ISSN: 2321-7782 (Olie) Volume 1, Issue 4, September 2013 Iteratioal Joural of Advace Research i Computer Sciece ad Maagemet Studies Research Paper Available olie at: www.ijarcsms.com Decisio Support System (DSS) for Capacity Plaig: A Case Study Jawahar Babu A 1 Raja Gopal K 2 Departmet of Mech. Egg Gudlavalleru Egieerig College Gudlavalleru Idia Departmet of Mech. Egg K.S.R.M. Egieerig College Kadapa - Idia Abstract: A Decisio Support System (DSS) refers to a class of systems which support the process of makig decisios. DSS is give the opportuity to overcome the cogitive limitatios of huma decisio makers ad offers effectiveess ad efficiecy i makig decisios. Capacity plaig is the process of determiig the ecessary people, machies, ad physical resources to meet the productio objectives of the firm. Without provisio of adequate capacity or recogitio of the existece of excess capacity, the beefits of a effective Maufacturig Plaig ad Cotrol system caot fully realize. The proposed DSS provides a correspodece betwee the capacity required to execute Master Productio Schedule (MPS) ad capacity available to implemet MPS. Keywords: Decisio Support System; Cogitive Limitatios; Capacity Plaig; Maufacturig Plaig ad Cotrol System; Master Productio Schedule. I. INTRODUCTION Decisio makig is a process of selectio from a set of alterative courses of actio to fulfil the objectives of the decisio problem more satisfactorily tha others [8, 9]. Decisios may be categorized ito structured, ustructured ad semi-structured decisios [3, 7]. Structured decisios ivolve situatios where the procedures to follow, whe a decisio is eeded ca be specified i advace. Such decisios are structured or programmed by the decisio procedures or decisio rules developed for them. Ustructured decisios ivolve decisio situatios where it is ot possible or desirable to specify i advace most of the decisio procedures to follow. I betwee the above categories there exist semi structured decisios, where some decisio procedures ca be pre specified, but ot eough to lead to a defiite recommeded decisio. Although a Decisio support system ca be used for a wide variety of decisios, the decisio support cocept applies better to semi structured ad ustructured problems. Gorry ad Scott [4] offer followig defiitio of a DSS: Decisio Support Systems couple the itellectual resources of idividuals with capabilities of the computer to improve the quality of decisios. They comprise a computer based support system for maagemet decisio makers who deal with semi-structured problems. George M. Marakas [8] defies DSS as a system uder the cotrol of oe or more decisio makers that assists i the activity of decisio makig by providig a orgaized set of tools iteded to impose structure o portios of the decisio makig situatio ad to improve the ultimate effectiveess of the decisio outcome. A Decisio Support System (DSS) refers to a class of systems which support the process of makig decisios. The emphasis is o support rather tha a automatio of decisios [1]. DSS allow decisio makers to retrieve data ad test alterative solutios durig the process of problem solvig. Effective problem solvig is iteractive ad is ehaced by dialogue betwee the user ad the system [3]. A DSS cosists of hardware, software, data, model, ad people. 2013, IJARCSMS All Rights Reserved 24 P a g e
