Single-machine scheduling with trade-off between number of tardy jobs and compression cost

Size: px
Start display at page:

Download "Single-machine scheduling with trade-off between number of tardy jobs and compression cost"

Transcription

1 Ths s the Pre-Publshed Verson. Sngle-machne schedulng wth trade-off between number of tardy jobs and compresson cost 1, 2, Yong He 1 Department of Mathematcs, Zhejang Unversty, Hangzhou , P.R. Chna 2 State Key Lab of CAD & CG, Zhejang Unversty, Hangzhou , P.R. Chna Q We 3 3 Nngbo Insttute of Technology, Zhejang Unversty, Nngbo , P.R. Chna T. C. E. Cheng 4, 4 Department of Logstcs, The Hong Kong Polytechnc Unversty, Kowloon, Hong Kong Abstract We consder a sngle-machne schedulng problem n whch the job processng tmes are controllable or compressble. The performance crtera are the compresson cost and the number of tardy jobs. For the problem where no tardy jobs are allowed and the objectve s to mnmze the total compresson cost, we present a strongly polynomal tme algorthm. For the problem to construct the trade-off curve between the number of tardy jobs and the maxmum compresson cost, we present a polynomal tme algorthm. Furthermore, we extend the problem to the case of dscrete controllable processng tmes where the processng Ths research was supported by the Teachng and Research Award Program for Outstandng Young Teachers n Hgher Educaton Insttutons of the MOE, Chna, and the Natonal Natural Scence Foundaton of Chna ( ). The thrd author was supported n part by The Hong Kong Polytechnc Unversty under a grant from the ASD n Chna Busness Servces. Emal: mathhey@zju.edu.cn Emal: weq@nt.zju.edu.cn Correspondng author; emal: LGTcheng@polyu.edu.hk 1

2 tme of a job can only take one of some gven dscrete values. We show that even some specal cases of the dscrete controllable verson wth the objectve of mnmzng the total compresson cost are NP-hard, but the general case s solvable n pseudo-polynomal tme. Moreover, we present a strongly polynomal tme algorthm to construct the trade-off curve between the number of tardy jobs and the maxmum compresson cost for the dscrete controllable verson. 1 Introducton Most classcal schedulng problems assume that the job processng tmes are fxed,.e., they are determned a pror and cannot be modfed by the scheduler. However, n real applcatons the processng of a job often requres resources, such as manpower, facltes, funds, raw materals, etc. By puttng more resources nto job processng, shorter job processng tmes may be accomplshed. Hence, t s possble to compress jobs (control the job processng tmes) by ncurrng extra costs. Schedulng problems wth controllable processng tmes have receved much attenton from researchers n the last two decades. Work n ths area was ntated by Vckson [11, 12] and Van Wassenhove and Baker [13]. For the related work on machne schedulng problems wth controllable processng tmes, the reader s referred to the survey by Nowck and Zdrzalka [9]. In ths paper we consder a sngle-machne model of jont job sequencng and resource allocaton wth the sequencng crteron beng the number of tardy jobs, whch was frst proposed by Danels and Sarn [5]. Formally, ths problem can be formulated as follows. We are gven a set J = {J 1, J 2,, J n } of ndependent jobs, whch must be scheduled on a sngle machne. The processng requrement of a job J s specfed by four non-negatve parameters p, d, u and w. Here p denotes the normal (ntal) processng tme of J, d denotes the due-date of J, u p s an upper bound on the compressblty of J, and w s the cost for unt-tme compresson of J, called the unt-compresson cost. By allocatng addtonal resources to the processng of J, the actual processng tme of J may be compressed from p to p = p x, where x [0, u ] s the amount of compresson. The cost of performng ths compresson equals w x. Let C denote the completon tme of job J under a gven schedule. Defne U = 0 f job J s early (C < d ) or on-tme (C = d ), and U = 1 f t s tardy (C > d ). We assume that all the jobs are avalable at tme zero. Moreover, let T CC = Σ n w x be the total compresson cost, and n T = Σ n U the number of tardy jobs. Then the objectve s to construct the trade-off curve between n T and T CC. We denote ths problem by P 0 : 1 contr, n T k T CC n ths paper. 2

3 For ths problem, Danels and Sarn [5] provded some theoretcal results. Cheng, Chen and L [2] proved that the problem s NP-hard, and presented a pseudo-polynomal tme algorthm. Ths paper s an extenson of the above work wth three man contrbutons. Frst, we wll resolve the specal case n T = 0 of the problem by presentng an optmal algorthm. Its tme complexty s O(n 2 ), and becomes O(n log n) f all w are the same. Ths problem, denoted by P 1 : 1 contr, n T = 0 T CC, models the followng scenaro. For some applcatons n logstcs and supply chan management, jobs denote orders from customers, and the machne denotes the manufacturer. In some cases, whle tardy jobs are allowed, havng tardy jobs may damage the goodwll of the customers, so the manufacturer s concerned wth the trade-off between n T and T CC. In other cases, customers wll accept no tardy jobs. So the manufacturer must fnd a soluton to mnmze the compresson cost wth the constrant that n T = 0. Note that whle the algorthm gven n [2] stll requres pseudo-polynomal tme for n T = 0, we present a strongly polynomal tme algorthm for ths case. Second, we consder the problem wth a new objectve. The objectve s to construct the trade-off curve between n T and the bottleneck objectve functon MCC = max,,n w x,.e., the maxmum compresson cost. Ths problem, denoted by P 2 : 1 contr, n T k MCC, models the followng stuaton: Gven that the resource s lmted, the decson maker (scheduler) seeks to fnd a soluton that balances the amount of resource used by each job under the constrant that the number of tardy jobs s no greater than a gven nonnegatve nteger k. We wll show that ths problem can be solved n O(n 2 log W ) tme, where W = max,,n w u. Thrd, we consder the problems wth dscrete controllable processng tmes. Schedulng problems wth dscrete controllable processng tmes have been studed by Chen, Lu and Tang [1], De et al. [3], [4], and Skutella [10], whch have many real-world applcatons. For ths stuaton, the allowed compresson amount x of job J s n a fnte set,.e., t must be one of some gven dscrete values, nstead of any value n the nterval [0, u ]. Hence, we assume that for each J J, the actual processng tme p = p x has l possble values a 1 = p > a 2 > > a l. Let w j be the compresson cost for the processng tme a j, = 1,, n, j = 1,, l. It s assumed that 0 = w 1 < w 2 < < w l. Ths assumpton s reasonable because to acheve smaller processng tmes requres more resources, hence ncurrng hgher costs [1]. We show that both P 3 : 1 dsc contr, n T k T CC and P 4 : 1 dsc contr, n T = 0 T CC are NP-hard even for a very specal case where all the jobs have the same due-date, all l = 2 and all unt-compresson costs w 2 /a 2, = 1,, n, are the same (.e., all the unt-compresson costs are the same for all the jobs), and present pseudo-polynomal tme algorthms based on dynamc programmng for the general case of P 3 and P 4. Moreover, we wll present a strongly polynomal tme algorthm 3

4 for the problem P 5 : 1 dsc contr, n T k MCC. The tme complexty s O(n 2 log(nl)), where l = max,,n l. Ths paper s organzed as follows: Sectons 2 and 3 consder the problems P 1 and P 2, respectvely, Secton 4 studes the problems P 3 and P 4, and Secton 5 studes the problem P 5. Fnal remarks are presented n Secton 6. 2 Strongly polynomal solvablty of the problem P 1 Defnton 2.1 For the problem P 1, a soluton s called feasble f t satsfes Σ n U = 0. Lemma 2.2 There exsts an optmal soluton such that all the jobs are scheduled n the Earlest Due-Date (EDD) order. Proof. It can be shown by applyng the standard parwse job nterchange argument. Hence, n the remander of ths secton, we assume that the jobs are re-ndexed such that d 1 d 2 d n. Snce u s the upper bound on the compressblty of job J, = 1, 2,, n, we have Corollary 2.3 The problem P 1 has a feasble soluton satsfyng the EDD order f and only f Σ j=1 (p j u j ) d, = 1, 2,, n. For each job J, = 1, 2,, n, let the actual compresson be x. Then job J s not tardy f and only f Σ j=1 (p j x j ) d,.e., Σ j=1 x j Σ j=1 p j d. Thus, the problem P 1 can be formulated as the followng lnear program: mn Σ n w x s.t. Σ j=1 x j Σ j=1 p j d, = 1, 2, n, 0 x u, = 1, 2,, n. It s clear that the lnear program can be solved n polynomal tme, so can P 1. In the followng, we present a strongly polynomal tme algorthm for ths problem. Algorthm A 1 : 1. Renumber all the jobs n the EDD order such that d 1 d 2 d n. Set x = 0, u = u, = 1, 2,, n, and k = 1. 4

