INTEGER | 2AMMING | est, of pAcdiwct Catry oldu. L -Agad 0:——?. — Un Lathan ose om LPP voli oly Nae | a pe onal bey are _Aegutrdd to | — S a Armpurn Oe reed _tnshee jacob lana -—————— A | prspraannslang — = || Te aekue oe | PP sag. tre _tusthing = plone |. ~ : { te | tol scritionn (se fanst drs, the. eae | Arsuikewenthe fh. obtaha the LP Avforxatme - = . al ave ¢ a loti ue 5 —o_fechiorod Rk, 4 f f La Gtrodueod 4, . oe 2 tke optima poe ty CP ‘ oo et agtnod prmblena ts £0lucdl— een aoe Are hin mre Nat vari He drone [reece kin. Cin aa 2a / ‘ol | of2 Yee vaniod Oo, as) 7 Scanned with CamScannerIntroduction:- In linear programming problem, the variables are allowed to take any real or fractional value, however in the real life problem the fractional values of variables has no significance, for example it does not make sense that 3.5 workers are required for the project, 2.4 machines required for the work shop ect., The integer solution is obtained by rounding off the optimal value of the variables to the nearest integer. This approach is very easy, however it may not satisfies all the given constrains and also the value of the objective function so obtained may not be optimal. This type of difficulties can be avoided if the given problem is solved by integer programming method. The integer linear programming problems are those in which some or all the variables are restricted to integer value (discrete value). Scanned with CamScannerStandard form of Pure Integer Programming Problem The pure integer programming problem in its standard form can be written as follows Max Z=c1X1+C2X2+....CnXn, Subject to the constraints a11X1+a12X2+.. .nXn=D1 21X1+822K2+.. ankn=D2 ASIF AG2K2+ seen ee3nXn= D3 MTX + AMAA .veeeseeemaXn= Dm and Xz, X2, X3,..Xn20 and are integers. Scanned with CamScannerCut A cut Is a linear constraint added to the given LP Problem, it is also called additional linear constraint.( or fractional cut) Enumeration and Cutting Plane Solution Concept The cutting plane method is used to solve integer linear programming problem The cutting plane method was developed by R.E. Gomory in the year 1956. The cutting plane method is based on the generation of sequence of linear inequality called cut. Scanned with CamScannerProcedure of Gomory's Algorithm. Step:-01 Solve the given LP problem using simplex method by ignoring integer requirements. Step:-02 If all the variables in the Xs column of the final simplex table is non negative integer, then the current solution is optimal. otherwise go to next step. Step:-03 If some of the basic variables do not have non negative integer value, then an additional linear constraints called Gomory's linear constraints (cut) is generated, after having generated linear constraints (cutting plane) , it is added to the bottom of the optimum simple table so that the solution no longer remains feasible. Step:-04 The new problem is then solved by dual simplex