Вы находитесь на странице: 1из 10

Heuristic Approach For n 2 Open Shop Scheduling Problem Including Transportation Time And Job Block Criteria, Probabilities

s Associated With Processing Time . Deepak Gupta Prof. & Head, Department of Mathematics, Maharishi Markandeshwar University, Mullana, Haryana, India guptadeepak2003@yahoo.co.in. Renuka Research Scholar, Department of Mathematics Maharishi Markandeshwar University, Mullana, Haryana, India renukajain7@gmail.com Payal Singla Research Scholar,Department of Mathematics, Maharishi Markandeshwar University, Mullana, Haryana, India Payalsingla86@gmail.com Abstract: In the present paper an attempt is made to obtain the optimal sequence for n jobs processing on two machines in which the order of processing is not given. The problem consists the processing time which are associated with probabilities, the concept of equivalent job for a job block and transportation time from one machine to another is also included. The main objective of the study is to obtain optimal sequence for n jobs in order to minimizing the total elapsed time with the help of heuristic approach. A numerical illustration is also given to clarify the algorithm. KeyWords: Open shop problem, Equivalent job, Elapsed time, Job block, Transportation time. 1.Introduction: Job sequence and scheduling problem has a wider application in the industrial areas, which minimizing the manufacturing cost and time or optimizes effectiveness by selecting the most suitable sequence. The present paper studies open job shop scheduling in which operations can be performed in any order but only once at a time i.e. for each job machine order is not fixed and not known in advance. So the machine is continuously available from time zero and operations are processed without preemption. The objective of the open shop scheduling problem is to determine a feasible combination of the machine orders and job orders(sequence), which minimizes the maximum completion time(makespan).The optimization algorithm for two or three multi stage flow shop problem was introduced by Johnson [7]. Further the work was developed by by Anup[1], D.Rebaine[2], Gupta Page 1

Deepak[3], Gupta D; Singla.P;SharmaS;et.al.[4],GuptaDeepakandSinghT.P.[5],HarbansLalandMaggu.P.L.[6],Johnson.S.M.[7],Ja ckson,J.R[8],Maggu&Das[9],M.DeelsAmico[10],Singh.T.P.[11],Singh.T.P.and Rastogi.V.K[12],Yoshida and Hitomi[13]byintroducing different parameters such as transportation time ,break down interval,weightage interval,weightage etc .Anup associate probabilities with processing time as the processing time are always not be taken exact . The Maggu & Das introduce the concept of equivalent job for a job block in which priority is given to one job over the another. In this Paper, we assume that there is a time lag (parameter) between the completion of an operation and the beginning of the next operation of the same job. We refer to this lag as the transportation time, under the assumption that the transporting device is always available. Hence the present paper combines the study made by Gupta Deepak, Singh.T.P., Maggu and Das, Anup, K.Rajendra and extends the basic study of Johnson(1954) by including the parameter transportation time. Thus the problem in the presentpaperhas wider and practically more applicable and provides suitable results. 2.Problem Formulation: Let n jobs 1,2,3,..n be processed through two machines A and B. Let Ai be the processing time of ith job (i=1,2,3.n) on machine A and pi be the probabilities associated with processing time Ai such that 0 pi 1 and pi=1 and Bi be the processing time of ith job on machine B and qi be the probabilities associated with processing time Bi such that 0 qi 1 and qi =1.Let Ti,12be the transportation time of ith job from 1st machines to the 2nd machine which is same as transportation time from 2nd to 1st i.e. Ti,12 is same asTi,21 et be an equivalent job for job block (k,m) in which k is given priority over job m, where k and m are any jobs among the given n jobs. The mathematical model of the problem can be stated in the matrix form as: Jobs i 1 2 3 . . . n Machine A Ai A1 A2 A3 . . . An pi p1 p2 p3 . . . pn Ti,12 Ti,21 T1,12 T2,12 T3,12 . . . Tn,12 Machine B Bi qi B1 B2 B3 . . . Bn q1 q2 q3 . . . qn

Tableau-1 Our objective is to find the optimal schedule of all the jobs which minimizes the total elapsed time of each machine, expected processing time may be in the terms of cost. 3. Assumptions: 1. 2. Two jobs cant be processed on a single machine at a time. Transportation time from machine first to second and second to first are same. Page 2

3. 4. 5. 6.

