalyss usng Resdual Lfe rguments Secton 3.. nalyss usng an Imbedded Marov Chan pproach Secton 3. 3. Method of Supplementary Varables done later!

More information

One-sided finite-difference approximations suitable for use with Richardson extrapolation

One-sided finite-difference approximations suitable for use with Richardson extrapolation Journal of Computatonal Physcs 219 (2006) 13 20 Short note One-sded fnte-dfference approxmatons sutable for use wth Rchardson extrapolaton Kumar Rahul, S.N. Bhattacharyya * Department of Mechancal Engneerng,

More information

Prof. Dr. I. Nasser Phys 630, T Aug-15 One_dimensional_Ising_Model

Prof. Dr. I. Nasser Phys 630, T Aug-15 One_dimensional_Ising_Model EXACT OE-DIMESIOAL ISIG MODEL The one-dmensonal Isng model conssts of a chan of spns, each spn nteractng only wth ts two nearest neghbors. The smple Isng problem n one dmenson can be solved drectly n several

More information

APPENDIX A Some Linear Algebra

APPENDIX A Some Linear Algebra APPENDIX A Some Lnear Algebra The collecton of m, n matrces A.1 Matrces a 1,1,..., a 1,n A = a m,1,..., a m,n wth real elements a,j s denoted by R m,n. If n = 1 then A s called a column vector. Smlarly,

More information

Global Sensitivity. Tuesday 20 th February, 2018

Global Sensitivity. Tuesday 20 th February, 2018 Global Senstvty Tuesday 2 th February, 28 ) Local Senstvty Most senstvty analyses [] are based on local estmates of senstvty, typcally by expandng the response n a Taylor seres about some specfc values

More information

CHAPTER 5 NUMERICAL EVALUATION OF DYNAMIC RESPONSE

CHAPTER 5 NUMERICAL EVALUATION OF DYNAMIC RESPONSE CHAPTER 5 NUMERICAL EVALUATION OF DYNAMIC RESPONSE Analytcal soluton s usually not possble when exctaton vares arbtrarly wth tme or f the system s nonlnear. Such problems can be solved by numercal tmesteppng

More information

C/CS/Phy191 Problem Set 3 Solutions Out: Oct 1, 2008., where ( 00. ), so the overall state of the system is ) ( ( ( ( 00 ± 11 ), Φ ± = 1

C/CS/Phy191 Problem Set 3 Solutions Out: Oct 1, 2008., where ( 00. ), so the overall state of the system is ) ( ( ( ( 00 ± 11 ), Φ ± = 1 C/CS/Phy9 Problem Set 3 Solutons Out: Oct, 8 Suppose you have two qubts n some arbtrary entangled state ψ You apply the teleportaton protocol to each of the qubts separately What s the resultng state obtaned

More information

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography CSc 6974 and ECSE 6966 Math. Tech. for Vson, Graphcs and Robotcs Lecture 21, Aprl 17, 2006 Estmatng A Plane Homography Overvew We contnue wth a dscusson of the major ssues, usng estmaton of plane projectve

More information

Adaptive Dynamical Polling in Wireless Networks

Adaptive Dynamical Polling in Wireless Networks BULGARIA ACADEMY OF SCIECES CYBERETICS AD IFORMATIO TECHOLOGIES Volume 8, o Sofa 28 Adaptve Dynamcal Pollng n Wreless etworks Vladmr Vshnevsky, Olga Semenova Insttute for Informaton Transmsson Problems

More information

Computation of Higher Order Moments from Two Multinomial Overdispersion Likelihood Models

Computation of Higher Order Moments from Two Multinomial Overdispersion Likelihood Models Computaton of Hgher Order Moments from Two Multnomal Overdsperson Lkelhood Models BY J. T. NEWCOMER, N. K. NEERCHAL Department of Mathematcs and Statstcs, Unversty of Maryland, Baltmore County, Baltmore,

More information

Difference Equations

Difference Equations Dfference Equatons c Jan Vrbk 1 Bascs Suppose a sequence of numbers, say a 0,a 1,a,a 3,... s defned by a certan general relatonshp between, say, three consecutve values of the sequence, e.g. a + +3a +1