Hardware resources iclude maagemet workstatios, departmet miicomputers, ad corporate maiframes. Software resources iclude software packages such as DSS geerators ad spreadsheet packages that perform database maagemet, model base maagemet ad dialogue geeratio ad maagemet. Data ad model resources iclude a database extracted from iteral, exteral ad persoal databases, ad a model base that is a collectio of mathematical models ad aalytical techiques. People resources iclude maagers ad staff specialists who explore decisio alteratives with the support of DSS. A DSS offers the followig types of support: Explores multiple perspectives of a decisio cotext. Geerates multiple ad higher quality alteratives for cosideratio. Explores ad tests multiple problem-solvig strategies. Icreases the decisio maker s ability to tackle complex problems. Improves respose time of decisio maker. Discourages premature decisio makig ad alterative selectio. II. CAPACITY PLANNING Productio capacity is defied as the maximum rate of output that a productio facility (or productio lie, work ceter, or group of work ceters) is able to produce uder a give set of assumed operatig coditios. The assumed operatig coditios refer to the umber of shifts per day, umber of days i the week (or moth) that the plat operates, employmet levels, ad so forth [5]. Capacity plaig is the process of determiig the ecessary people, machies, ad physical resources to meet the productio objectives of the firm [10]. A master productio schedule details usually o a mothly, weekly or eve daily basis, the quatity of specific products or product groups to be produced. To assess the realistic ature of the MPS, its resources must be checked. Capacity plaig compares the available capacity with required capacity. Capacity plaig may be doe i two stages [6]. Rough Cut Capacity Plaig (RCCP) Capacity Requiremet Plaig (CRP) 1) A Rough Cut Capacity Plaig: RCCP is used to examie the effects of proposed MPS o key work ceters, departmets ad machies. Durig RCCP work ceters whose capacity is isufficiet are idetified. The capacity required at each selected work ceter is the calculated usig the proposed MPS. If the required capacity exceeds the available capacity at oe or more ceters, the MPS must be modified. Every time the MPS is ameded its effects o key resources must be re-examied. Thus RCCP is a iterative process for which the computer is well suited. The process eds whe the MPS appears to be feasible. RCCP may be accomplished usig followig three methods. Overall Factors Method. Capacity Bill Method. Resource Profiles. The overall factor method uses past data to determie the percetage of total productio hours that ca be attributed to each work ceter. These percetages are used to estimate the expected workload at each work ceter for each MPS time period. Followig equatio has bee used to obtai the total umber of productio hours required for a particular week [6]. 2013, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Olie) 25 P a g e
H t = q pt h p P=1 l wt = H t r w Where H t = Total umber of productio hours required for the period t. q pt = Number of uits of product p to be produced durig period t. h p = Total productio hours required by product p. = Number of products to be produced. l wt = Expected work load at work ceter w durig period t. r w = Percetage of total productio hours assiged to work ceter w durig previous period. Assumptio: i this method is, the product mix remai same. This method is easy to use ad requires a miimal amout of data. Capacity Bill Method: This method estimates the expected work load at each work ceter w for each MPS time period t usig the followig equatio [6]. l w, t = (q p,t )(h p,w ) p=1 l w, t = Expected work load at work ceter w for the period t q p, t = Number of uits of product p to be produced durig period t h p, w = Number of productio hours required by product p at work ceter w = Number of products to be produced. I resource profile method lead-time to maufacture the various compoets, subassemblies will be take ito cosideratio to project the required capacities at various work ceters. This method is ot cosidered i the preset work. 2) Capacity Requiremet Plaig: I capacity requiremet plaig lot sizes of the compoets to be maufactured ad the lead times required to maufacture the compoets will also be take ito cosideratio which will give the realistic