5 2. Schedule the jobs n the order of J 1, J 2,, J n wth the actual processng tmes p 1 x 1, p 2 x 2,, p n x n, respectvely. If there are no tardy jobs after processng all the jobs, then go to 4; otherwse let J tk be the frst tardy job wth completon tme C tk, go to Assume that the unt-compresson weghts of the job set {J 1, J 2,, J tk } satsfy w j1 w j2 w jtk (that s to say, J j s the job wth the -th smallest unt-compresson cost n {J 1, J 2,, J tk }). Note that the current bounds on the compressblty of these jobs are u 1, u 2,, u t k, respectvely. Let s = mn{ Σ p=1 u j p > C tk d tk, 1 t k }. Then for = j 1, j 2,, j s 1, set x x +u and u 0; for = j s, set x x +C tk d tk Σ s 1 p=1 u j p and u u x. Set k k + 1, go to Output x, = 1, 2,, n, as the fnal compresson of job J, and the objectve value Σ n w x, stop. Lemma 2.4 Consder the followng lnear program: mn Σ t w y s.t. Σ t y L, (1) 0 y l, = 1, 2,, t. where L s a postve number and 0 w j1 w j2 w jt. Let s = mn{ Σ p=1 l j > L, 1 j t}. Then s an optmal soluton. l j, f 1 s 1, y j = L Σ s 1 l j, f = s, 0, f s + 1 t Proof. It s a contnuous relaxaton of the mnmzaton verson of a specal bounded Knapsack problem [8], hence the lemma holds. Theorem 2.5 Algorthm A 1 produces an optmal soluton to the problem P 1, and runs n tme O(n 2 ). Proof. Frst, t s clear that the soluton produced by algorthm A 1 s feasble. We next prove ts optmalty. 5

6 Let x k and x denote the compresson of job J, = 1, 2,, n, rght after the k-th teraton of the algorthm, and the compresson of job J n an optmal soluton satsfyng the EDD order, respectvely. We prove by nducton on k that Σ t k w x Σt k w x k, and Σt k x k s exactly the mnmum possble total compresson to assure no jobs n {J 1, J 2,, J tk } are tardy, and thus Σ t k x Σ t k x k for every k 1. It states that the algorthm yelds a soluton that has the objectve value and the total compresson no greater than those of the optmal soluton, and thus s optmal. For k = 1, n order to guarantee that job J t1 s not tardy, Σ t 1 x C t1 d t1 must hold. From Lemma 2.4, we conclude that x 1 1, x1 2,, x1 t 1 s an optmal soluton to the followng lnear program: mn Σ t 1 w y s.t. Σ t 1 y C t1 d t1, (2) 0 y u, = 1, 2,, t 1. On the other hand, x 1, x 2,, x t 1 s a feasble soluton to (2), hence Σ t 1 w x Σ t 1 w x 1. Because Σ t 1 x 1 = C t 1 d t1, we know that Σ t 1 x 1 s the mnmum possble total compresson such that job J t1 s not tardy. Hence, the result s true for k = 1. In general, suppose that the result s true for k 1, that s, Σ t k 1 w x Σt k 1 w x k 1 and Σ t k 1 x Σ t k 1 xk 1, and no jobs n {J 1, J 2,, J tk 1 } are tardy wth the processng tmes p x k 1, = 1, 2,, t k 1. Snce Σ t k 1 xk 1 s the mnmum possble total compresson that guarantees that no jobs n {J 1, J 2,, J tk 1 } are tardy, to make job J tk not tardy, the new compresson must be at least C tk d tk. From the algorthm, we know that Σ t k x k = Σ t k 1 xk 1 + C tk d tk. It mples that Σ t k x k s exactly the mnmum possble total compresson to ensure that no jobs n {J 1, J 2,, J tk } are tardy, and thus Σ t k x Σ t k x k. (3) To show that Σ t k w x Σ t k w x k, we dvde the compressons x, = 1, 2,, t k nto two parts x and x, and prove the result by verfyng that the total cost of the frst parts of all jobs s no less than that of the compressons x k 1 1, x k 1 2,, x k 1 t k 1, and the cost of the second parts of all jobs s no less than that of the new compressons n the k-th teraton. = x k 1 xk 1 1, x k 2 determned such that: x k 1. = x k 2 xk 1 2,, x k. t k = x k tk x k 1. Frst, x, = 1, 2,, t k 1 are 6 t k

7 ) If x xk 1, then x x xk 1, f x < xk 1, then x = x. ) If the processng tme of job J, = 1, 2,, t k 1 s p = p x, then the jobs J 1, J 2,, J tk 1 are all not tardy by the EDD rule, and Σ t k 1 x = Σ t k 1 xk 1. (4) Furthermore, let x = 0, = t k 1 + 1,, t k. Thus, let the second part of x 1, 2,, t k. be x = x x, = In fact, ths constructon can be performed as follows: Let z = mn{x, xk 1 }, = 1, 2,, t k 1 and T = Σ t k 1 xk 1 Σ t k 1 z. Let v = mn{ Σ j=1 (x j z j) > T, 1 t k 1 }. We then defne x, = 1, 2,, t k 1 as follows (note that x = 0 as above, = t k 1 + 1,, t k ): x, for 1 v 1, x = z + T Σ v 1 (x z ), for = v, (5) z, for v < t k 1. Obvously, x, = 1, 2,, t k 1 satsfy ) and (4). Hence, we only need to prove that no jobs n {J 1, J 2,, J tk 1 } wth processng tmes p = p x are tardy. We show ths result by contradcton. Suppose that there s a tardy job. From the algorthm, we know that job J tk 1 s an on-tme job after the k 1-th teraton n the algorthm wth processng tmes p x k 1, = 1,, t k 1, thus d tk 1 = Σ t k 1 (p x k 1 ). (6) Snce Σ t k 1 x = Σt k 1 xk 1, (6) mples d tk 1 = Σ t k 1 (p x ), and J t k 1 s not tardy wth the processng tmes p, = 1,, t k 1. Thus, let J tj, j k 2 be the frst tardy job accordng to the order from J tk 1 to J 1 wth processng tmes p, = 1,, t k 1. If t j < v, then from the constructon of x we conclude that Σt j (p x ) = Σt j (p x ) d t j, whch mples that J tj s not tardy, a contradcton. Thus, t j v. From the algorthm, we know that job J tj s an on-tme job rght after the j-th teraton wth the processng tmes p x j, = 1,, t j, thus d tj = Σ t j (p x j ). Hence, combnng ths wth (6), we have d tk 1 d tj = Σ t k 1 (p x k 1 ) Σ t j (p x j ) = Σ t k 1 =t j +1 p Σ t j (xk 1 x j ) Σt k 1 =t j +1 xk 1 Σ t k 1 =t j +1 p Σ t k 1 =t j +1 xk 1 Σ t k 1 =t j +1 p Σ t k 1 =t j +1 z (snce j < k 1 mples x k 1 x j 0 ) (snce z = mn{x, xk 1 } x k 1, = 1, 2,, t k 1 ) = Σ t k 1 =t j +1 p Σ t k 1 =t j +1 x. (by (5)) (7) 7