More information

x = , so that calculated

x = , so that calculated Stat 4, secton Sngle Factor ANOVA notes by Tm Plachowsk n chapter 8 we conducted hypothess tests n whch we compared a sngle sample s mean or proporton to some hypotheszed value Chapter 9 expanded ths to

More information

MATH 5630: Discrete Time-Space Model Hung Phan, UMass Lowell March 1, 2018

MATH 5630: Discrete Time-Space Model Hung Phan, UMass Lowell March 1, 2018 MATH 5630: Dscrete Tme-Space Model Hung Phan, UMass Lowell March, 08 Newton s Law of Coolng Consder the coolng of a well strred coffee so that the temperature does not depend on space Newton s law of collng

More information

Lectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix

Lectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix Lectures - Week 4 Matrx norms, Condtonng, Vector Spaces, Lnear Independence, Spannng sets and Bass, Null space and Range of a Matrx Matrx Norms Now we turn to assocatng a number to each matrx. We could

More information

Bounds on the bias terms for the Markov reward approach

Bounds on the bias terms for the Markov reward approach Bounds on the bas terms for the Markov reward approach Xnwe Ba 1 and Jasper Goselng 1 arxv:1901.00677v1 [math.pr] 3 Jan 2019 1 Department of Appled Mathematcs, Unversty of Twente, P.O. Box 217, 7500 AE

More information

Notes on Frequency Estimation in Data Streams

Notes on Frequency Estimation in Data Streams Notes on Frequency Estmaton n Data Streams In (one of) the data streamng model(s), the data s a sequence of arrvals a 1, a 2,..., a m of the form a j = (, v) where s the dentty of the tem and belongs to

More information

Meenu Gupta, Man Singh & Deepak Gupta

Meenu Gupta, Man Singh & Deepak Gupta IJS, Vol., o. 3-4, (July-December 0, pp. 489-497 Serals Publcatons ISS: 097-754X THE STEADY-STATE SOLUTIOS OF ULTIPLE PARALLEL CHAELS I SERIES AD O-SERIAL ULTIPLE PARALLEL CHAELS BOTH WITH BALKIG & REEGIG

More information

CHAPTER 14 GENERAL PERTURBATION THEORY

CHAPTER 14 GENERAL PERTURBATION THEORY CHAPTER 4 GENERAL PERTURBATION THEORY 4 Introducton A partcle n orbt around a pont mass or a sphercally symmetrc mass dstrbuton s movng n a gravtatonal potental of the form GM / r In ths potental t moves

More information

A Hybrid Variational Iteration Method for Blasius Equation

A Hybrid Variational Iteration Method for Blasius Equation Avalable at http://pvamu.edu/aam Appl. Appl. Math. ISSN: 1932-9466 Vol. 10, Issue 1 (June 2015), pp. 223-229 Applcatons and Appled Mathematcs: An Internatonal Journal (AAM) A Hybrd Varatonal Iteraton Method

More information

CHAPTER III Neural Networks as Associative Memory

CHAPTER III Neural Networks as Associative Memory CHAPTER III Neural Networs as Assocatve Memory Introducton One of the prmary functons of the bran s assocatve memory. We assocate the faces wth names, letters wth sounds, or we can recognze the people

More information

NP-Completeness : Proofs

NP-Completeness : Proofs NP-Completeness : Proofs Proof Methods A method to show a decson problem Π NP-complete s as follows. (1) Show Π NP. (2) Choose an NP-complete problem Π. (3) Show Π Π. A method to show an optmzaton problem

More information

Module 9. Lecture 6. Duality in Assignment Problems

Module 9. Lecture 6. Duality in Assignment Problems Module 9 1 Lecture 6 Dualty n Assgnment Problems In ths lecture we attempt to answer few other mportant questons posed n earler lecture for (AP) and see how some of them can be explaned through the concept

More information

Lecture 4: November 17, Part 1 Single Buffer Management

Lecture 4: November 17, Part 1 Single Buffer Management Lecturer: Ad Rosén Algorthms for the anagement of Networs Fall 2003-2004 Lecture 4: November 7, 2003 Scrbe: Guy Grebla Part Sngle Buffer anagement In the prevous lecture we taled about the Combned Input