Priority is given to the i1 job over i2.ik in job block (i1,i2ik). The second job will be processed on a machine when first job is completed. Transporting device is always available. pi=1, qi=1, 0pi1, 0qi1

4. Algorithm: For n*2 open shop scheduling problem, the heuristic algorithm is given as: Step1: Define expected processing time A'i and B'i on machine A and B respectively as follows(a) A'i = Ai*pi (b) B'i= Bi*qi Step2: Define two fictitious machines namely G and H with processing time Gi and Hi as follows :Gi =A'i +Ti,12 ; Hi =G'i+Ti,21

Step3: Define processing time for the equivalent job = (k,m) as follows:For the order G (a)G' = G'k + G'm - min(G'm, H'k) (b)H' =H'K +H'm min(G'm,H'k) For the order H (a)H' = H'k + H'm min(G'k,H'm) (b)G' = G'k +G'm min(G'K,H'm) Step4: For G flow shop problem, Construct a set SG of all the processing time G'i where G'i H'i and S'G of all the processing time G where G'i H'i. Step5: Let S1 denote a sub optimal sequence of jobs corresponding to non decreasing times SG and similarly a sequence S2 corresponding to set S'G. Step6: The augmented ordered sequence (S1, S2) gives optimal or near optimal sequence for processing the jobs on machine A for the given problem. Step7: For the order H G flow shop problem
I

Construct the set SH and S'H of processing times H'i Where H'i G'i and of processing times H H'i G'i respectively .

where

Step8: LetS2 denote a sub optimal sequence of jobs corresponding to the non decreasing processing times in the set SH .. similarlly S'2 corresponding to S'H. Page 3

Step9: The augmented ordered sequence (S'1, S'2) gives the optimal or near optimal sequence for processing the jobs on the machine B for the given problem. Step10: Prepare in-out tables of sequences (S1, S2) and (S'1,S'2) in the order A and B respectively by including transportation time.

Numerical Illustration: Consider 5 jobs, 2 machines namely A and B open shop problem with processing time Ai and Bi associated with their respected probabilities pi and qi and transportation time Ti,12 or Ti,21 given in the following table. The jobs 2 and 4 are processed as a group job (2,4). jobs i 1 2 3 4 5 Machine A Ti,12 Ti,21 Ai Pi 21 0.2 2 16 0.1 5 13 0.3 3 17 0.2 4 11 0.2 5 Tableau-2 Machine B Bi qi 14 0.1 15 0.2 18 0.3 10 0.1 18 0.3

Our object is to find the optimal sequence which minimize the total elapsed time. Solution:As per Step1: Define expected processing time A'i and B following formulae:(a) A'i = Ai*pi (b) B'i = Bi*qi Jobs 1 2 3 4 5 Machine A 4.2 1.6 3.9 3.4 2.2 Tableau-3 Machine B 1.4 3.0 5.4 1.0 5.4
i

on machine A and B respectively by

As per Step2: Define two fictitious machines G and H with their processing times Gi and Hi using: Gi = A'i + Ti,12 ; Jobs 1 2 3 4 5 Gi 6.2 6.6 6.9 7.4 7.2 Tableau-4 Hi 3.4 8.0 8.4 5.0 10.4 Hi = B'i + Ti,21

Page 4

As per Step3: For Flow shop G Define processing time for the equivalent job = (2,4) using: (a) G' =G'k +G'm min(H'k,G'm) G' = 6.6 +7.4 min(8.0, 7.4) =14.0 -7.4 =6.6 (b) H' =H'k +H'm min(H'k,G'm) H' = 8.0 +5.0 min(8.0, 7.4) =13.0-7.4 =5.6 As per Step4: Reduced processing times Gi and Hi are as follows : Jobs 1 3 5 G'i 6.2 6.6 6.9 7.2 H'i 3.4 5.6 8.4 10.4

