" Ene -BA, Aco) Gata. ' cass -8Y.BBA(S) | | FACULTY NAME : MANISHABEN SHAH | | suevect , : MATHS FOR MGT junit B : LINEAR PROGRAMMING i JL ote tp B29 eeMaths for Mgt. Class : FYBBA (Gen) Faculty Name : Manisha Shah Introduction: t 4s was handed over the task of During the Second World War, a team of scien! developing scientific technique for obtaining eo a {ie. output or efficiency) using limited resources. The techniques developed by the scientists at that time for obtaining advantage"in the military operations cgme to be known as ‘Operations Research" (or O.R.) Thus operations research is a Bgllections of scientific technique Used to arive af an optimum decision in the problems. éf¥yanagerial nature. m getionier period its use Following the success of O.R. techniques in war, spread in the industrial and other administrative revolution. Theré was a large scale industrial product activities was required fo achieve the ulti jum cost. The techniques of operat ‘the Blanning ot different ieving maximum profit,or g to achieve this goal. telhod, technique and tools to provide to those in control of the “Operation Research is the appl problems involving the operationgof otem so system with optimum solutions fo IN prgblems": problem of filocating scarce resource fo the ac tivates The Programmixg problem; fo be performed. Such a» robles are concerned with optimizing | Le. maximizing or jeclives such os minimizing cost, maximizing profit etc. the minimizing ) the desir collection of te; ysedsfor solving problems is called mathematical ranting, a simple and most commonly used technique, is @ programming. tiebr pl Particulareg: sal programming. ‘mathematical modell. a set of equations and inequaiitie’ ) liner’ means that all the the powers of all the Liner Programmiigg u to describe theNgortemed problem. The adjective’ mathematical equdlions or inequalities in this model are liner i.e. variables involved are not greater than one and there are no terms involving the products of the variable in them. The word ‘programming’ is another wora for Thus, liner programming involves the planning of activates 10 oblain an planning’ optimal(best) result from emong all feasible alteiative. ‘Shri V.Z. Patel Commerce Collegetees! rs Maths for Mgt. A In any industry the limited resource and capital should be allocated in sucha way that . eptimum_adyanfage can be achieved. This ailvantage can be Maximum profit or i As the resource are limited, we have various Constraints oF limitations regarding available resource. The constraints can mathematically be expressed in the form of inequalities, leaning of Liner Programming: Linear programming involves optin Certain liner function called the objec For example, a manager of a factory linear constraints iding thé appropriate Productmix of three producis. He Products, has data regar ofitability-of these the requirement if raw materials, avail ] labour etc. His problem can be formulated as a linear programttt Iblen{faking the objective is mdite pr t-mix keeping in ‘ations (1. maximizations or ive functions subject fo certain May face a problem of dec} minimizations) of od functions as the maximisation of profits obtainable view the various constraints - the availability of raw maf cl. capital. Labour supply etc, 3 In short, linear Programming is a technique for Scere ‘an optimum schedule of interdependent activities in view of the ovo eo ariables,8e nor shegalive in not vandt'large: ar. \ Assumptions of near Prograniming (1) The value of the decision v (2) The number of cons (3] The objective functidfs ipdihe (4) The constraints vec, s [5] The numbeigoR cggiiguctiables is grater then or equal to the number of i constraints : ine ramming : 4 lo express the real life objective function press ihe real lle constcints of limitations os finer inequalities, fficien!s of decision variables are probabilistic or uncertain, the of linear Brogramming cannot be applied. (4) The technique of linear programming takes int. '! cannot be applied in the case of a problem 4s a linear function. lo account a single objective only, having multiple objectivesMaths for Mg available for solving linear progiamming problems involving @ large number o' Zecision variables be ; Te of tiner Programming: 1 inear programming is the popular and widely used technique of‘decision making in Dusiness and industry. Some of the oreas of ils applications are discuss below. i) Industry : The main purpose of linear programming is allocation of limited Fesources so a5 10 maximize profil or minimize cost. As we know, in industry the fesource of capital, man-power, machine-time ele, are limited, Linear programming help in the optimum allocation of these limited resource Product produced by various industries are stocke Ge" jhouses and they are to be distributed as per demand requirements ‘dgfilels. or retail stores Generally the source of demand