method, If the optimum solution , so obtained is integer. Which is the required one, otherwise repeat Step: 03 until all basic variables assume non negative integer values. Scanned with CamScannerExample-1: Find the optimum integer solution to the following linear programming problem: Max. z=x,+x2 Subject to the constraints 3x, +2x2 55 x2 <2 where x12 2 0 and x;,x2 are integers Solution: Convert all constraints into equations by using slack variabless, = 0,5 2 0, we get Max. Z = 3x, + 5x2 + Os; + 0s Scanned with CamScannerSubject to the constraints 3x, +2x2 +5, =5 x2 +5, =2 X1X2,S1,52 20 Table-12.1: Initial Simplex Table B Cs|G]!] 1 | 0] 0 [%/m m>0 Xp | x1 | x2 | Si | Sz nv ° Sy 0/5/23) 5/3— s of;2/o] 1 fo]t Z=0 |df-1t] -! | 0] 0 [Xp/m m>0 Xy 145/3] 1 2/3 | 1/3 | -2/3 | 5/2 s o|2/0/PJ]oj]1 |2a Z=0 4, | 0 | -13T) 13 | 1/3 xy 1/13) 1 0 | 13 | -2/3 x {2/0/11 [o0]1t Z= 7/3 0} 0/18] 13 Since all Aj> 0, so the optimal solution is xy = 1/3,x2 = 2 and Max. 2=7/3 To obtain an optimum integer solution, we have to add Gomorian Constraints: 1 Mm=f, %2=2 Max. Fractional (2,0) = 3 Since, } corresponding to the first row i.e. x3 is the source row. ‘ ; a m2 ; Now, expressing the negative fraction + in the source row as a sum of negative integer and positive fraction. 1 =-14} Since, x, is the source row, we have 1 1 1 fox +0n+is,4+(a+))s, 3 Scanned with CamScanner1 1 1 gauitOr+5s1—s2 +582 eyed 3taF5 The fractional cut constraint is given by 2,2 514 4 _ 2 ot 3t3F3 Psat Gas Where, G is the Gomory Slack. Adding Gomory Constraint at the bottom of optimum simplex table 12.1. Table-12,2 B Ce) G }1it 0 0 |} 0 Xp | x1} %2/ S21 | S2 | Gr xy 1] 13/1/04] 1/3 |-23] 0 x t/2/o;li|ofi lo G,0f-13] 0] 0 fema|-13 7 Z= 73 | a |0/0/iat| iB] oO Apply Dual Simplex Method: Outgoing Vector Xee = "" (pu; Xa < 0} = Min{-1/3} = —1/3(xp1) Incoming Vector roa fay, = Max{24., 44) = —Max{-1,-1) or Gag Bn<0} = Mente as , There is a tie, so we can select anyone of them incoming vector. So, we choose sybe incoming vector. And, applying usual simplex method. Table-12.3 Bo Cp/G)1)1)0};0];0 Xp | a | X2| 81] 52 |G mm 1/0|1/ojo;-afi x -!/2/0/1|oJ/1fo Scanned with CamScannero|o0 Since all 4;> 0 and all Xz > Oand all restricted variables x,,x2 are integers. Hence, an optimum integer solution is x, = 0, x2 = 2 and Max. Z=2 Example-2: The owner of a ready-made garments stores two types of shirts known as Zee- Shirts and Button-down shirts. He makes a profit of Rs. 1 and Rs. 4 per shirt on Zee-shirts and Button-shirts respectively. He has two Tailors A and B at his disposal to stitch the shirts. Tailor A and Tailor B_ can devote at the most 7 hours and 15 hours per-day respectively. Both these shirts are to be stitched by both tailors. Tailor A and Tailor B spend two hours and five hours respectively in stitching Zee-shirt, and four hours and three hours respectively in stitching a Button-down shirt. How many shirts of both the types should be