Tableau-5 As per Step5: Construct a set SG and S'G S G = {6.9,7.2} As per Step6: S1 = {3, 5} S'1 ={1 ,} S'G = {6.2, 6.6}

As per Step7: Augmented ordered sequence = {3, 5, 1, } i.e. {3 ,5, 1, 2, 4} For flow shop H As in step 1 Jobs 1 2 3 4 5 B'i 1.4 3.0 5.4 1.0 5.4 Ti,21 2 5 3 4 5 Tableau-6 A'i 4.2 1.6 3.9 3.4 2.2 :

Page 5

As per Step2: Define two fictitious machines H and G with processing time Hi and Gi for job i on machines H and G respectively as: Hi = B'i + Ti,21 Gi = A'i + Ti,12 Jobs 1 2 3 4 5 Hi 3.4 8.0 8.4 5.0 10.4 Tableau-7 Gi 6.2 6.6 6.9 7.4 7.2

As per Step3: Define processing time for equivalent job =(2,4) H' = H'k +H'm min (H'm,G'k) =8.0 +5.0 min(6.6,5.0) = 13.0 5.0 =8.0 G' =G'k +G'm min (H'm, G'k) =6.6 +7.4 min (6.6, 5.0) =14.0 -5.0 =9.0 As per Step8: For G Jobs 1 3 5 flow shop, construct the set SH and S'H: Hi 3.4 8.0 8.4 10.4 Tableau-8 S'H ={8.4,10.4} S2 ={1,} S'2 ={3,5} Gi 6.2 9.0 6.9 7.2

SH ={3.4,8.0} As per Step9:

As per Step10: Augmented ordered sequence is = {1, , 3, 5} i.e. {1,2,4,3,5} This gives the exact or near optimal sequence for processing the jobs on machine G. Now we calculate the total production times on machine A and B for the sequence (3,5,1,2,4) and(1,2,4,3,5 )respectively. As per Step11: In Out table for flow shop A B Page 6

For flow shop AB Jobs 3 5 1 2 4 Machine A In Out 0 - 3.9 3.9 - 6.1 6.1 - 10.3 10.3 - 11.9 11.9 - 15.3 Tableau-9 Machine B In 6.9 12.3 17.7 19.1 21.1 Out 12.3 17.7 19.1 21.1 22.1

In Out table for flow shop B A For flow shop B Machine B In 0 4.2 5.8 9.2 13.1 Out - 4.2 - 5.8 - 9.2 -13.1 - 15.3 Tableau-10

jobs 1 2 4 5 3

Machine A In 6.2 10.8 13.8 16.1 21.5 Out 7.6 13.8 14.8 21.5 26.9

The total expected elapsed time when the flow is from A to B for the sequence (3, 5, 1, 2, 4) is 22.1 units and for the sequence (1, 2, 4, 3, 5) is 26.9 units when flow is B to A. Hence the optimal sequence of all the jobs which minimizing the total elapsed time of each machine is (3, 5, 1, 2, 4) for flow shop A B. Remarks 1.If the probabilities are not associated with processing time then the results tally with[6]. 2.If probabilities are associate with processing time then the results tally with[5]. 3.If both probabilities and transportation time are not included then results tally with[6]. 4. Equivalent job formulation is associative and non commutative in nature. 5. The study on n*2 open shop scheduling may be further extended by including various parameters such as break down intervals, weightage of jobs etc. 6. The study may further be extended for n jobs 3 machine open shop problem.