8 Because of the assumpton that job J tj s tardy wth the processng tmes p, = 1,, t k 1, we have Σ t j (p x ) > d t j. Combnng ths wth (7), we obtan Σ t k (p x ) > d t j + Σ t k 1 =t j +1 (p x ) > d t k 1. Thus J tk 1 s tardy wth the processng tmes p, = 1,, t k 1, a contradcton. Therefore, we conclude that no jobs n {J 1, J 2,, J tk 1 } are tardy. Now we return to the proof of the theorem. On the one hand, the compressons x 1, x 2,, x t k 1 make the jobs J 1, J 2,, J tk 1 not tardy; hence, from the nducton assumpton, we know that Σ t k 1 w x Σ t k 1 wk 1 x k 1. (8) On the other hand, subtractng (4) from (3), we get Σ t k x Σ t k x k = C tk d tk. Because satsfes ), we have x By defnton, we know 0 x = x x u x u x k 1, = 1, 2,, t k 1. 0 x = x u, = t k 1 + 1,, t k. Therefore, x, = 1, 2,, t k, s a feasble soluton to the followng lnear program: mn Σ t k w y s.t. Σ t k y C tk d tk, 0 y u x k 1, = 1, 2,, t k 1, (9) 0 y u, = t k 1 + 1, t k. From Lemma 2.4 and the algorthm, we conclude that x k, = 1, 2,, t k s an optmal soluton to the above lnear program (9). Thus, we obtan Σ t k w x k Σ t k w x. (10) Addng (8) and (10), we get Σ t k w x k Σt k w x. Thus, the soluton produced by algorthm A 1 s an optmal soluton to the problem P 1. We next study the tme complexty of algorthm A 1. It s clear that Step 1 takes O(n log n) tme. Steps 2-3 may terate at most n tmes and each teraton takes O(n) tme. Hence, algorthm A 1 runs n O(n 2 ) n the worst case and s a strongly polynomal algorthm. If all unt-compresson costs are the same,.e., w = w, = 1,, n, t can be shown smlarly that the followng smplfed verson of A 1 yelds an optmal soluton n tme O(n log n). 8

9 Algorthm A 1 : 1. Renumber the jobs n the EDD order such that d 1 d 2 d n. 2. Compute r such that Σ r p d r = max{σ j p d j j = 1, 2,, n}. 3. Let s = mn{j Σ j u > Σ r p d r, 1 j n}. Then, for 1 s 1, let x = u ; for = s, let x = Σ r p d r Σ s 1 u ; and for s + 1 n, let x = Output x 1, x 2,, x n and Σ n w x, stop. 3 Polynomal solvablty of the problem P 2 To solve the problem P 2, we assume that all w x, = 1, 2,, n are ntegers, thus the objectve value max{w x, = 1, 2,, n} s also an nteger. Ths assumpton should not be a serous lmtaton [2], snce the amount of allocaton can always be expressed n the smallest unt of resource for all practcal purposes, whch makes the costs ntegers. Let max{w u, = 1, 2,, n} = W. It s well-known that Moore s algorthm can solve the problem 1 n T n O(n 2 ) tme [7]. In the followng we present an optmal algorthm for P 2 by combnng Moore s algorthm wth the bsecton method. Defnton 3.1 Consder an nstance of the problem 1 n T wth the processng tmes {p = 1, 2,, n}. If ts optmal objectve value s n T k, then we say that {p = 1, 2,, n} s a feasble soluton to the problem P 2. The problem P 2 s called feasble f t has at least one feasble soluton. In the remander of ths secton, we use p t to denote the set consstng of processng tmes p mn{t/w, u }, = 1, 2,, n, where p s the ntal processng tme of job J. Specfcally, p 0 = {p = p = 1, 2,, n} and p W = {p = p u = 1, 2,, n}. The followng result s trval. Lemma 3.2 (1) Let {p = 1, 2,, n} be a feasble soluton to the problem P 2. If p p, = 1, 2,, n, then {p = 1, 2,, n} s a feasble soluton to the problem P 2, too. (2) The problem P 2 has a feasble soluton f and only f p W s a feasble soluton. Algorthm A 2 : 9

10 1. Invoke Moore s algorthm to solve the nstance of 1 n T wth actual processng tmes p 0. If n T k, then the current soluton s optmal wth the objectve value MCC = 0, stop; otherwse, go to Invoke Moore s algorthm to solve the nstance of 1 n T wth actual processng tmes p W. If n T > k, then output that the problem P 2 has no feasble soluton, stop; Otherwse, let t = W and t = If t t = 1, then output that the soluton p t s optmal wth the objectve value MCC = t, stop; otherwse set l = (t + t )/2, and nvoke Moore s algorthm to solve the nstance of 1 n T wth actual processng tmes p l. If n T k, then set t = l and go back to 3; otherwse, set t = l and go to 3. Theorem 3.3 If the problem P 2 s feasble, then the soluton obtaned by algorthm A 2 s an optmal soluton, and the tme complexty of algorthm A 2 s O(n 2 log W ). Proof. If the algorthm stops at Step 2, then p W s not a feasble soluton, and hence by Lemma 3.2(2), the problem s nfeasble. We now consder the case that the problem P 2 s feasble. If the algorthm stops at Step 1, then a soluton p 0 s obtaned wth the objectve value 0, whch s trvally an optmal soluton. Hence, we suppose n the followng that the optmal objectve value s not 0. Then we can clam that p W s a feasble soluton, whereas p 0 s not. So, by the bsecton procedure, algorthm A 2 can get t and t such that t t = 1, p t s feasble, but p t = p t 1 s not. Thus algorthm A 2 outputs a feasble soluton p t wth the objectve value MCC = t when t stops. Let {p = 1, 2,, n} be any feasble soluton of the problem P 2 wth the objectve value MCC = max{(p p )w = 1, 2,, n}. We now prove by contradcton that MCC = t MCC and hence p t must be the optmal soluton. Suppose MCC > MCC. As there s no nteger between t 1 and t, MCC > MCC mples that t 1 MCC. So, by Lemma 3.3(1), p t 1 s also feasble, However t = t 1, and we know that p t s not feasble, a contradcton. Thus, MCC > MCC s not true and p t s an optmal soluton to the problem P 2. Because Step 3 may repeat for at most log W tmes and the tme complexty of Moore s algorthm s O(n 2 ), we conclude that the tme complexty of algorthm A 2 s O(n 2 log W ). 10

11 4 NP-hardness of the problems P 3 and P 4 Cheng, Chen and L [2] proved that the problem P 0 : 1 contr, n T k T CC s NP-hard. Now we dscuss the problem P 3 where the possble compressons of each job are some dscrete values. Theorem 4.1 The problem P 3 s NP hard even f all the jobs have the same due-date, all l = 2, and all the unt-compresson costs are the same. Proof. Snce all the unt-compresson costs are the same, the objectve value of the problem P 3 s equvalent to the total compresson. We show the result by reducng the Partton problem [6] to ths problem. Gven an nstance I of the Partton problem wth a set of postve ntegers {h 1, h 2, h n } and 2B = Σ n h, we construct an nstance II of the problem P 4 as follows: Assocated wth each h, = 1,, n, s job J wth p = h, p = p x {h, 0}, d = B, = 1, 2,, n. In addton, we construct n more jobs J n+1,, J 2n wth p = 2B, p = p x {2B, 0}, d = B, = n + 1, n + 2,, 2n. Defne k = n and threshold UB = B. We prove that nstance I has a soluton f and only f nstance II has a soluton wth n T k and the objectve s no greater than UB. If I has a soluton, then there exst two subsets H 1 and H 2 of H such that H 1 H 2 = H, H 1 H 2 = and h H 1 h = h H 2 h = B. Construct a soluton for nstance II as follows: Let the compresson of J, H 1, be x = h for each H 1, and x = 0 for all other jobs. Then the frst n jobs can be completed early or on-tme whle the last n jobs are tardy. Hence, the number of tardy jobs s k = n, and the total compresson s exactly B. Next, suppose II has a soluton such that n T k = n and the objectve value (.e., the total compresson) s no greater than UB. If a job J, n + 1 2n, s compressed, then the total compresson s at least 2B > UB, a contradcton. Thus, none of the last n jobs can be compressed, and all of them are tardy. That s to say, the frst n jobs must be completed early or on tme. If the total compresson of the jobs J 1, J 2,, J n s less than B, then ther total actual processng tme s more than 2B B = B, hence there exsts at least a tardy job, a contradcton. Snce UB = B, the total compresson of the jobs J 1, J 2,, J n s exactly B. It follows that there exsts a subset H {1, 2,, n} such that Σ H a = B. We have shown that 1 contr, n T = 0 T CC s strongly polynomal tme solvable. However, the dscrete controllable case becomes N P -hard. 11