and the source o 03 it geographically distant place. Hence the distribution of the stocked Costs, time and other such condition deleimine minimunt cls Involves transportation ost shipping progremme. l2agicuture : The citferent pol of land in Bp ogicule feim con be olocoled 10 _dillerent.crops, in order 10 get mgximum baefit of db optimum muro ot alftewsy ferllizers can be determined so aso gel matipugfProduce. The technique of linear Programming is quit useful if such problem. ® the bes! strategy using limited number of Janes and-toldiers in a war. Even the transportation ye ore Useful in some problems related to war, Sy, | {3) War : Liner programming help in déciaf /9nms-ammunations, vehi technique and assignment (4) Production may WALigg>r programming solves production management problems suc! luction scheduling. product mix, assembly fine balancing, Pro uatior e PRBbIeM stc. (s)Fingli iSPmanagerepNahen problem of allocation of available internal or extemal fund for, yar lodetion activity in the optimum manner can be solved by inoor progiditing Wechikique. Aso the problem of portfolio selection for investment in various compari in order fo get maximum return. or 10 have minimum risk canbe solved by lin8@? programming, (4)Petsonal management : the pioblem of allocation of proper persons to proper jobs so @5 10 maximum folal efficiency can be solved by linear prcgramming techniques Shri V.Z. Patel Commerce College Page 3al fore i Maths for Mgt. Thus linear programming can be used successtully in the solvation of problem concerned with blending of material, disibution and transportation of raw materials of Produced goads. allocation of personnel anil investment portfolios. ‘Mathematical form of a general linear programming problem: Suppose miinear inequalities involving n variable x1,x2___ x» are given: We wish to find non negative | either positive or zero } values of these n variable which optimize ( ‘maximize or minimum } a given linear function of these n variables. {in general terms, a linear programming problem | LP.P,) can mal matically be writen 3 follows, Find the valves of the variables x3,2__ «which Zeer tea arte ten te Subject to the restrictions yy Xy + Oz Xa tt gy TS Oy e faxton tt tan tes day |S \ ® (2) Oma 4 + Oa X24 and BEERS 20a 3 Poy 1own constants The above problem can be of minimization of (1 also in which case < sign in (2) are to be replaced BY® signs; other expression remain as they are. (3) We can write. the above linear programming problem in the matrix form also. Suppose Shri V.Z. Patel Commerce College aOy Ou ~~ te El Cr eer coy « ba Now, the above finer programming problem can be stated as follows. Find a colsrnn vectgso a1 to Vorimize Zm etx Subject fo the ts Arsh * And ne 0 Some d (1) Constraints p The near inequatiies of equations representing tne liitations dee $f Ihe Given near programming problem are knows a1 constraints, (2) Objective functions : The Inear function of the variable which flo be maximized or fled the © ve fine (3) Solution ; A set of w ing the constraints (2) is cated a Shri V Z Patel Commerce College Page 5uy Maths for Mgt. ie (4) Feasible solution : A set of non-negative values of the variables wich satishies the Constraints (2) is called a feasible solution. OR A set of. values of the variable which sabsfy the constraints (2) and the non negativity restrictions (3) is called a feasible solution. (5)Basie solution : A basic solution is a solulion obtained by setting any n~m vanable equal to zero and solving the m equations for the remain m variables, provided the determinant of the coefficients of these m variables is nonzero | as we know, Ine number of variables nis greater then the number of equations:zm ). Such m variable are called basic variables and the remaining n—m variablasyare called non-basic voriables. & (4)Basic feasible solution : A basic solution which sal normggaalivity restrictions {3)is called a basic feasible solution. ‘OR # ‘A basic solution in which the valve of the basidlyariablegpre positive or zero is called o basic feasible solution, (7)Optimum basic feasible solution : A basi sible solution which maximizes or minimizes the objective function of IhagiveMlnear progromming problem is called the optimim basic feasiy olution. Thus, 0 batle soliton whi egies (3) ond polimises the objective function in (1) este solution (8) Non-degener feGsible solullan : A basic solution in which all the m basic variable, i pottiye aNd remaining n-m variables are zero is called a non- degenerdia pattysatsible solution. (9) Degenerdte,basic feasible solution : A basic feasible solution in which one or more baxic variables oe zero is called a degenerate basi. feosble solution, Basie theorem of liner programming The maximum and minimum value cf a linear function ax, +bx;+¢ defime in a convex polygon $ occuss at (1) One of the vertices of S or ‘Shri V.Z. Patel Commerce