More information

More metrics on cartesian products

More metrics on cartesian products More metrcs on cartesan products If (X, d ) are metrc spaces for 1 n, then n Secton II4 of the lecture notes we defned three metrcs on X whose underlyng topologes are the product topology The purpose of

More information

1 Matrix representations of canonical matrices

1 Matrix representations of canonical matrices 1 Matrx representatons of canoncal matrces 2-d rotaton around the orgn: ( ) cos θ sn θ R 0 = sn θ cos θ 3-d rotaton around the x-axs: R x = 1 0 0 0 cos θ sn θ 0 sn θ cos θ 3-d rotaton around the y-axs:

More information

Convexity preserving interpolation by splines of arbitrary degree

Convexity preserving interpolation by splines of arbitrary degree Computer Scence Journal of Moldova, vol.18, no.1(52), 2010 Convexty preservng nterpolaton by splnes of arbtrary degree Igor Verlan Abstract In the present paper an algorthm of C 2 nterpolaton of dscrete

More information

Lecture 10 Support Vector Machines II

Lecture 10 Support Vector Machines II Lecture 10 Support Vector Machnes II 22 February 2016 Taylor B. Arnold Yale Statstcs STAT 365/665 1/28 Notes: Problem 3 s posted and due ths upcomng Frday There was an early bug n the fake-test data; fxed

More information

Stanford University CS359G: Graph Partitioning and Expanders Handout 4 Luca Trevisan January 13, 2011

Stanford University CS359G: Graph Partitioning and Expanders Handout 4 Luca Trevisan January 13, 2011 Stanford Unversty CS359G: Graph Parttonng and Expanders Handout 4 Luca Trevsan January 3, 0 Lecture 4 In whch we prove the dffcult drecton of Cheeger s nequalty. As n the past lectures, consder an undrected

More information

Simultaneous Optimization of Berth Allocation, Quay Crane Assignment and Quay Crane Scheduling Problems in Container Terminals

Simultaneous Optimization of Berth Allocation, Quay Crane Assignment and Quay Crane Scheduling Problems in Container Terminals Smultaneous Optmzaton of Berth Allocaton, Quay Crane Assgnment and Quay Crane Schedulng Problems n Contaner Termnals Necat Aras, Yavuz Türkoğulları, Z. Caner Taşkın, Kuban Altınel Abstract In ths work,

More information

Notes prepared by Prof Mrs) M.J. Gholba Class M.Sc Part(I) Information Technology

Notes prepared by Prof Mrs) M.J. Gholba Class M.Sc Part(I) Information Technology Inverse transformatons Generaton of random observatons from gven dstrbutons Assume that random numbers,,, are readly avalable, where each tself s a random varable whch s unformly dstrbuted over the range(,).

More information

The Feynman path integral

The Feynman path integral The Feynman path ntegral Aprl 3, 205 Hesenberg and Schrödnger pctures The Schrödnger wave functon places the tme dependence of a physcal system n the state, ψ, t, where the state s a vector n Hlbert space

More information

Lecture Notes on Linear Regression

Lecture Notes on Linear Regression Lecture Notes on Lnear Regresson Feng L fl@sdueducn Shandong Unversty, Chna Lnear Regresson Problem In regresson problem, we am at predct a contnuous target value gven an nput feature vector We assume

More information

3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X

3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X Statstcs 1: Probablty Theory II 37 3 EPECTATION OF SEVERAL RANDOM VARIABLES As n Probablty Theory I, the nterest n most stuatons les not on the actual dstrbuton of a random vector, but rather on a number

More information

Distributions /06. G.Serazzi 05/06 Dimensionamento degli Impianti Informatici distrib - 1

Distributions /06. G.Serazzi 05/06 Dimensionamento degli Impianti Informatici distrib - 1 Dstrbutons 8/03/06 /06 G.Serazz 05/06 Dmensonamento degl Impant Informatc dstrb - outlne densty, dstrbuton, moments unform dstrbuton Posson process, eponental dstrbuton Pareto functon densty and dstrbuton