12 Theorem 4.2 The problem P 4 s NP hard even f all the jobs have the same due-date, all l = 2, and all the unt-compresson costs are the same. Proof. Smlarly, all the unt-compresson costs beng the same, the objectve value of the problem P 4 s equvalent to the total compresson. We agan show the result by reducng the Partton problem to ths problem. Gven an nstance I of the Partton problem wth a set of postve ntegers {h 1, h 2, h n } and 2B = Σ n h, we construct an nstance II of the problem P 3 as follows: Assocated wth each h, = 1,, n, s job J wth p = 2h, p = p x {2h, 3h /2}, d = 7B/2, = 1, 2,, n. Defne the threshold UB = B/2. It can easly be shown that nstance I has a soluton f and only f nstance II has a soluton wth n T = 0 and the objectve value s no greater than UB. Next we present a pseudo-polynomal tme algorthm that solves the general case of the problem P 3 optmally. Note that f the problem P 3 s feasble (.e., the number of tardy jobs s no greater than k), there must exst an optmal soluton such that the early and on-tme jobs are scheduled n the EDD order, and the tardy jobs are scheduled n any order followng all the early and on-tme jobs. Furthermore, the tardy jobs are not compressed snce we are to mnmze the compresson cost. Hence, we assume that d 1 d 2 d n. Algorthm A 3 : 1. Let f(, t, q) be the mnmum total cost of a partal soluton contanng the frst jobs, J 1, J 2,, J, gven that the completon tme of the early and on-tme jobs n ths partal soluton s exactly t, and the number of tardy jobs s exactly q (1 n, 0 t d n, 0 q k). 2. Recursve relatons: For = 2, 3,, n, t = 0, 1,, d n and q = 0, 1,, k: mn{f( 1, t, q 1), mn{f( 1, t a j, q) + w j, = 1, 2,, l }}, f t d 1, mn{f( 1, t a j, q) + w j, = 1, 2,, l }, f(, t, q) = f d 1 < t d,, f t > d. { p a j, f f(, t, q) = f( 1, t a j, q) + w j, x = 0, otherwse. 12

13 3. Intal values: For t = 0, 1,, d n : { w1j, f d 1 t = a 1j, f(1, t, 0) =, otherwse, x 1 = f(1, t, 1) = 0, t = 0, 1,, d n. { p1 a j, f f(1, t, 0) = w 1j, 0, otherwse. 4. An optmal soluton can be obtaned by computng mn{f(n, t, 0), f(n, t, 1), f(n, t, k) t = 0,, d n }. Remark 4.3 Let l = max,2,,n l, the tme complexty of algorthm A 2 s O(nlkd n + n log n) snce we try all possble values of ( = 1,, n), all possble values of t (t = 0, 1,, d n ), and all possble values of q (q = 0, 1,, k), and the computaton of f(, t, q) needs O(l) tme for each possble state (, t, q). In addton, sortng jobs n non-decreasng order of due-date takes O(n log n) tme. Therefore, algorthm A 3 s pseudo-polynomal tme, mplyng that the problem P 3 s only NP-hard n the ordnary sense. Remark 4.4 Algorthm A 3 can also be used to solve P 4 by keepng q = k = 0. Hence, the tme complexty of the algorthm becomes O(nld n +n log n), and P 4 s NP -hard n the ordnary sense. 5 Strongly polynomal solvablty of the problem P 5 The man dea of the algorthm for P 5 s smlar to that for P 2. However, we can construct a strongly polynomal tme algorthm by makng mnor modfcatons. Before we present the algorthm, let w 0 = 0, and re-arrange the compresson costs of all the jobs n non-ncreasng order of ther values. Then we express ther dfferent values as follows: 0 = w 0 < w 1 < w 2 < < w L. Defnton 5.1 For a gven postve number S, defne j = max{j w j S, j = 1, 2,, l } for every = 1,, n, and let p S = {p = a j = 1, 2,, n}. Specfcally, p w 0 = {p = a 1 = p = 1, 2,, n} and p w L = {p = a 1 = p = 1, 2,, n}. Algorthm A 4 : 1. Invoke Moore s algorthm to compute the objectve value of the nstance of 1 n T wth actual processng tmes p w 0. If the objectve value n T k, then output MCC = 0, stop; otherwse, go to 2. 13

14 2. Invoke Moore s algorthm to compute the objectve value of the nstance of 1 n T wth actual processng tmes p w L. If the objectve value n T > k, then output that the problem P 5 has no a feasble soluton, stop; otherwse, let t = L and t = If t t = 1, then output that the soluton p wt s optmal wth objectve value MCC = w t, stop; otherwse set l = (t + t )/2, and nvoke Moore s algorthm to compute the nstance of 1 n T wth actual processng tmes p w l. If n T k, then set t = l and go back to 3; otherwse, set t = l and go back to 3. Theorem 5.2 If the problem P 5 has feasble solutons, then the soluton obtaned by algorthm A 4 s an optmal soluton, and the tme complexty of algorthm A 4 s O(n 2 log(nl)), where l = max,2,,n l. Proof. Wth an argument analogous to the proof of Theorem 3.3, we can show that the soluton produced by algorthm A 4 s optmal. Snce the compresson costs of all the jobs have at most L nl dfferent values, Step 3 terates at most L tmes. Therefore, we conclude that the tme complexty of algorthm A 4 s O(n 2 log(nl)). 6 Conclusons In ths paper we consdered sngle-machne schedulng wth contnuously and dscretely controllable processng tmes. The goal s to construct the trade-off curve between the number of tardy jobs and the total or maxmum compresson cost. We found that the levels of dffculty between the sum objectve (.e., total compresson cost) and bottleneck objectve cases, and between the contnuous and dscrete models, are quet dfferent. For the sum objectve case, the problems 1 contr, n T k T CC, 1 dsc contr, n T k T CC and 1 dsc contr, n T = 0 T CC are NP-hard, but for the bottleneck case, both the problems 1 contr, n T k MCC and 1 dsc contr, n T k MCC are polynomal solvable. For the contnuous model, the problem 1 contr, n T = 0 T CC s strongly polynomal solvable, but for the dscrete model, even the specal case of the problem 1 dsc contr, n T = 0 T CC s NP-hard. References [1] Z.-L. Chen, Q. Lu, G.C. Tang, Sngle machne schedulng wth dscretely controllable processng tmes, Operatons Research Letters, 21(1997),

15 [2] T.C.E. Cheng, Z.-L. Chen, C.-L. L, Sngle-machne schedulng wth trade-off between number of tardy jobs and resource allocaton, Operatons Research Letters, 19(1996), [3] P. De, E.J. Dunne, J.B. Ghosh, C.E. Wells, The dscrete tme-cost trade-off problem revsted, European Journal of Operatonal Research, 81(1995), [4] P. De, E.J. Dunne, J.B. Ghosh, C.E. Wells, Complexty of the dscrete tme-cost trade-off problem for project networks, Operatons Research, 45(1997), [5] R.L. Danels and R.K. Sarn, Sngle machne schedulng wth controllable processng tmes and number of jobs tardy, Operatons Research, 37(1989), [6] M.R. Garey, D.S. Johnson, Computers and Intractablty: A Gude to the Theory of NPhardness, Freeman, San Francsco, CA, [7] J.M. Moore, An n job, one-machne sequencng algorthm for mnmzng the number of late jobs, Management Scence, 15(1968), [8] S. Martello, P. Toth, Knapsack Problems: Algorthm and Computer Implementatons, John Wley & Sons, Chchester, [9] E. Nowck, S. Zdrzalka, A survey of results for sequencng problems wth controllable processng tmes, Dscrete Appled Mathematcs, 25(1990), [10] M. Skutella, Approxmaton algorthms for the dscrete tme-cost trade-off problem, Mathematcs of Operatons Research, 23(1998), [11] R.G. Vckson, Choosng the job sequence and processng tmes to mnmze total processng plus flow cost on a sngle machne, Operatons Research, 28(1980), [12] R.G. Vckson, Two sngle machne sequencng problems nvolvng controllable job processng tmes, IIE Transactons, 12(1980), [13] L.N. Van Wassenhove, K.R. Baker, A bcrteron approach to tme/cost trade-offs n sequencng, European Journal of Operatonal Research, 11(1982),

This is the Pre-Published Version.