Collegeaths tor Mgt ~ {2} ft occurs al any tow adjacent vertices of §, then if occurs af ail fhe point of the line segment joining the two vertices. Mathematical formulation of a linear programming p'oblem : We consider the following ilustrion to discuss he mathematical formulation of a finer programming problem. Mlustration: Gujarat Chemical manufactures low product say, naptha OH) ‘and urec 4) in its plant located in the outskirts of Vadodara. fii go! a Prosi offs. 7@,per unit for naptha and Rs 100 per unit for urea, The time requiemenls for edch pioductiand the total avaiable time in each department are show in the folloggPy}cble ‘oducl hes fo poss through two department in the paint . Department Total required per unit (in falta fa month hours) Haphia [Urea - 1 30 40 * & 2.100 2 40 3.0 * 2.100. Foriher, he demand for these prodget rest fp progion To G maximum Of 450 unit of each product. Express this Eneor pr@grammin mathematically (PU. March 2000) * Suppose Gujarat Chemicals manvtactures’tX,) unit of naptha and (¥4) units of urea. IfZ is the total profit obtain, th Yar function of the profit? wil be as follows. Z= 10x, + 1002; 2 Me tPlo department gives us the following linear restrictions. 4x, + 327182: The eee, of the two products results into the following restriction. x, $450 x, $450 Shri V.Z. Patel Commerce College Page? G ( a7 oOAWFiths for Mgt. As the unit of production of the two chemicals cannot be negative. We have the following non-negative restrictions, x, 20, x, 20 Thus given problem can mathematically be expressed as follows Maximize Z = 70x, + 100x;... Up Subject to the restrictions 3x, 44x; $ 2100, (2) 58 (3) fe (4) 4x, + 3x; $2100... 2 $450... x, $450 x, 20,2, 20. (6) Methods of solving linear pr aS 19 probheg $y: There are two method for solving ingly prjamming Problems (1) Graphical method (2) Simple meth Simple linear progray ms involving only two variable can be solved by the graphical sab ie graphical method has limited scope due to iis handling of Faritpie at time. The, In iterative procedure for solving the linear programming lene Tht ‘as developed by G.B. Dantzig. This method consists of basicfei ioh to the equations { Which are obtained from the given ineaualilies byfadding suitable slack or surplus variables } and then testing this solution il is optimal of not. If the first basic feasible solution in not optimal then i is to be improved by a sel of rules. This process is repeated until {an optimal solution is obtained. The computation procedure requires m non- zero basic variables at any time. If at any stage we get than variables. We come across a degenerate problem. m non-ze Shri V.Z. Patel Commerce Coll Page 8Maths for Mgt Graphical Method: bles in a liner programming problem. Graphi can be used; We first convert the inequalities in the given linear programminc problem into equations by taking equality signs in place of inequality sign. These linear equations represent straight lines. There graphs are drawg, on the same graph paper. As per the non-negalivily restrictions +, 2 0 and. x, 20, the feasible solutions occur in the first quadrant. The set of points which are common to the regions salisfied by the given constraints ( inequelfjes:) is a convex polygon in the first quadrant. As per basic theorem of lin ‘ogramining, the optimum feasible solution of the liner programmigg, proftety, oSgurs at one of. the vertical of this convex polygon. If the og luyentzoccurs at two ‘adjacent vertices of this polygon, any point on D semeh joining these two vertices also gives the optimum feasible solutio mM - When there two varial We obtain the co-ordinates of thegtelfices of the Belygon by solving the equations of the corresponding sides. WeMhen fing the value of the objective function at different verticesby subsputing She co-ordinates’ of the corresponding vertex in theobjedye func! values of the variables x, and x, for which the ceca n is apfimum (i.e. maximum or minimum) represent the optimal solution of the A liner linear programming probiem. Now we solve the liner 9 ming problem given in ‘the earlier illustration by graphical method. i. Maths for Mgt. ———————————————— The graph of equation (7) isa straight line. We requite at least two points to draw @ line, Taking x, =0 in (7), we get x4, =700.Also taking x,=0 in (7) we get x, = 525, Thus the line represented by equation (7) can be obtained by joining the points (700.0 } and (0,525 } Similarly to draw the graph of the line represented by equation (8). We take x; = 2.and get x; = 525. Alo x, = 0 gives x; = 700. Thus the line represented by ) aad (0,700) equation (8) can be obtained by joining the points (525 The line represented by the equation (10) is a straight ling passing through the point (450,0) and perpendicular to the X-axis The line represented by the equation (10) is a’ point (0,450) and parallel to the x-axis. Now . we obtain the regions represented hy different intg 4 s x Substituting the co ordinates of the origin Nyt Ax, $2100,we find 0+05 2100 or 0S 2100 which i irue, The sel ofgWhls represented by this inequality is the region on the origin side of the fine, ‘sx, + 4x, 5 2100. at-Jhe set of poial represented by the inequality Similarly, we can see 3 fn ori {he side of the line 4x, + 3x; S 2100. 