stitched in order to maximize daily profit? (a) Set-up and solve the linear programming problem. (b) If the optimal solution is not integer-valued, use Gomory’s technique to derive the optimal integer solution. Solution: (a) Let x,=Zee-shirts and x2 =Button-down shirts ‘Then the linear programming problem becomes Max. z=x,+4x2 Subject to the constraints 2x, + 4x2 <7 5x, 43x, <15 where x1,x, 2 O and x1,X2 are integers (b)Convert all constraints into equations by using slack variables s, > 0,s2 >0 and then solving the linear programming problem by usual simplex method, the optimum solution is given by in the following table, Table-12.4 Bo Gp G |1]4]o]o Xp X1 | Xz Ss 52 mm 4 | 7 |i2|1 lial o Sz 0 39/4 | 7/2) 0 | -34] 1 Z=7 a |! lolifo Since all A;> 0. So optimal solution is x2 = 4,5, = O and Max.Z=7 To obtain an optimum integer solution, we have to add Gomorian Constraints: 7 =0, x2=2 (+3) Scanned with CamScannerMax. Fractional (0, Since, Zcorresponding to the first row i.e. x3 is the source row. Since, x, is the source row, we have P=B4r tis, +05, Belge? The factions cut constraint is given by Seine pe -B-intG Where, G, is the Gomory Slack. Adding Gomory Constraint atthe bottom of optimum simplex table 12.4. Table-12.5 B G]G |! ]4]0]0]o Xp | x1 [x2] $1 | 52 | Gr x 4/74] %l1{ molto s 0 [394 0]-34]1 [0 GO| -34 0|-4foyt Z=7 |g} jolt fofo Apply Dual Simplex Method: Outgoing Vector Xpe = “Y" (xa; Xp < 0} = Min{—-3/4) = —3/4(xp5) Incoming Vector Maxey i Rar gi *Bys0 = Max{. Apply usual simplex method ‘Table-12.6 Bo Gp/G][1[4] 0 ;oyo Scanned with CamScannerx2 4/1 }0/1] 0 /0]1 s. 0/92]0 0 xy 1 [32/1 ]0/12]/0/-2 0 Z= 11/2) 4, | 0 2}0| 2 Since, x, is still not an integer ke., x,is source row. To obtain an optimum integer solution, we have to add Gomorian Constraints: ae! m=30+) Max. Fractional = ; Since, Jeorresponding to the third row i.e. x, is the source row. Since, x, is the source row, we have 1 1 3 =%1 + 0x2 +55; + 0s, + 06, i oa tet 21= 2 The fractional cut constraint is given by 1 1 pa tG Where, G2 is the Gomory Slack. Adding Gomory Constraint at the bottom of optimum simplex table 12.6, Table-12.7 B Gs]G]1]4] 0 ]0,0]0 Xp [xi[x2| s | 82|G:| Ge xz 4/1 {0/10 jolifo s 0)/92]0 T{7]{o x 1 (32/1 o;2]}0 G0 /-I2/0 ofoft Z= 12] 4, [0] 0/iatfol2/0 Apply Dual Simplex Method: Outgoing Vector Xax = "I" (xp; Xp < 0} = Min{-1/2} = -1/2 Scanned with CamScannerIncoming Vector Max; 4 Xp = {Fiix 0 and all Xp > Oand all restricted variables x,,x2 are integers. Hence, an optimum integer solution is xy = 1, x2 = 1 and Max. Z=5. ‘Thus, the owner of ready-made garments should produce one Zee-shirt and also one Button- down shirt in order to get the maximum profit of Rs. 5. Scanned with CamScanner_—ss | The mewn of A Ate ocky roche. urn ty__ateie |! p2ckts trum types ef plenftiny shiath, Brown a5 | | een ashisty ke Staa- is htule. He propos % To Hl PR 20D LV 3O0_per shirt om zee 2 Stor} _ Pabst _Acap. Me hoy