This is the Pre-Published Version. Ths s the Pre-Publshed Verson. Abstract In ths paper we consder the problem of schedulng obs wth equal processng tmes on a sngle batch processng machne so as to mnmze a prmary and a secondary crtera. We

More information

The Minimum Universal Cost Flow in an Infeasible Flow Network

The Minimum Universal Cost Flow in an Infeasible Flow Network Journal of Scences, Islamc Republc of Iran 17(2): 175-180 (2006) Unversty of Tehran, ISSN 1016-1104 http://jscencesutacr The Mnmum Unversal Cost Flow n an Infeasble Flow Network H Saleh Fathabad * M Bagheran

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

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

On the Multicriteria Integer Network Flow Problem

On the Multicriteria Integer Network Flow Problem BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 5, No 2 Sofa 2005 On the Multcrtera Integer Network Flow Problem Vassl Vasslev, Marana Nkolova, Maryana Vassleva Insttute of

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

Yong Joon Ryang. 1. Introduction Consider the multicommodity transportation problem with convex quadratic cost function. 1 2 (x x0 ) T Q(x x 0 )

Yong Joon Ryang. 1. Introduction Consider the multicommodity transportation problem with convex quadratic cost function. 1 2 (x x0 ) T Q(x x 0 ) Kangweon-Kyungk Math. Jour. 4 1996), No. 1, pp. 7 16 AN ITERATIVE ROW-ACTION METHOD FOR MULTICOMMODITY TRANSPORTATION PROBLEMS Yong Joon Ryang Abstract. The optmzaton problems wth quadratc constrants often

More information

princeton univ. F 17 cos 521: Advanced Algorithm Design Lecture 7: LP Duality Lecturer: Matt Weinberg

princeton univ. F 17 cos 521: Advanced Algorithm Design Lecture 7: LP Duality Lecturer: Matt Weinberg prnceton unv. F 17 cos 521: Advanced Algorthm Desgn Lecture 7: LP Dualty Lecturer: Matt Wenberg Scrbe: LP Dualty s an extremely useful tool for analyzng structural propertes of lnear programs. Whle there

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

Foundations of Arithmetic

Foundations of Arithmetic Foundatons of Arthmetc Notaton We shall denote the sum and product of numbers n the usual notaton as a 2 + a 2 + a 3 + + a = a, a 1 a 2 a 3 a = a The notaton a b means a dvdes b,.e. ac = b where c s an

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

An Admission Control Algorithm in Cloud Computing Systems

An Admission Control Algorithm in Cloud Computing Systems An Admsson Control Algorthm n Cloud Computng Systems Authors: Frank Yeong-Sung Ln Department of Informaton Management Natonal Tawan Unversty Tape, Tawan, R.O.C. ysln@m.ntu.edu.tw Yngje Lan Management Scence

More information

Maximizing the number of nonnegative subsets

Maximizing the number of nonnegative subsets Maxmzng the number of nonnegatve subsets Noga Alon Hao Huang December 1, 213 Abstract Gven a set of n real numbers, f the sum of elements of every subset of sze larger than k s negatve, what s the maxmum

More information

On the correction of the h-index for career length

On the correction of the h-index for career length 1 On the correcton of the h-ndex for career length by L. Egghe Unverstet Hasselt (UHasselt), Campus Depenbeek, Agoralaan, B-3590 Depenbeek, Belgum 1 and Unverstet Antwerpen (UA), IBW, Stadscampus, Venusstraat

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 14: Bandits with Budget Constraints

Lecture 14: Bandits with Budget Constraints IEOR 8100-001: Learnng and Optmzaton for Sequental Decson Makng 03/07/16 Lecture 14: andts wth udget Constrants Instructor: Shpra Agrawal Scrbed by: Zhpeng Lu 1 Problem defnton In the regular Mult-armed

More information

VARIATION OF CONSTANT SUM CONSTRAINT FOR INTEGER MODEL WITH NON UNIFORM VARIABLES

VARIATION OF CONSTANT SUM CONSTRAINT FOR INTEGER MODEL WITH NON UNIFORM VARIABLES VARIATION OF CONSTANT SUM CONSTRAINT FOR INTEGER MODEL WITH NON UNIFORM VARIABLES BÂRZĂ, Slvu Faculty of Mathematcs-Informatcs Spru Haret Unversty barza_slvu@yahoo.com Abstract Ths paper wants to contnue

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

THE CHINESE REMAINDER THEOREM. We should thank the Chinese for their wonderful remainder theorem. Glenn Stevens

THE CHINESE REMAINDER THEOREM. We should thank the Chinese for their wonderful remainder theorem. Glenn Stevens THE CHINESE REMAINDER THEOREM KEITH CONRAD We should thank the Chnese for ther wonderful remander theorem. Glenn Stevens 1. Introducton The Chnese remander theorem says we can unquely solve any par of

More information

2.3 Nilpotent endomorphisms

2.3 Nilpotent endomorphisms s a block dagonal matrx, wth A Mat dm U (C) In fact, we can assume that B = B 1 B k, wth B an ordered bass of U, and that A = [f U ] B, where f U : U U s the restrcton of f to U 40 23 Nlpotent endomorphsms

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

Single-Facility Scheduling over Long Time Horizons by Logic-based Benders Decomposition

Single-Facility Scheduling over Long Time Horizons by Logic-based Benders Decomposition Sngle-Faclty Schedulng over Long Tme Horzons by Logc-based Benders Decomposton Elvn Coban and J. N. Hooker Tepper School of Busness, Carnege Mellon Unversty ecoban@andrew.cmu.edu, john@hooker.tepper.cmu.edu

More information

FUZZY GOAL PROGRAMMING VS ORDINARY FUZZY PROGRAMMING APPROACH FOR MULTI OBJECTIVE PROGRAMMING PROBLEM

FUZZY GOAL PROGRAMMING VS ORDINARY FUZZY PROGRAMMING APPROACH FOR MULTI OBJECTIVE PROGRAMMING PROBLEM Internatonal Conference on Ceramcs, Bkaner, Inda Internatonal Journal of Modern Physcs: Conference Seres Vol. 22 (2013) 757 761 World Scentfc Publshng Company DOI: 10.1142/S2010194513010982 FUZZY GOAL

More information

Example: (13320, 22140) =? Solution #1: The divisors of are 1, 2, 3, 4, 5, 6, 9, 10, 12, 15, 18, 20, 27, 30, 36, 41,

Example: (13320, 22140) =? Solution #1: The divisors of are 1, 2, 3, 4, 5, 6, 9, 10, 12, 15, 18, 20, 27, 30, 36, 41, The greatest common dvsor of two ntegers a and b (not both zero) s the largest nteger whch s a common factor of both a and b. We denote ths number by gcd(a, b), or smply (a, b) when there s no confuson

More information

The lower and upper bounds on Perron root of nonnegative irreducible matrices

The lower and upper bounds on Perron root of nonnegative irreducible matrices Journal of Computatonal Appled Mathematcs 217 (2008) 259 267 wwwelsevercom/locate/cam The lower upper bounds on Perron root of nonnegatve rreducble matrces Guang-Xn Huang a,, Feng Yn b,keguo a a College