4x, + 3x; S 2100is substituting the ovat of O (0.6) in x, $450 which is true. Hence the set of points by", < 450 is the 1egion on the origin side of the line x, 345 The ineQuoltyxke 0, gives a reason consisting of the x, =0, and the set of ht hand side of x; axis. Similarly the inequality x, 20, given o point off the. region consisting of the line x, = 0, and the set of point on the upper side of xy 20, and axis. The intersection of the regions given by the two inequalities x, = x; 2 0,is the quadrant. Shn V.Z. Patel Commerce Collegeven by the inequalities (2). (3). (4). (5) and {4) 1 a the figure by the shaded aiea. The intersection of the region gi : convex polygon OABCDE which is shown in : ale of A are (450 0}. The co- Now , the co-ordinate of @ are [0 0). The colordin co ordinate of point B can be obtained by solving x, = 450 and 4x, + 3x; Which ore { 450 100).The co-ordinale of C can be obiain by solving the equations 3x, + 4x; = 2100 and 4x1 43x; = 2100 which are { 00 ool The co- ordinates of D can be oblained by solving x, = 450 and 3x + 4x4'= 2100 , which are (100,450). The co-ordinates of E are (0,450) Now . as we have seen earlier, the optimum solugi of one of the vertices of the convex polygor tegions satisfied by the given inequaliies. Hen! objective function Z = 70x, + 100x, a (ferent ver OABCDE 29 AT 0( 0,0) Z=70 X0+100 %0=0 2 At A(450,0) 2 =70 x 450. = 315 At B(450,100) Z= 70 x 450-4100 At €(300300) z= 70 x so F100 x 300 = 51,000 At D(100,450) Z= Chea = 52,000" it £(0.4! & 10 X 450 = 45,000 sons 5 s orth The aye Yation we can see that the maximum profit is Rs. 52000 andi iscoblaigble%at the point D(100.450). We can verify that all the four constrain and}pe%non-negalivily restrictions are satisfied by the co-ordinates of this pointy, this point gives optimum solution. the company should manufacture 100 units of naptha and 450 unils of urea in order to ec maximum profit of Rs 52,000 ae Slack and Super variables : Let_one of the constraints of an L.P.P. be Shri Vz Patel Commerce College - : PagelGem GL rycun > sony Syca tm’) 9 _TABNBPORTATION PROBLEM 9.1 INTRODUCTION One important application of linear programming is in the area of physical distribution (transportation) ‘of goods and services from several supply origins to several demand destinations. It is easy to express 1 transportation problem mathematically in terms of an LP model, which can be solved by the simplex method, But because it involves a large number of variables and constraints, it takes a long time to solve However, transportation algorithms, namely the Stepping Stone Method and the MODI (modified distribution) Afethod, have been developed for this purpose. ‘The structure of transportation problem involves a large number of shipping routes from several supply origins to several demand destinations. The objective is to determine the number of units which should be shipped from each origin to each destination in order to satisfy the required quantity of goods or services ‘at each demand destination, within the limited quantity of goods or services available at each supply origin, at the minimum transportation cost and/or time. ‘The transportation algorithm discussed in this chapter is applied to minimise the total cost of transporting a homogeneous commodity from supply origins to demand destinations. However, it can also bbe applied to the maximisation of some total value or utility, for example, financial resource’ are distributed in such a way that the profitable return is maximised. ‘There ate various types of transportation models and the simplest of them was first presented by F L Hitchcock (1941). It was further developed by T C Koopmans (1949) and G B Dantzig (1951). Several extensions of transportation model and methods have been subsequently developed. 