two toile, AA By! ——| cat hig abort te stich the shins. Tailor, LA deanbe a otal of oheruis|day while lene give ot tra mest (5 houry a doy: typo ef ALTA Ore stitched “by both the falters. The tyro waded for eg see tla no eua by faite Ae 3 Itaifes § + Hes oo ahwds ot. Caal type | Lwherdod be atihed ¢ ekdos_fo_masina’3@ dott 2 Corwulote Gb as an 1PP | 1 Pee 1 OP SAO lAhintz be 2; & 4, Aersb. The aboue ay |dote Coim he oberpieded ax m= Moyinnt pe 2 = Dov x +300 2 Dobie t tr eee eel Bn, ob 3x, S 15 Hy yh 20 tarkey2d Tee LP Aetogntinn pf tte prebten — Scanned with CamScanner| Moriwnise. " 2 2on4, + 300%, -__ Awlfect to Be pe Moe, ANT BES estas eS) XG FO BISESSst tits LP Ardagatton with bivaposy aowetherd . | Qnty oduc¢ M3 the Alack Vartotles we bawre Marivakee “Zs 200%, +30 %, +05, +08, Aubject te DH p+ Wy +A = IT Boy + Bx, + A, = IS Ai % 4, 5, 20 Scanned with CamScannercy | 200 300 5 ° Se” | c-B- | BV my Ae Bz, ae ibe { 300 | % | O | a. af | i 200 | 4% \ ° —Y2 2/2 3a i _z | 200 | 300. | 50 | Ise/s | ias0 | | G74} D1 _o =50_|-190/3 | 1 = Bat in Pa nett — ga the pptinad sshclinn | al 2 i850) [Sie ee satus ol tre voniobles aro net Le deen net Aebrorosh the gatictor _| fit the IPP. Scanned with CamScanner |Ie apply the calking— place mac thd 5. ise epta by choosing Sc “Compbrsin’ the the LP jPAelarcsctien's u \ tha -ofptinaa sh —___ sctucten in which oa bante Santable JA deactimal. ‘ Calpe Ones ee lee ee lee eo) 2 3 We wil nous ene barn CONT ass in the fon o A+ 0)n, 0%, + (=) +¥e)6.+ orale le Pode. aK dn | iain Aida b olf termd unt the _| proc Morol re pptete sect ma the 1-H: % OASeb — 1 _ = = +’ 1h hs b+) my .) . Lon trink Pf thy 29 oa. “Yn, 2/9 by = Mat Bone fact ict per. _| | rp ee erat tna ' pomae, Creches peat pant ae boyge a 7 Yo by + 43 £5 2 Yo. Scanned with CamScanner— The costting. franc “tain nor Augjeate d Haddiong. eet te hee oe d i Acta xation's optimal ADL bobo ar =o 6-7/2 A, <= - Yo : a a4 3b, tts oa ot Such a comtsaint is Bnew aro Gunatiny Hemaxbraint oA cut. Etast ol these, (a thot tho tunrowt opXiroal sel” te “the LP relaxation phalé Inet Aartts fy tho Ciab- EL. thro tet” tn Table 2 bos 0 £2 4, =0, tnrore VoIuos de not Aorta fy. the, cut wileieh ALgArtnen Yo b, + 2/3 6, 2 Yo. bt ty tre. | Pps Whi ott ts added tothe LP LArtonotion, we e¢bect te obtara a_s9e”; Volues . Tp Ue ty 66, use Aare focunol the! | whore, al? varia blor assumo datezer | | lL epti mad 4el™ to te Alaina 2 | PP. Scanned with CamScanner° a ¥ - = Ty the vess obtimal snl™ hos ati seme | Ana cH mand values s. nother. CL wy -genenadeal a | © tie precoss ta Comtinueds ‘Scanned with CamScannerNeus. pawiee ding a A tn the. hata’ LP prcthowa , wre get the Aorsdredk +nbhe ar — Scanned with CamScannerat Wan hy Ay |} detrocdtting ot, Crate —2—ALoeR vont pte [fn the, abe Sy vse. 9e ~ Scanned with CamScannero fe eeo x, 1 a =i 10 1} oyu f= O° ' | r cy | 200] 300] o | 0 o | © a” T Fee. fave] x | me | ar | Az [bs | Ay be 