More information

Singular Value Decomposition: Theory and Applications

Singular Value Decomposition: Theory and Applications Sngular Value Decomposton: Theory and Applcatons Danel Khashab Sprng 2015 Last Update: March 2, 2015 1 Introducton A = UDV where columns of U and V are orthonormal and matrx D s dagonal wth postve real

More information

Kernel Methods and SVMs Extension

Kernel Methods and SVMs Extension Kernel Methods and SVMs Extenson The purpose of ths document s to revew materal covered n Machne Learnng 1 Supervsed Learnng regardng support vector machnes (SVMs). Ths document also provdes a general

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

Interactive Bi-Level Multi-Objective Integer. Non-linear Programming Problem

Interactive Bi-Level Multi-Objective Integer. Non-linear Programming Problem Appled Mathematcal Scences Vol 5 0 no 65 3 33 Interactve B-Level Mult-Objectve Integer Non-lnear Programmng Problem O E Emam Department of Informaton Systems aculty of Computer Scence and nformaton Helwan

More information

Amiri s Supply Chain Model. System Engineering b Department of Mathematics and Statistics c Odette School of Business

Amiri s Supply Chain Model. System Engineering b Department of Mathematics and Statistics c Odette School of Business Amr s Supply Chan Model by S. Ashtab a,, R.J. Caron b E. Selvarajah c a Department of Industral Manufacturng System Engneerng b Department of Mathematcs Statstcs c Odette School of Busness Unversty of

More information

Integrated approach in solving parallel machine scheduling and location (ScheLoc) problem

Integrated approach in solving parallel machine scheduling and location (ScheLoc) problem Internatonal Journal of Industral Engneerng Computatons 7 (2016) 573 584 Contents lsts avalable at GrowngScence Internatonal Journal of Industral Engneerng Computatons homepage: www.growngscence.com/ec

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

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

A Single-Machine Deteriorating Job Scheduling Problem of Minimizing the Makespan with Release Times

A Single-Machine Deteriorating Job Scheduling Problem of Minimizing the Makespan with Release Times A Sngle-Machne Deteroratng Job Schedulng Problem of Mnmzng the Makespan wth Release Tmes Wen-Chung Lee, Chn-Cha Wu, and Yu-Hsang Chung Abstract In ths paper, we study a sngle-machne deteroratng ob schedulng

More information

Feature Selection: Part 1

Feature Selection: Part 1 CSE 546: Machne Learnng Lecture 5 Feature Selecton: Part 1 Instructor: Sham Kakade 1 Regresson n the hgh dmensonal settng How do we learn when the number of features d s greater than the sample sze n?

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

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

Outline and Reading. Dynamic Programming. Dynamic Programming revealed. Computing Fibonacci. The General Dynamic Programming Technique

Outline and Reading. Dynamic Programming. Dynamic Programming revealed. Computing Fibonacci. The General Dynamic Programming Technique Outlne and Readng Dynamc Programmng The General Technque ( 5.3.2) -1 Knapsac Problem ( 5.3.3) Matrx Chan-Product ( 5.3.1) Dynamc Programmng verson 1.4 1 Dynamc Programmng verson 1.4 2 Dynamc Programmng

More information

EEL 6266 Power System Operation and Control. Chapter 3 Economic Dispatch Using Dynamic Programming

EEL 6266 Power System Operation and Control. Chapter 3 Economic Dispatch Using Dynamic Programming EEL 6266 Power System Operaton and Control Chapter 3 Economc Dspatch Usng Dynamc Programmng Pecewse Lnear Cost Functons Common practce many utltes prefer to represent ther generator cost functons as sngle-

More information

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification E395 - Pattern Recognton Solutons to Introducton to Pattern Recognton, Chapter : Bayesan pattern classfcaton Preface Ths document s a soluton manual for selected exercses from Introducton to Pattern Recognton

More information

Calculation of time complexity (3%)

Calculation of time complexity (3%) Problem 1. (30%) Calculaton of tme complexty (3%) Gven n ctes, usng exhaust search to see every result takes O(n!). Calculaton of tme needed to solve the problem (2%) 40 ctes:40! dfferent tours 40 add

More information

Ranking the vertices of a complete multipartite paired comparison digraph

Ranking the vertices of a complete multipartite paired comparison digraph Rankng the vertces of a complete multpartte pared comparson dgraph Gregory Gutn Anders Yeo Department of Mathematcs and Computer Scence Odense Unversty, Denmark September 1, 2003 Abstract A pared comparson

More information

ECE559VV Project Report

ECE559VV Project Report ECE559VV Project Report (Supplementary Notes Loc Xuan Bu I. MAX SUM-RATE SCHEDULING: THE UPLINK CASE We have seen (n the presentaton that, for downlnk (broadcast channels, the strategy maxmzng the sum-rate

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

Computing Correlated Equilibria in Multi-Player Games

Computing Correlated Equilibria in Multi-Player Games Computng Correlated Equlbra n Mult-Player Games Chrstos H. Papadmtrou Presented by Zhanxang Huang December 7th, 2005 1 The Author Dr. Chrstos H. Papadmtrou CS professor at UC Berkley (taught at Harvard,

More information

Embedded Systems. 4. Aperiodic and Periodic Tasks

Embedded Systems. 4. Aperiodic and Periodic Tasks Embedded Systems 4. Aperodc and Perodc Tasks Lothar Thele 4-1 Contents of Course 1. Embedded Systems Introducton 2. Software Introducton 7. System Components 10. Models 3. Real-Tme Models 4. Perodc/Aperodc

More information

Anti-van der Waerden numbers of 3-term arithmetic progressions.

Anti-van der Waerden numbers of 3-term arithmetic progressions. Ant-van der Waerden numbers of 3-term arthmetc progressons. Zhanar Berkkyzy, Alex Schulte, and Mchael Young Aprl 24, 2016 Abstract The ant-van der Waerden number, denoted by aw([n], k), s the smallest

More information

Real-Time Systems. Multiprocessor scheduling. Multiprocessor scheduling. Multiprocessor scheduling

Real-Time Systems. Multiprocessor scheduling. Multiprocessor scheduling. Multiprocessor scheduling Real-Tme Systems Multprocessor schedulng Specfcaton Implementaton Verfcaton Multprocessor schedulng -- -- Global schedulng How are tasks assgned to processors? Statc assgnment The processor(s) used for

More information

MMA and GCMMA two methods for nonlinear optimization

MMA and GCMMA two methods for nonlinear optimization MMA and GCMMA two methods for nonlnear optmzaton Krster Svanberg Optmzaton and Systems Theory, KTH, Stockholm, Sweden. krlle@math.kth.se Ths note descrbes the algorthms used n the author s 2007 mplementatons

More information

Optimal Scheduling Algorithms to Minimize Total Flowtime on a Two-Machine Permutation Flowshop with Limited Waiting Times and Ready Times of Jobs

Optimal Scheduling Algorithms to Minimize Total Flowtime on a Two-Machine Permutation Flowshop with Limited Waiting Times and Ready Times of Jobs Optmal Schedulng Algorthms to Mnmze Total Flowtme on a Two-Machne Permutaton Flowshop wth Lmted Watng Tmes and Ready Tmes of Jobs Seong-Woo Cho Dept. Of Busness Admnstraton, Kyongg Unversty, Suwon-s, 443-760,

More information

= z 20 z n. (k 20) + 4 z k = 4

= z 20 z n. (k 20) + 4 z k = 4 Problem Set #7 solutons 7.2.. (a Fnd the coeffcent of z k n (z + z 5 + z 6 + z 7 + 5, k 20. We use the known seres expanson ( n+l ( z l l z n below: (z + z 5 + z 6 + z 7 + 5 (z 5 ( + z + z 2 + z + 5 5

More information

Some modelling aspects for the Matlab implementation of MMA

Some modelling aspects for the Matlab implementation of MMA Some modellng aspects for the Matlab mplementaton of MMA Krster Svanberg krlle@math.kth.se Optmzaton and Systems Theory Department of Mathematcs KTH, SE 10044 Stockholm September 2004 1. Consdered optmzaton