projectio of the capacities. I the preset work CRP has ot bee icorporated. III. NEED FOR DSS FOR CAPACITY PLANNING Maufacturig firms have made cosiderable progress i the istallatio ad use of Material Requiremet plaig (MRP) systems to pla ad cotrol fabricatio ad assembly operatios. But to maage effectively the MRP systems two road blocks must be overcome. They are: 1) Preparatio of Master Productio Schedule (MPS) ad 2) The developmet of capacity plas to support that schedule May firms fail to implemet MPS because of poor plaig of capacity at idividual work ceters. A critical activity that parallels the developmet of the material plas is the developmet of capacity plas. Without the provisio of adequate capacity or recogitio of the existece of excess capacity, the beefits of a effective Maufacturig Plaig ad Cotrol (MPC) system caot be fully realized, because isufficiet capacity will quickly lead to deterioratig delivery performace, escalatig work-i-process ivetories. O the other had, excess capacity may be a eedless expese that ca be reduced. Firms must be equipped with iformatio system which provides appropriate work ceter capacities (11). With this objective a DSS for 2013, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Olie) 26 P a g e
Capacity Plaig is proposed. The proposed DSS provides a correspodece betwee the capacity required to execute a give material pla ad that made available to execute the pla. If this correspodece does ot exist, the pla will either be impossible to execute or iefficietly executed. The proposed DSS has bee implemeted for a small-scale idustry, which maufacture Trasformers. A. Developmet of DSS: Although several models are available, for the developmet of DSS for capacity plaig oly two models, amely capacities plaig usig overall factors (CPOF) ad bill of capacity method have bee used. Data regardig Master Productio Schedule, time required for maufacture of each model of trasformer, historical data regardig usage of various ceters, bill of capacity for each model have bee collected from a idustry which maufactures trasformers. B. Case Study (for CPOF): A SSI uit, Maufacturers of trasformers had the Master Productio Schedule (MPS) as show i Table-I. To determie the percetage of total time for each work ceter historical data of four periods has bee obtaied. Each period is oe moth. Table-3 shows the percetage of total time for differet work ceters. The time required at idividual work ceters based o the overall factor method is show i the table-iv. C. Case Study (for Bill of Capacity method) To implemet the bill of capacity method i additio to Master Productio Schedule, Bill of materials ad Routig ad operatio time data have bee obtaied. Bill of capacity for each product is the sum of the operatio timigs required at differet workstatios to produce that product. By usig MPS ad Bill of capacities the work load at idividual work ceters has bee calculated. For the selected SSI uit Bill of capacity for differet models of trasformers are obtaied ad preseted i Table-V. Ad required capacity for various work ceters is preseted i Table-VI. TABLE-I Master Productio Schedule for the year 2011-12. Ed Product 1 2 3 4 5 6 7 8 9 10 11 12 M-1 400 440 400 300 600 400 450 400 360 250 400 400 M-2 150 120 150 150-250 200 240 160 170 180 170 M-3-150 120 170 150 - - 120 120 200-140 M-4 44 45 30 45 30 35 60-45 50 55 20 TABLE-II Direct Labour Time for Each Model: M = Model Sr.No Ed product Total direct labour time i stadard hours/uit 1 M-1 5.58 2 M-2 6.92 3 M-3 9.33 4 M-4 17.00 2013, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Olie) 27 P a g e