More information

Lecture 3. Ax x i a i. i i

Lecture 3. Ax x i a i. i i 18.409 The Behavor of Algorthms n Practce 2/14/2 Lecturer: Dan Spelman Lecture 3 Scrbe: Arvnd Sankar 1 Largest sngular value In order to bound the condton number, we need an upper bound on the largest

More information

ELASTIC WAVE PROPAGATION IN A CONTINUOUS MEDIUM

ELASTIC WAVE PROPAGATION IN A CONTINUOUS MEDIUM ELASTIC WAVE PROPAGATION IN A CONTINUOUS MEDIUM An elastc wave s a deformaton of the body that travels throughout the body n all drectons. We can examne the deformaton over a perod of tme by fxng our look

More information

6. Stochastic processes (2)

6. Stochastic processes (2) 6. Stochastc processes () Lect6.ppt S-38.45 - Introducton to Teletraffc Theory Sprng 5 6. Stochastc processes () Contents Markov processes Brth-death processes 6. Stochastc processes () Markov process

More information

Lectures on Stochastic Stability. Sergey FOSS. Heriot-Watt University. Lecture 5. Monotonicity and Saturation Rule

Lectures on Stochastic Stability. Sergey FOSS. Heriot-Watt University. Lecture 5. Monotonicity and Saturation Rule Lectures on Stochastc Stablty Sergey FOSS Herot-Watt Unversty Lecture 5 Monotoncty and Saturaton Rule Introducton The paper of Loynes [8] was the frst to consder a system (sngle server queue, and, later,

More information

6. Stochastic processes (2)

6. Stochastic processes (2) Contents Markov processes Brth-death processes Lect6.ppt S-38.45 - Introducton to Teletraffc Theory Sprng 5 Markov process Consder a contnuous-tme and dscrete-state stochastc process X(t) wth state space

More information

This column is a continuation of our previous column

This column is a continuation of our previous column Comparson of Goodness of Ft Statstcs for Lnear Regresson, Part II The authors contnue ther dscusson of the correlaton coeffcent n developng a calbraton for quanttatve analyss. Jerome Workman Jr. and Howard

More information

Problem Set 9 Solutions

Problem Set 9 Solutions Desgn and Analyss of Algorthms May 4, 2015 Massachusetts Insttute of Technology 6.046J/18.410J Profs. Erk Demane, Srn Devadas, and Nancy Lynch Problem Set 9 Solutons Problem Set 9 Solutons Ths problem

More information

Formulas for the Determinant

Formulas for the Determinant page 224 224 CHAPTER 3 Determnants e t te t e 2t 38 A = e t 2te t e 2t e t te t 2e 2t 39 If 123 A = 345, 456 compute the matrx product A adj(a) What can you conclude about det(a)? For Problems 40 43, use

More information

CS-433: Simulation and Modeling Modeling and Probability Review

CS-433: Simulation and Modeling Modeling and Probability Review CS-433: Smulaton and Modelng Modelng and Probablty Revew Exercse 1. (Probablty of Smple Events) Exercse 1.1 The owner of a camera shop receves a shpment of fve cameras from a camera manufacturer. Unknown

More information

ANSWERS. Problem 1. and the moment generating function (mgf) by. defined for any real t. Use this to show that E( U) var( U)

ANSWERS. Problem 1. and the moment generating function (mgf) by. defined for any real t. Use this to show that E( U) var( U) Econ 413 Exam 13 H ANSWERS Settet er nndelt 9 deloppgaver, A,B,C, som alle anbefales å telle lkt for å gøre det ltt lettere å stå. Svar er gtt . Unfortunately, there s a prntng error n the hnt of

More information

Resource Allocation with a Budget Constraint for Computing Independent Tasks in the Cloud

Resource Allocation with a Budget Constraint for Computing Independent Tasks in the Cloud Resource Allocaton wth a Budget Constrant for Computng Independent Tasks n the Cloud Wemng Sh and Bo Hong School of Electrcal and Computer Engneerng Georga Insttute of Technology, USA 2nd IEEE Internatonal

More information

Introduction to Vapor/Liquid Equilibrium, part 2. Raoult s Law:

Introduction to Vapor/Liquid Equilibrium, part 2. Raoult s Law: CE304, Sprng 2004 Lecture 4 Introducton to Vapor/Lqud Equlbrum, part 2 Raoult s Law: The smplest model that allows us do VLE calculatons s obtaned when we assume that the vapor phase s an deal gas, and

More information

Problem Set 9 - Solutions Due: April 27, 2005

Problem Set 9 - Solutions Due: April 27, 2005 Problem Set - Solutons Due: Aprl 27, 2005. (a) Frst note that spam messages, nvtatons and other e-mal are all ndependent Posson processes, at rates pλ, qλ, and ( p q)λ. The event of the tme T at whch you

More information

LINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity

LINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity LINEAR REGRESSION ANALYSIS MODULE IX Lecture - 30 Multcollnearty Dr. Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur 2 Remedes for multcollnearty Varous technques have

More information

Inductance Calculation for Conductors of Arbitrary Shape

Inductance Calculation for Conductors of Arbitrary Shape CRYO/02/028 Aprl 5, 2002 Inductance Calculaton for Conductors of Arbtrary Shape L. Bottura Dstrbuton: Internal Summary In ths note we descrbe a method for the numercal calculaton of nductances among conductors

More information

Entropy of Markov Information Sources and Capacity of Discrete Input Constrained Channels (from Immink, Coding Techniques for Digital Recorders)

Entropy of Markov Information Sources and Capacity of Discrete Input Constrained Channels (from Immink, Coding Techniques for Digital Recorders) Entropy of Marov Informaton Sources and Capacty of Dscrete Input Constraned Channels (from Immn, Codng Technques for Dgtal Recorders). Entropy of Marov Chans We have already ntroduced the noton of entropy

More information

Errors for Linear Systems

Errors for Linear Systems Errors for Lnear Systems When we solve a lnear system Ax b we often do not know A and b exactly, but have only approxmatons  and ˆb avalable. Then the best thng we can do s to solve ˆx ˆb exactly whch

More information

BMAP/M/C Bulk Service Queue with Randomly Varying Environment

BMAP/M/C Bulk Service Queue with Randomly Varying Environment IOSR Journal of Engneerng (IOSRJEN) ISSN (e): 2250-3021, ISSN (p): 2278-8719 Vol. 05, Issue 03 (March. 2015), V1 PP 33-47 www.osrjen.org BMAP/M/C Bulk Servce Queue wth Randomly Varyng Envronment Rama.G,

More information

NON-CENTRAL 7-POINT FORMULA IN THE METHOD OF LINES FOR PARABOLIC AND BURGERS' EQUATIONS

NON-CENTRAL 7-POINT FORMULA IN THE METHOD OF LINES FOR PARABOLIC AND BURGERS' EQUATIONS IJRRAS 8 (3 September 011 www.arpapress.com/volumes/vol8issue3/ijrras_8_3_08.pdf NON-CENTRAL 7-POINT FORMULA IN THE METHOD OF LINES FOR PARABOLIC AND BURGERS' EQUATIONS H.O. Bakodah Dept. of Mathematc

More information

AGC Introduction

AGC Introduction . Introducton AGC 3 The prmary controller response to a load/generaton mbalance results n generaton adjustment so as to mantan load/generaton balance. However, due to droop, t also results n a non-zero

More information

The Order Relation and Trace Inequalities for. Hermitian Operators

The Order Relation and Trace Inequalities for. Hermitian Operators Internatonal Mathematcal Forum, Vol 3, 08, no, 507-57 HIKARI Ltd, wwwm-hkarcom https://doorg/0988/mf088055 The Order Relaton and Trace Inequaltes for Hermtan Operators Y Huang School of Informaton Scence

More information

Transient results for M/M/1/c queues via path counting. M. Hlynka*, L.M. Hurajt and M. Cylwa

Transient results for M/M/1/c queues via path counting. M. Hlynka*, L.M. Hurajt and M. Cylwa Int. J. Mathematcs n Operatonal Research, Vol. X, No. X, xxxx 1 Transent results for M/M/1/c queues va path countng M. Hlynka*, L.M. Hurajt and M. Cylwa Department of Mathematcs and Statstcs, Unversty