200 || %. Oo t oO 2 Il) t 3 | | 200] m4 | 1 o | o | o 15/3 |B} 2 [ [ols toto }o fl 1 I-1 |-2} © 4 Ay O ° \ oO [23 |-%/3 \ Zj_| 200 | 300 | 0 O | joo | 19913 | (B00 f Kj-z] © | o | 0 | 6 _[H1o0/s |-to0/3 _L—}--_+_F f Ero the schuton it weay be chhersed that — eat ef ne boar variobtes has om inbeges Ae” value. Accordingly, the sol” ix the dsasired. obtimal sel” te the IPP. \ Mo? x, 23 b Zz 1200 Scanned with CamScannerGAIA Srna ctr of toys maken duor tupen $j ey, A&B. Pracennl to — Gf there ture teya is dene en “hop wachinen K & Y. Toy A Acysines 4 hewes 0 machine % La six% ews ien metrincY, iabite Tay 8 Acpinen | 8 heurs on marhing % £ Keurs er wathine. Y Thete_one thivby,

---

Of tanche gon _ |] We comp tder tip LP relaxation £ are by dtvenphey wre Hod May traiae Ze 30% jb 30%, OA) + On Scanned with CamScannerTobe | eal G 40 30 0 ° sot” | Dvn | ree | B-V- mm My Ay Ar br oti o) 8 4, / 4 | 8s} | | 9 4 3 é o | 42 | © | © ° | 30 — 23; 1 o oO ° © 8 G-25 | 20 30 oO a fet : - Foe 2D — q | 30 [30] © O_s[ soe | mast cb ev mH | x | Ar Ae | bi | Rak rb ° May] v7 [-2/a | 12 | 87% 30 s L ° A ° V6 | S | 6 | | zi_| 30 a } | Ss | !5d | | ICs — ° 5 Oo | -5 3 at Tobe 3 j 2} © | sot | A Pe To | By | =q | 18/4 _| 5/28 | 2/7 _| 20/5 IST | 30/4 j nyolq | een |-30/3 { LP As fa och ton y fo Ee p22 = 1a ke Z = \\Yo/7 Scanned with CamScanner[cinco the variables ore nat integer, it | dees nat reprsdromt Une pol tothe | PP. the put Sood cians. thre cut frown Cass —_ _ Ox, bo +3 i py dy calles 34 4 fte6\p, = 24% dye (tts + 34, + (tS \ by Ih Zs, + 6 5 = Ya [a+s, -%,) nt 2 7 9 Acvording hey Boxe 6p DS Vo wT oe ot =3 A, = 6 24 < —4G M ss eu =r = es a, £ By + Ba Me tr orsined tabla ob Testa Gl27 130} of o lo | se" | muh [ Ce ew | 4 1% | A, [oe on ae jRetle | 30 || % vt tuyl=Va! oo Batis | 0 || xy O4-5)231 2f7) 9 |aofy | — 7) 0 Ast ° o IC -4/4 | 1 a | 8/3 sal | | ef 30 | 50 | stl Fo Paw pO Gz o | o [9/4 |-3o/4}-0 [> duakerr | - | -| sts fo yo™ - eae Te | ~~ T Scanned with CamScannerTable S ant | a ee ee ff Loy te} 3°) 9} of og = c8 bv | % me 2 =i A, Ay i | x0 | x | O | 1. fo tly i 30 [ou rpo to fi belo, | Oo | 4. | 0 5 | MEM | aps | 2:.| 30 [30 |-o [0 |e | 10 | OS ae c-3 | oO lo} 0 | o = rn Again , the borle vuplables dno hosing v - - . Fra ctoyol velies. se = Creakor — 4 + lg, - 5 6, = lof ln, + Dx, + Os, £18 = + Oke ¢él +e = SAL (It Ox, 40x, +04, + (14 Ota +f + : | (eu /250 Se) | 6 3 i | \ ae a tS | 6 3 | A -Vé 4, < -V2 f OV dy +dy = ~/2 Scanned with CamScannerre sf! Tr » | L o |o | Soe C8 Ax 5 by bo 30 ell é i Oo i 30 fond x t SS -23 las | -—6 2 Lo 30 150 -30} 1) Scanned with CamScannerTy — 2a pS) G3 || The dicteian at the local ene | Ab blannteg the breakfast Irena for. the anatey Me ing Smear ‘ ‘ Cena ty Lumad patients. She is peawntong | , Or Abe af) seme fottemsing Wel & has chosen stage CAcese. & Acraambtaol eqy for brah past + She [in_prio erly tance rast wits Vitoria 6 2 Aen negutiencenatn— bA _ploeamniing. the -braokife i L tre Stake Meditcas Ana Biot LSMA) mane mothers mamrb gek ot Sense (emg of Vitwaalia E & 24 vaiph of Utena b- MA has dbook Aeberts "| eh tolnn | en Vidura inn € Se 3 meg of tA dept. Kyrotes that a ACsvp Of corttewpe tn00 be. Crvty "EQ fe Ata. af Acrasnb0od 235-—} oS af otth (okt cian is CN Rencptivt — 46 Abledmine tre opty rtm brorjast ren that Dodson of the scpdronects + HL mninindren TO. The tem taniath that AC can ALnase. fooda by AC oD, than neceosite| pecking. Etenen. bpp 4 Scanned with CamScannerLet 2, & 0, be thy es ee of crthage lL cheese Qe scramabied ey. Aeaheettay ATA _—— Dinuwae ZZ = 2 9G + 29) —dubject to 3x, + 2x, 2 - 3, + A,, A, 20 inthe penr. i don Lang. the LP Aol od on of the Gisston Pp C deliing & by. Airaptere wretna d. vs a Miws 2, + 2%, aie . Hi +2%y 2 | 2 3m, +8% 224 7 ees Hn, 2 O Mans otPon ye hae - As TEs Fotrchisting tre bursts b arti cot plead ge 2H, tp 2m, Ob, +08, OK, » Aubiytd ts: EMA MA, “MAY —2.Ay tr 2, 3%) + Be, Scanned with CamScanner| Scanned with CamScanner ott ia 1 4 i * 3) 2) | JI ed | UY gia] vol 1 = Sl Z| wo) Ol N = if ‘=| 5 ela ale S| | AF & 8) | Obata - ‘el ol RAG lS i Ty f. eI] 7] oo f£]° w [FI) g 0 €]/L}°} o] -| clo é @ x] PIS £|° E) €} -| o| ©] <]o Ole ee z¢ 9) &) Of of TLS E of <|=) ele 4 of a] ze] 2] oRelY] ‘| 4 “|| 2 [stl se" o| <] 7 zl< FL 0) <] 7] o} of Ele a T= — A} g} Oo} -| of al] o 81 2S |} f] o -fojafo elf |r| col Alar ele] Se al x] Of of -f-d} o a o] 2] FOL zs Ny onleileg | «| oN sal 4 a slg My Ja] 4 21 Se] x] Rk] NT La oS NS oe J —~T | 3] el dle s] ej alaT | sot” _ & | -S5y" oo | 3/3, - 2%] 4 Te shiiwality Aeachad the optic I aap 34 w= 8) A, 22 £ Z= 28/3 + oxy Mota the ect po horg buat i ArytAL rh alne aa the integer Stved as the Cut jain tre 59 Cetra ttn — Noe, + On, — m Be etd +04, = 8/ I (ec\y, tox + (ay 4 7 = tiie, +04, = 2 Yh, a & 2+ = Ss i = I gta te = 2 4(2-%, +4,) tc Beene Sb, 4 Scanned with CamScanner{ ae a aa Trdrodu ting thee tuk Table geb te Aetred table on — a : _ Tobhe- 5 Te 2| 0 olo Lo | [ c-8-f] Bev | % | %> A, | 42 | Ay | 4y | Lo ls,| © | o Fal vals [oT 2 lie l ve|-“% | o | a | \ 2x] 1 | o |v} Ya} © | © [o {2 1 | o =SAICA eo ft | zl 2} 2 [3/9i-¥aio | © | -z| 0 | 0 |e4{%a lo] o | Qustert | o | — |-t j=i |= | O | i eps __t- SRA Te got “re act” stented bo cane of winimise rTP e Aisidea p= 24 Votes. by. te. Ing (= we wad tte [econ Se ahs ai lad rape. Apaprlevt as tre earckenin vanrtable 7 { _ Hero. eaten ing vantabte = Ao bud gating, orLable — Sy L News Pete tlory ort a cam of merwral . po nae ee 4 tob2o 0» — —t Scanned with CamScanner| | [ GS 2 m Oo} 9] : om Gok ‘S _ fe@ffaw | xm |u| dM] se] Ae | es bi | a OM [Sas Ove (eles es Om Re eee Omg) i = 2a, oft] 1 tolo [x] 3 | | 2 | x I oO al ° ° —4 a5 [- 32 |°-0 oO. vd of | ~ \ q 6 | | bei 2)1 215 1o |o |-w| jo | 5-25 ole flo fo o J 1s | The Chitlned ae” of toe gon IPP i ab ir Ls XK = 3 | (2a PUP Oe VEY) ! : = Ye Ty - = | =Flo Scanned with CamScanner