TABLE-III: Historical Data to Determie the Percetage of Total Time Used by Work Ceters. S.o Work ceter Number 0f hours Percetage of Total hours 1 Sheet metal cuttig 190 1.12 2 Bedig machie 166 0.98 3 Drillig machie 590 3.48 4 Gas cuttig machie 870 5.13 5 Arc weldig machie 2740 16.17 6 Core buildig statio 2600 15.34 7 L.V.widig machie 1230 7.26 8 H.V.Widig machie 2740 16.17 9 Sub assembly workstatio 3360 19.83 10 Fial assembly workstatio 1810 10.68 11 Spray paitig work ceter 650 3.84 TABLE-IV: Capacity Plaig Usig Overall Factors Wc 1 2 3 4 5 6 7 8 9 10 11 12 1 45 61 54.8 56.7 58.8 51 55 56.1 56 59.2 49.4 56.6 2 39.3 53.4 48 49.6 51.5 44.6 48.1 49.1 49 51.8 43.2 49.5 3 139.8 189.6 170.5 176.2 182.9 158.5 171 174.4 174 184 153.5 175.9 4 206.1 279.5 251.3 259.7 269.7 233.7 252.1 257.1 256.5 271.2 226.3 259.3 5 649.7 881.2 792.2 818.6 850.1 736.8 794.7 810.5 808.6 854.9 713.5 817.3 6 616.3 836 751.6 776.6 806.5 699 753.9 768.9 767 811 676.8 775.3 7 291.7 395.6 355.7 367.5 381.7 330.8 356.8 363.9 363 383.8 320.3 366.92 8 649.7 881.2 792.2 818.6 850.1 736.8 794.7 810.5 808.6 854.9 713.5 817.3 9 796.7 1080.7 971.6 1004 1042.5 903.6 974.6 993.9 991.6 1048.5 875 1002.3 10 429.1 582 523.2 540.7 561.5 846.6 524.9 535.3 534 564.6 471.2 539.8 11 154.2 209.2 188.1 194.4 201.8 174.9 188.7 192.4 192 203 169.4 194 Tpc 4018 5450 4899 5063 5257 4557 4915 5012 5000 5287 4412 5054 Wc=Work ceter Tpc=Total plat capacity(i hours) TABLE -V: Bill of Capacity for Differet Models of Trasformers: S.o Work ceter M-1 M-2 M-3 M-4 1 Sheet metal cuttig 3.75 3.75 3.75 3.75 2 Bedig machie 3.25 3.25 3.25 3.25 3 Drillig machie 11 17 20 33 4 Gas cuttig machie 18.75 21.75 25.75 38.75 5 Arc weldig machie 60 75 85 160 6 Core buildig statio 55 68 75 129 7 L.V. Widig machie 21 30 48 96 8 H.V. Widig machie 48 60 90 243 9 Sub assembly workstatio 66.5 80.5 110.5 176 2013, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Olie) 28 P a g e
10 Fial assembly workstatio 35.5 40.5 80.5 111 11 Spray paitig work ceter 12.5 15.5 18.5 26 Total time per ui 335.25 415.25 560.25 1019.75 (I miutes) TABLE-VI Capacity Plaig usig Capacity Bills (i hours) Wc 1 2 3 4 5 6 7 8 9 10 11 12 1 37.1 47.1 43.7 41.5 48.7 42.7 44.3 47.5 42.8 41.8 39.6 45.62 2 32.1 40.9 37.9 36 42.2 37.1 38.4 41.7 37.1 36.3 34.4 39.5 3 140 189.4 172.3 178.9 176.5 163.4 172.1 181.3 176 188.1 154.8 179.1 4 207.8 274.4 250.2 250.1 271.2 238.2 251.8 263.5 251 257.8 225.7 259.6 5 704.8 922.5 837.5 848.3 892.5 805.8 860 870 850 879.1 771.6 864.1 6 631.2 823.5 751.1 754.2 802 725.2 768.1 788.6 758 779.3 688.9 777.3 7 285.4 406 359 388 378 321 353.5 356 374 412.5 318 369 8 648.2 879.2 771.5 827.2 826.5 711.7 803 740 810.2 872.5 722.7 781 9 773.6 1056.9 953.5 978.8 1029.2 881.4 943 986.3 966.6 1020.1 846.1 987.9 10 419.3 625.8 554.4 590 611.7 470.1 512.2 559.6 565.2 623.5 459.9 626.2 11 141.1 188.4 172 173 184.2 163 171.4 182.3 172.8 179.3 153.6 179 TPC 4020.9 5454.4 4903.5 5066.5 5263 4560 4918.2 5016.5 5004.1 5290.7 4415.5 5058.7 TABLE-VII: Available Capacities of Various Work ceters Work Ceter 1 2 3 4 5 6 7 8 9 10 11 Number of Machies 1 1 1 2 5 6 3 8 7 5 1 Capacity per moth(hrs) 140 120 180 260 900 840 360 960 980 700 180 IV. RESULTS AND DISCUSSIONS Capacity plaig by CPOF ad capacity bills has bee preseted i the table-iv ad table-vi respectively. I CPOF procedure, data requiremets are miimal ad calculatios are straight forward. However the CPOF approximatios of capacity requiremets at idividual work ceters are oly valid to the extet that product mixes or historical divisios of work betwee work ceters remai costat. The capacity bill method provides a much more direct lik betwee idividual ed products i the MPS ad capacity required for idividual work ceters. It takes ito accout ay shifts i product mix. Cosequetly it requires more data tha the CPOF procedure. Results of CPOF ad capacity bill method reveal the differeces i work ceter estimates for each time period. These differeces are far more importat i firms which experiece sigificat period-to-period product mix variatios tha i those that have a relatively costat patter of work. Compariso of required capacities with the available capacities of idividual work ceters gives the maagers to take a apt decisio regardig the strategies to be followed wheever the required capacity exceeds the available capacity. 2013, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Olie) 29 P a g e