More information

A New Refinement of Jacobi Method for Solution of Linear System Equations AX=b

A New Refinement of Jacobi Method for Solution of Linear System Equations AX=b Int J Contemp Math Scences, Vol 3, 28, no 17, 819-827 A New Refnement of Jacob Method for Soluton of Lnear System Equatons AX=b F Naem Dafchah Department of Mathematcs, Faculty of Scences Unversty of Gulan,

More information

FACTORIZATION IN KRULL MONOIDS WITH INFINITE CLASS GROUP

FACTORIZATION IN KRULL MONOIDS WITH INFINITE CLASS GROUP C O L L O Q U I U M M A T H E M A T I C U M VOL. 80 1999 NO. 1 FACTORIZATION IN KRULL MONOIDS WITH INFINITE CLASS GROUP BY FLORIAN K A I N R A T H (GRAZ) Abstract. Let H be a Krull monod wth nfnte class

More information

arxiv: v1 [math.co] 1 Mar 2014

arxiv: v1 [math.co] 1 Mar 2014 Unon-ntersectng set systems Gyula O.H. Katona and Dánel T. Nagy March 4, 014 arxv:1403.0088v1 [math.co] 1 Mar 014 Abstract Three ntersecton theorems are proved. Frst, we determne the sze of the largest

More information

A new construction of 3-separable matrices via an improved decoding of Macula s construction

A new construction of 3-separable matrices via an improved decoding of Macula s construction Dscrete Optmzaton 5 008 700 704 Contents lsts avalable at ScenceDrect Dscrete Optmzaton journal homepage: wwwelsevercom/locate/dsopt A new constructon of 3-separable matrces va an mproved decodng of Macula

More information

For now, let us focus on a specific model of neurons. These are simplified from reality but can achieve remarkable results.

For now, let us focus on a specific model of neurons. These are simplified from reality but can achieve remarkable results. Neural Networks : Dervaton compled by Alvn Wan from Professor Jtendra Malk s lecture Ths type of computaton s called deep learnng and s the most popular method for many problems, such as computer vson

More information

Perfect Competition and the Nash Bargaining Solution

Perfect Competition and the Nash Bargaining Solution Perfect Competton and the Nash Barganng Soluton Renhard John Department of Economcs Unversty of Bonn Adenauerallee 24-42 53113 Bonn, Germany emal: rohn@un-bonn.de May 2005 Abstract For a lnear exchange

More information

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017 U.C. Berkeley CS94: Beyond Worst-Case Analyss Handout 4s Luca Trevsan September 5, 07 Summary of Lecture 4 In whch we ntroduce semdefnte programmng and apply t to Max Cut. Semdefnte Programmng Recall that

More information

A 2D Bounded Linear Program (H,c) 2D Linear Programming

A 2D Bounded Linear Program (H,c) 2D Linear Programming A 2D Bounded Lnear Program (H,c) h 3 v h 8 h 5 c h 4 h h 6 h 7 h 2 2D Lnear Programmng C s a polygonal regon, the ntersecton of n halfplanes. (H, c) s nfeasble, as C s empty. Feasble regon C s unbounded

More information

General viscosity iterative method for a sequence of quasi-nonexpansive mappings

General viscosity iterative method for a sequence of quasi-nonexpansive mappings Avalable onlne at www.tjnsa.com J. Nonlnear Sc. Appl. 9 (2016), 5672 5682 Research Artcle General vscosty teratve method for a sequence of quas-nonexpansve mappngs Cuje Zhang, Ynan Wang College of Scence,

More information

On the Interval Zoro Symmetric Single-step Procedure for Simultaneous Finding of Polynomial Zeros

On the Interval Zoro Symmetric Single-step Procedure for Simultaneous Finding of Polynomial Zeros Appled Mathematcal Scences, Vol. 5, 2011, no. 75, 3693-3706 On the Interval Zoro Symmetrc Sngle-step Procedure for Smultaneous Fndng of Polynomal Zeros S. F. M. Rusl, M. Mons, M. A. Hassan and W. J. Leong

More information

EXPONENTIAL ERGODICITY FOR SINGLE-BIRTH PROCESSES

EXPONENTIAL ERGODICITY FOR SINGLE-BIRTH PROCESSES J. Appl. Prob. 4, 022 032 (2004) Prnted n Israel Appled Probablty Trust 2004 EXPONENTIAL ERGODICITY FOR SINGLE-BIRTH PROCESSES YONG-HUA MAO and YU-HUI ZHANG, Beng Normal Unversty Abstract An explct, computable,

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

Affine transformations and convexity

Affine transformations and convexity Affne transformatons and convexty The purpose of ths document s to prove some basc propertes of affne transformatons nvolvng convex sets. Here are a few onlne references for background nformaton: http://math.ucr.edu/

More information

Refined Coding Bounds for Network Error Correction

Refined Coding Bounds for Network Error Correction Refned Codng Bounds for Network Error Correcton Shenghao Yang Department of Informaton Engneerng The Chnese Unversty of Hong Kong Shatn, N.T., Hong Kong shyang5@e.cuhk.edu.hk Raymond W. Yeung Department

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

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

n ). This is tight for all admissible values of t, k and n. k t + + n t

n ). This is tight for all admissible values of t, k and n. k t + + n t MAXIMIZING THE NUMBER OF NONNEGATIVE SUBSETS NOGA ALON, HAROUT AYDINIAN, AND HAO HUANG Abstract. Gven a set of n real numbers, f the sum of elements of every subset of sze larger than k s negatve, what

More information

Solutions to exam in SF1811 Optimization, Jan 14, 2015

Solutions to exam in SF1811 Optimization, Jan 14, 2015 Solutons to exam n SF8 Optmzaton, Jan 4, 25 3 3 O------O -4 \ / \ / The network: \/ where all lnks go from left to rght. /\ / \ / \ 6 O------O -5 2 4.(a) Let x = ( x 3, x 4, x 23, x 24 ) T, where the varable

More information

Valuated Binary Tree: A New Approach in Study of Integers

Valuated Binary Tree: A New Approach in Study of Integers Internatonal Journal of Scentfc Innovatve Mathematcal Research (IJSIMR) Volume 4, Issue 3, March 6, PP 63-67 ISS 347-37X (Prnt) & ISS 347-34 (Onlne) wwwarcournalsorg Valuated Bnary Tree: A ew Approach

More information

arxiv: v1 [math.ho] 18 May 2008

arxiv: v1 [math.ho] 18 May 2008 Recurrence Formulas for Fbonacc Sums Adlson J. V. Brandão, João L. Martns 2 arxv:0805.2707v [math.ho] 8 May 2008 Abstract. In ths artcle we present a new recurrence formula for a fnte sum nvolvng the Fbonacc

More information

arxiv: v1 [math.co] 12 Sep 2014

arxiv: v1 [math.co] 12 Sep 2014 arxv:1409.3707v1 [math.co] 12 Sep 2014 On the bnomal sums of Horadam sequence Nazmye Ylmaz and Necat Taskara Department of Mathematcs, Scence Faculty, Selcuk Unversty, 42075, Campus, Konya, Turkey March

More information

Capacity Constraints Across Nests in Assortment Optimization Under the Nested Logit Model

Capacity Constraints Across Nests in Assortment Optimization Under the Nested Logit Model Capacty Constrants Across Nests n Assortment Optmzaton Under the Nested Logt Model Jacob B. Feldman School of Operatons Research and Informaton Engneerng, Cornell Unversty, Ithaca, New York 14853, USA

More information

Appendix for Causal Interaction in Factorial Experiments: Application to Conjoint Analysis

Appendix for Causal Interaction in Factorial Experiments: Application to Conjoint Analysis A Appendx for Causal Interacton n Factoral Experments: Applcaton to Conjont Analyss Mathematcal Appendx: Proofs of Theorems A. Lemmas Below, we descrbe all the lemmas, whch are used to prove the man theorems

More information

Lecture Space-Bounded Derandomization

Lecture Space-Bounded Derandomization Notes on Complexty Theory Last updated: October, 2008 Jonathan Katz Lecture Space-Bounded Derandomzaton 1 Space-Bounded Derandomzaton We now dscuss derandomzaton of space-bounded algorthms. Here non-trval