More information

Simulated Power of the Discrete Cramér-von Mises Goodness-of-Fit Tests

Simulated Power of the Discrete Cramér-von Mises Goodness-of-Fit Tests Smulated of the Cramér-von Mses Goodness-of-Ft Tests Steele, M., Chaselng, J. and 3 Hurst, C. School of Mathematcal and Physcal Scences, James Cook Unversty, Australan School of Envronmental Studes, Grffth

More information

Assortment Optimization under MNL

Assortment Optimization under MNL Assortment Optmzaton under MNL Haotan Song Aprl 30, 2017 1 Introducton The assortment optmzaton problem ams to fnd the revenue-maxmzng assortment of products to offer when the prces of products are fxed.

More information

Lecture 3: Probability Distributions

Lecture 3: Probability Distributions Lecture 3: Probablty Dstrbutons Random Varables Let us begn by defnng a sample space as a set of outcomes from an experment. We denote ths by S. A random varable s a functon whch maps outcomes nto the

More information

Convergence of random processes

Convergence of random processes DS-GA 12 Lecture notes 6 Fall 216 Convergence of random processes 1 Introducton In these notes we study convergence of dscrete random processes. Ths allows to characterze phenomena such as the law of large

More information

Continuous Time Markov Chain

Continuous Time Markov Chain Contnuous Tme Markov Chan Hu Jn Department of Electroncs and Communcaton Engneerng Hanyang Unversty ERICA Campus Contents Contnuous tme Markov Chan (CTMC) Propertes of sojourn tme Relatons Transton probablty

More information

Assignment 5. Simulation for Logistics. Monti, N.E. Yunita, T.

Assignment 5. Simulation for Logistics. Monti, N.E. Yunita, T. Assgnment 5 Smulaton for Logstcs Mont, N.E. Yunta, T. November 26, 2007 1. Smulaton Desgn The frst objectve of ths assgnment s to derve a 90% two-sded Confdence Interval (CI) for the average watng tme

More information

Min Cut, Fast Cut, Polynomial Identities

Min Cut, Fast Cut, Polynomial Identities Randomzed Algorthms, Summer 016 Mn Cut, Fast Cut, Polynomal Identtes Instructor: Thomas Kesselhem and Kurt Mehlhorn 1 Mn Cuts n Graphs Lecture (5 pages) Throughout ths secton, G = (V, E) s a mult-graph.

More information

Power law and dimension of the maximum value for belief distribution with the max Deng entropy

Power law and dimension of the maximum value for belief distribution with the max Deng entropy Power law and dmenson of the maxmum value for belef dstrbuton wth the max Deng entropy Bngy Kang a, a College of Informaton Engneerng, Northwest A&F Unversty, Yanglng, Shaanx, 712100, Chna. Abstract Deng

More information

MODELING TRAFFIC LIGHTS IN INTERSECTION USING PETRI NETS

MODELING TRAFFIC LIGHTS IN INTERSECTION USING PETRI NETS The 3 rd Internatonal Conference on Mathematcs and Statstcs (ICoMS-3) Insttut Pertanan Bogor, Indonesa, 5-6 August 28 MODELING TRAFFIC LIGHTS IN INTERSECTION USING PETRI NETS 1 Deky Adzkya and 2 Subono

More information

Analysis of Queuing Delay in Multimedia Gateway Call Routing

Analysis of Queuing Delay in Multimedia Gateway Call Routing Analyss of Queung Delay n Multmeda ateway Call Routng Qwe Huang UTtarcom Inc, 33 Wood Ave. outh Iseln, NJ 08830, U..A Errol Lloyd Computer Informaton cences Department, Unv. of Delaware, Newark, DE 976,

More information

The (Q, r) Inventory Policy in Production- Inventory Systems

The (Q, r) Inventory Policy in Production- Inventory Systems Proceedngs of SKISE Fall Conference, Nov 14-15, 2008 The ( Inventory Polcy n Producton- Inventory Systems Joon-Seok Km Sejong Unversty Seoul, Korea Abstract We examne the effectveness of the conventonal

More information

Turing Machines (intro)

Turing Machines (intro) CHAPTER 3 The Church-Turng Thess Contents Turng Machnes defntons, examples, Turng-recognzable and Turng-decdable languages Varants of Turng Machne Multtape Turng machnes, non-determnstc Turng Machnes,