References Page 7

[1].Anup (2002),"On two machine flow shop problem in which processing time assume Probabilities and there exists equivalent for an ordered job block,JISSO,VolXXIII No.1-4,pp.41-44. [2].D.Rebaine, V.A.Strusevich(1998), Two machine open shop scheduling with special transportation times,CASSM R&D paper 15,University of Greenwich, London,UK. [3]. Gupta Deepak & Singh.T.P.(2004), Published in the Proceedings of International conference on Industrial & Applied Mathematics at India International Center,New Delhi, 4-6. Dec. [4]. Gupta D; Singla P; Sharma S;et.al(2011), Minimizing Rental Cost for n-jobs 2-Machines flow shop scheduling , processing time association with probabilities including Transportation time and job block criteria,Information and Knowledge Management,ISSN 2224 5758(paper) ISSN 2224896x(online),Vol 1,No.3,2011. [5]. Gupta Deepak, Singh T.P.(2005),"On job block open shop scheduling The processing time associated with probabilities.J.Indian Soc. Stat.Opers.Res. Vol.XXVI,No.1-4. [6]. Harbans Lal and Maggu.P.L. (1989), On job block open shop scheduling problem ,PAMS pp 45-51. Vol.XXIX. [7]. Johnson.S.M.(1954), Optimal two and three stage production schedule with set up times Included.Nay. Res. Log. Quart. Vol.1.pp -61-68. [8]. Jackson,J.R.(1956), An extention of Johnsons result on job scheduling,Nav.Res.Log.Quar.,3, 201203. [9]. Maggu. P .L. & Das G.(1977), Equivalent jobs for job block in job sequencing , Opsearch, Vol.14.No .4,pp.277-281(1977). [10]. M. DellAmico(1996), Shop problems with two machines and time lags,Oper.Res.44,777-787. [11]. Singh T.P.(1985),"On n*2 shop problem involving job block, transportation time & break down machine times,PAMS Vol.XXI. [12]. ]. Singh T.P. & Rastogi V.K. (1998), Group job restrictions in a flowshop problem involving separated sequences independent set-up times and transportation time. The Mathematics Education Vol.XXXII.No.1.pp 12-16. [13]. Yoshida and Hitomi(1979) , Optimal two stage production schedule with set up timeSeparated,AIIE Transactions. Vol.II.pp261-263. .-

Page 8

[2].Gupta Deepak, Singh T.P.(2005),"On job block open shop scheduling The processing time associated with probabilities.J.Indian Soc. Stat.Opers.Res. Vol.XXVI,No.1-4. [3]. Johnson.S.M.(1954), Optimal two and three stage production schedule with set up times Included.Nay. Res. Log. Quart. Vol.1.pp -61-68. [4].Jackson,J.R.(1956),An extention of Johnsons result on job scheduling,Nav.Res.Log.Quar.,3, 201-203. [5]. Maggu. P .L. & Das G.(1977), Equivalent jobs for job block in job sequencing , Opsearch, Vol.14.No .4,pp.277-281(1977). [6]. Maggu .P. I. and Harbans Lal (1989), On job block open shop scheduling problem ,PAMS .-pp 45-51. Vol.XXIX [7].Singh T.P.(1985),"On n*2 shop problem involving job block, transportation time & break down machine times,PAMS Vol.XXI. [8]. Singh T.P. & Rastogi V.K. (1998), Group job restrictions in a flowshop problem involving separated sequences independent set-up times and transportation time. The Mathematics Education Vol.XXXII.No.1.pp 12-16. [9] Gupta D; Singla P; Sharma S;et.al(2011), Minimizing Rental Cost for n-jobs 2-Machines flow shop scheduling , processing time association with probabilities including Transportation time and job block criteria,Information and Knowledge Management,ISSN 2224 5758(paper) ISSN 2224-

Page 9

896x(online),Vol 1,No.3,2011. [10]. Gupta Deepak & Singh.T.P.(2004), Published in the Proceedings of International conference on Industrial & Applied Mathematics at India International Center,New Delhi, 4-6. Dec. [11].yoshida and Hitomi(1979) , Optimal two stage production schedule with set up times Separated,AIIE Transactions. Vol.II.pp261-263.

Page 10

Вам также может понравиться