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

A New Algorithm for Finding a Fuzzy Optimal. Solution for Fuzzy Transportation Problems

A New Algorithm for Finding a Fuzzy Optimal. Solution for Fuzzy Transportation Problems Appled Mathematcal Scences, Vol. 4, 200, no. 2, 79-90 A New Algorthm for Fndng a Fuzzy Optmal Soluton for Fuzzy Transportaton Problems P. Pandan and G. Nataraan Department of Mathematcs, School of Scence

More information

INTEGER NETWORK SYNTHESIS PROBLEM FOR HOP CONSTRAINED FLOWS

INTEGER NETWORK SYNTHESIS PROBLEM FOR HOP CONSTRAINED FLOWS INTEGER NETWORK SYNTHESIS PROBLEM FOR HOP CONSTRAINED FLOWS SANTOSH N. KABADI AND K.P.K. NAIR Abstract. Hop constrant s assocated wth modern communcaton network flows. We consder the problem of desgnng

More information

Vickrey Auction VCG Combinatorial Auctions. Mechanism Design. Algorithms and Data Structures. Winter 2016

Vickrey Auction VCG Combinatorial Auctions. Mechanism Design. Algorithms and Data Structures. Winter 2016 Mechansm Desgn Algorthms and Data Structures Wnter 2016 1 / 39 Vckrey Aucton Vckrey-Clarke-Groves Mechansms Sngle-Mnded Combnatoral Auctons 2 / 39 Mechansm Desgn (wth Money) Set A of outcomes to choose

More information

ON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EQUATION

ON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EQUATION Advanced Mathematcal Models & Applcatons Vol.3, No.3, 2018, pp.215-222 ON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EUATION

More information

Neryškioji dichotominių testo klausimų ir socialinių rodiklių diferencijavimo savybių klasifikacija

Neryškioji dichotominių testo klausimų ir socialinių rodiklių diferencijavimo savybių klasifikacija Neryškoj dchotomnų testo klausmų r socalnų rodklų dferencjavmo savybų klasfkacja Aleksandras KRYLOVAS, Natalja KOSAREVA, Julja KARALIŪNAITĖ Technologcal and Economc Development of Economy Receved 9 May

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

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

Randić Energy and Randić Estrada Index of a Graph

Randić Energy and Randić Estrada Index of a Graph EUROPEAN JOURNAL OF PURE AND APPLIED MATHEMATICS Vol. 5, No., 202, 88-96 ISSN 307-5543 www.ejpam.com SPECIAL ISSUE FOR THE INTERNATIONAL CONFERENCE ON APPLIED ANALYSIS AND ALGEBRA 29 JUNE -02JULY 20, ISTANBUL

More information

Y. Guo. A. Liu, T. Liu, Q. Ma UDC

Y. Guo. A. Liu, T. Liu, Q. Ma UDC UDC 517. 9 OSCILLATION OF A CLASS OF NONLINEAR PARTIAL DIFFERENCE EQUATIONS WITH CONTINUOUS VARIABLES* ОСЦИЛЯЦIЯ КЛАСУ НЕЛIНIЙНИХ ЧАСТКОВО РIЗНИЦЕВИХ РIВНЯНЬ З НЕПЕРЕРВНИМИ ЗМIННИМИ Y. Guo Graduate School

More information

ISSN: ISO 9001:2008 Certified International Journal of Engineering and Innovative Technology (IJEIT) Volume 3, Issue 1, July 2013

ISSN: ISO 9001:2008 Certified International Journal of Engineering and Innovative Technology (IJEIT) Volume 3, Issue 1, July 2013 ISSN: 2277-375 Constructon of Trend Free Run Orders for Orthogonal rrays Usng Codes bstract: Sometmes when the expermental runs are carred out n a tme order sequence, the response can depend on the run

More information

TRAPEZOIDAL FUZZY NUMBERS FOR THE TRANSPORTATION PROBLEM. Abstract

TRAPEZOIDAL FUZZY NUMBERS FOR THE TRANSPORTATION PROBLEM. Abstract TRAPEZOIDAL FUZZY NUMBERS FOR THE TRANSPORTATION PROBLEM ARINDAM CHAUDHURI* Lecturer (Mathematcs & Computer Scence) Meghnad Saha Insttute of Technology, Kolkata, Inda arndam_chau@yahoo.co.n *correspondng

More information

Multilayer Perceptron (MLP)

Multilayer Perceptron (MLP) Multlayer Perceptron (MLP) Seungjn Cho Department of Computer Scence and Engneerng Pohang Unversty of Scence and Technology 77 Cheongam-ro, Nam-gu, Pohang 37673, Korea seungjn@postech.ac.kr 1 / 20 Outlne

More information

The Second Anti-Mathima on Game Theory

The Second Anti-Mathima on Game Theory The Second Ant-Mathma on Game Theory Ath. Kehagas December 1 2006 1 Introducton In ths note we wll examne the noton of game equlbrum for three types of games 1. 2-player 2-acton zero-sum games 2. 2-player

More information

The L(2, 1)-Labeling on -Product of Graphs

The L(2, 1)-Labeling on -Product of Graphs Annals of Pure and Appled Mathematcs Vol 0, No, 05, 9-39 ISSN: 79-087X (P, 79-0888(onlne Publshed on 7 Aprl 05 wwwresearchmathscorg Annals of The L(, -Labelng on -Product of Graphs P Pradhan and Kamesh

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

Matrix Approximation via Sampling, Subspace Embedding. 1 Solving Linear Systems Using SVD

Matrix Approximation via Sampling, Subspace Embedding. 1 Solving Linear Systems Using SVD Matrx Approxmaton va Samplng, Subspace Embeddng Lecturer: Anup Rao Scrbe: Rashth Sharma, Peng Zhang 0/01/016 1 Solvng Lnear Systems Usng SVD Two applcatons of SVD have been covered so far. Today we loo

More information

A Dioid Linear Algebra Approach to Study a Class of Continuous Petri Nets

A Dioid Linear Algebra Approach to Study a Class of Continuous Petri Nets A Dod Lnear Algebra Approach to Study a Class of Contnuous Petr Nets Duan Zhang, Huapng Da, Youxan Sun Natonal Laboratory of Industral Control Technology, Zhejang Unversty, Hangzhou, P.R.Chna, 327 {dzhang,

More information

A Simple Inventory System

A Simple Inventory System A Smple Inventory System Lawrence M. Leems and Stephen K. Park, Dscrete-Event Smulaton: A Frst Course, Prentce Hall, 2006 Hu Chen Computer Scence Vrgna State Unversty Petersburg, Vrgna February 8, 2017

More information

HMMT February 2016 February 20, 2016

HMMT February 2016 February 20, 2016 HMMT February 016 February 0, 016 Combnatorcs 1. For postve ntegers n, let S n be the set of ntegers x such that n dstnct lnes, no three concurrent, can dvde a plane nto x regons (for example, S = {3,

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

Lecture 11. minimize. c j x j. j=1. 1 x j 0 j. +, b R m + and c R n +

Lecture 11. minimize. c j x j. j=1. 1 x j 0 j. +, b R m + and c R n + Topcs n Theoretcal Computer Scence May 4, 2015 Lecturer: Ola Svensson Lecture 11 Scrbes: Vncent Eggerlng, Smon Rodrguez 1 Introducton In the last lecture we covered the ellpsod method and ts applcaton

More information

CS 331 DESIGN AND ANALYSIS OF ALGORITHMS DYNAMIC PROGRAMMING. Dr. Daisy Tang

CS 331 DESIGN AND ANALYSIS OF ALGORITHMS DYNAMIC PROGRAMMING. Dr. Daisy Tang CS DESIGN ND NLYSIS OF LGORITHMS DYNMIC PROGRMMING Dr. Dasy Tang Dynamc Programmng Idea: Problems can be dvded nto stages Soluton s a sequence o decsons and the decson at the current stage s based on the

More information