More information

Integrals and Invariants of Euler-Lagrange Equations

Integrals and Invariants of Euler-Lagrange Equations Lecture 16 Integrals and Invarants of Euler-Lagrange Equatons ME 256 at the Indan Insttute of Scence, Bengaluru Varatonal Methods and Structural Optmzaton G. K. Ananthasuresh Professor, Mechancal Engneerng,

More information

LOW BIAS INTEGRATED PATH ESTIMATORS. James M. Calvin

LOW BIAS INTEGRATED PATH ESTIMATORS. James M. Calvin Proceedngs of the 007 Wnter Smulaton Conference S G Henderson, B Bller, M-H Hseh, J Shortle, J D Tew, and R R Barton, eds LOW BIAS INTEGRATED PATH ESTIMATORS James M Calvn Department of Computer Scence

More information

Numerical Solution of Ordinary Differential Equations

Numerical Solution of Ordinary Differential Equations Numercal Methods (CENG 00) CHAPTER-VI Numercal Soluton of Ordnar Dfferental Equatons 6 Introducton Dfferental equatons are equatons composed of an unknown functon and ts dervatves The followng are examples

More information

Comparison of Regression Lines

Comparison of Regression Lines STATGRAPHICS Rev. 9/13/2013 Comparson of Regresson Lnes Summary... 1 Data Input... 3 Analyss Summary... 4 Plot of Ftted Model... 6 Condtonal Sums of Squares... 6 Analyss Optons... 7 Forecasts... 8 Confdence

More information

A note on almost sure behavior of randomly weighted sums of φ-mixing random variables with φ-mixing weights

A note on almost sure behavior of randomly weighted sums of φ-mixing random variables with φ-mixing weights ACTA ET COMMENTATIONES UNIVERSITATIS TARTUENSIS DE MATHEMATICA Volume 7, Number 2, December 203 Avalable onlne at http://acutm.math.ut.ee A note on almost sure behavor of randomly weghted sums of φ-mxng

More information

Transfer Functions. Convenient representation of a linear, dynamic model. A transfer function (TF) relates one input and one output: ( ) system

Transfer Functions. Convenient representation of a linear, dynamic model. A transfer function (TF) relates one input and one output: ( ) system Transfer Functons Convenent representaton of a lnear, dynamc model. A transfer functon (TF) relates one nput and one output: x t X s y t system Y s The followng termnology s used: x y nput output forcng

More information

A new Approach for Solving Linear Ordinary Differential Equations

A new Approach for Solving Linear Ordinary Differential Equations , ISSN 974-57X (Onlne), ISSN 974-5718 (Prnt), Vol. ; Issue No. 1; Year 14, Copyrght 13-14 by CESER PUBLICATIONS A new Approach for Solvng Lnear Ordnary Dfferental Equatons Fawz Abdelwahd Department of

More information

On the Throughput of Clustered Photolithography Tools:

On the Throughput of Clustered Photolithography Tools: On the hroughput of lustered Photolthography ools: Wafer Advancement and Intrnsc Equpment Loss Maruth Kumar Mutnur James R. Morrson, Ph.D. September 23, 2007 Presentaton Outlne Motvaton Model : Synchronous

More information

Probability and Random Variable Primer

Probability and Random Variable Primer B. Maddah ENMG 622 Smulaton 2/22/ Probablty and Random Varable Prmer Sample space and Events Suppose that an eperment wth an uncertan outcome s performed (e.g., rollng a de). Whle the outcome of the eperment

More information

Section 8.3 Polar Form of Complex Numbers

Section 8.3 Polar Form of Complex Numbers 80 Chapter 8 Secton 8 Polar Form of Complex Numbers From prevous classes, you may have encountered magnary numbers the square roots of negatve numbers and, more generally, complex numbers whch are the

More information

Error Probability for M Signals

Error Probability for M Signals Chapter 3 rror Probablty for M Sgnals In ths chapter we dscuss the error probablty n decdng whch of M sgnals was transmtted over an arbtrary channel. We assume the sgnals are represented by a set of orthonormal

More information