V. CONCLUSION Makig decisios regardig complex system's strais huma s cogitive capabilities. There is a substatial amout of empirical evidece that huma ituitive judgmet ad decisio makig ca be far from optimal ad it deteriorates further with complexity. Decisio Support Systems ca aid huma cogitive deficiecies by itegratig various sources of iformatio, providig itelliget access to relevat kowledge, aidig the process of structurig decisios. The model based DSS for capacity plaig gives the required capacities at various work ceters. If there is a mismatch betwee available capacity ad required capacity the maager has to take a appropriate decisio to bridge the gap i order to realize the Master Productio Schedule. ACKNOWLEDGEMENT The authors would like to express a deep sese of gratitude to M/s Shirdi Sai Electricals, Idustrial Estate, Kadapa, Adhra Pradesh for their cooperatio ad providig valuable data to carry out our work. Refereces 1. Alter.S.L, 1980, Decisio Support Systems: Curret Practices ad Cotiuig Challeges, Addiso-Wesley, Readig, M.A. 2. Berry. W.L, Thomas G. Schmitt, CPIM, Thomas E. Vollma, 1982, Capacity Plaig Techiques for Maufacturig Cotrol Systems: Iformatio Requiremets ad Operatioal Features, Joural of operatios Maagemet, Vol. 3 (1), 13-25. 3. Davis, G.B, Margreth H.Olso, 2000, Maagemet Iformatio Systems, Coceptual foudatios, Structure ad Developmet, Tata Mc Graw-Hill editio. 4. Gorry, G.A, ad M.S. Scott Morto. 1989, A Frame work for Maagemet Iformatio Systems, Sloa Maagemet Review, 13 (1), 49-62. 5. Groover M.P, 2003, Automatio, Productio Systems ad Computer Itegrated Maufacturig, secod editio, Pearso Educatio. 6. Hamid Noori, Russel Radford, 1995, Productio ad Operatios Maagemet: Total Quality resposiveess Mc Graw-hill Iteratioal editio. 7. James A O Brie, 1998 Maagemet Iformatio Systems. A Maagerial Ed user perspective, Galgotia publicatios pvt Ltd. 8. Marakas, G.M, 2003, Decisio Support Systems i the 21 st Cetury, Secod Editio, Pearso Educatio. 9. Mallach, E.G, 2002, Decisio Support ad Data ware house Systems Tata Mc Graw-Hill publishig Compay. 10. Seetha Ram L. Narasimha, Deis W. Mc Leavey, Peter J. Billigto, 2003, Productio Plaig ad Ivetory Cotrol, Pretice Hall of Idia Pvt. Ltd. 11. Vollma, T.E, William L. Berry ad D. Clay Whybark, 2001, Maufacturig Plaig ad Cotrol Systems, Galgotia Publicatios Pvt Ltd. AUTHOR PROFILE Jawahar Babu Agalakuditi received his B.Tech i Mechaical Egieerig from Sri Vekateswara Uiversity-Tirupathi i 1988 ad M.E i Productio Egieerig from Bharathiar Uiversity-Coimbatore i 1991. He received his doctorial degree i Idustrial Egieerig from JNTU-Hyderabad i 2006. He has more tha two decades of teachig experiece. His areas of iterest are decisio support systems ad CAD/CAM. Presetly he is guidig oe Ph.D scholar i the area of Idustrial Egieerig. He is workig as Professor ad Head of Mechaical Egieerig at Gudlavalleru Egieerig College, Gudlavalleru, Krisha District, Adhra Pradesh, Idia. Rajagopal Kurool received his B.Tech i Mechaical Egieerig from JNTU-Hyderabad. He received his doctorial degree i Idustrial egieerig from Sri Vekateswara Uiversity-Tirupathi. He has published several papers i cofereces ad jourals of atioal ad iteratioal repute.he has guided three Ph.D scholars. His active areas of research are decisio support systems, productio plaig ad operatios research. Presetly he is workig as Professor ad Head of the Mechaical Egieerig Departmet at KSRM College of Egieerig, Kadapa, Adhra Pradesh, Idia. 2013, IJARCSMS All Rights Reserved ISSN: 2321-7782 (Olie) 30 P a g e