9.2. MATHEMATICAL MODEL OF TRANSPORTATION PROBLEM Let us consider Example 9.1 to illustrate the mathematical model formulation of transportation problem of transporting single commodity from several supply origins, (also called capacity centres) to several demand destinations (also called requirement centres. Example 9.1 A company has thre factories 5, 5; and S; with production capacity of 7, 9 and 18 units (in 100s) per week of a product, respectively. These units are to be shipped to four warehouses Ds, Ds, Dy and Dy with requirement of 5, 6, 7 and 14 units (in 100's) per week, respectively. The transportation ‘costs (in rupees) per unit between factories to warehouses are given in the table below: Dr D Ds Di | Capacity Si. 19 30 50 10 q Sy 70 30 40 60 9 Ss 40 8 70 20 18 Demand 3 es | 14 a4 Formulate this transportation problem as an LP mode! to minimise the total transportation cost.OLE maaan pastes fe Techniques from factory i(1= 1, 286 Quanta th e te tye produet to be transpor Let, ty = number of units 0 Model formulation y cara ee Aeta) P model as follows: sblem is stated as an u — + 70x, + 302 + 40m ‘The transportation ae ‘inimis i = + 30x12 4 50%13 4 Minimise (total transportation cost) Z ea ee hs; “+ Brgy + 70833 + 20354 subject to the constraints (j) Capacity constraints =7 ay ton tan ta ay tn Xp 4207 ay tan tay a7 18 (ii) Requirement constraints 0 ay tan tant ay tan tin xy xn +39 7 mate t= sj 2 0 for {and J. and In the above LP model, there are man = 344 = 12 decision Variables, xy and m+n = 7 constants, where m are the numberof rows and are the number ‘of coluinns in a general transportation table. vn general, let there bem sources Sy Si -» Sw having a (1 Is respectively tobe transported among n destinations, Dip Da» Da 4 = 1,2, on) units of demand (or requirement) respectively. Let cy be the cost of shipping one unit of the commodity from source # or reration j for each route. fx, Tepresents number of unis shipped per rove from source to desti- ation , the problem is to determine the transportation schedule <0 as We minimise the total transportation raion ng supply and demand conditions. Mathematically, the problem in general may be stated as , m) waits of supply (or capacity) follows: Minimise (tt z= Ebay jnimise (tal cos) 2 = FE eyar a subject to the constraints Sony nau, 19 1 2 om Cuply consis) @ Boog aby J 102s (demand contains) @ and ay 2 0 for all i and Oy For easy presentation and solution, a transportation problem is generally presented 2s shown in Table 9.1 { Existence of feasible solution A necessary and sufficient condition for the existence of a fea ie ion fe il & : ee jit istence of feasible solutionTransportation Problem 267 Total supply = Total demand Ea = Es (also called rim conditions 1 ‘That is, the total capacity (or supply) must equal total requirement (or demand), ‘Table 9.1 General Transportation Table To Dy D a D, Supply From . ; 4 a ou @ en 8 3 Cn © i 5 ar ® or © = Gn 6 ai a ta 6 5 ton © a ane by by bn In this problem, there are (m + n) cofistraints and m * m variables. Since all (m + n) constraints are equations, there is one extra (redundant) equation on account of two way rim conditions (supply and demand conditions). The extra constraint can be derived from the other constraints without affecting the feasible solution. It follows that any feasible solution for a transportation problem must have exactly (m +m — 1) non-negative basic decision variables xy (or allocations) satisfying rim conditions. Remarks 1. When the total supply equals total demand, the problem is called a balanced transportation ‘problem, otherwise an unbalanced transportation problem. The unbalanced transportation problem can be ‘made balanced by adding a dumuny supply centre (row) or a dummy demand centre (column) as the need arses. 2. When the number of positive allocations (values of decision variables) at any stage of the feasible solution is less than the requited number (rows + columns - 1), ie, number of independent constraint ‘equations, the solution is said to be degenerate, otherwise non-degenerate 3. Cells in the transportation table having positive allocation will be called occupied cells, otherwise emply or non-occupied cells 9.3 THE TRANSPORTATION METHOD ‘The solution algorithm to a transporation problem may be summarised into the following, steps. nd set up in the form Step 1 Formulate the problen “The formulation of the transportation problem is similar to the L function is the (otal transportation cost and the constiaints ate the supply and demand available at each the objective problem formulation, 1 source and destination, respectively— 288 Quantitative Techniques Step 2 Obtain an initial basic feasible solution in a ion: In this chapter, we shall discuss following three different methods to obtain an initial solu @ North-West Corner Method, (ii) Least Cost Method, and (iii) Vogel's Approximation (or Penalty) Method. ‘The initial solution obtained by any of the three methods must satisfy the following conditions: (The solution must be feasible, i. it must satisfy al the supply and demand constraints (also called rim conditions), Gii) The number of positive allocations must be equal to m +n 1, when m is the number of rows and 1 is the number of columns. ‘Any solution that satisfies the above conditions is called non-degenerate basic feasible solution, otherwise, degenerate solution ‘Step 3. Test the initial solution for optimality is chapter, we shall discuss the Modified tion (MODI) method to test the optimality of the ion is optimal, then stop. Otherwise, determine a new improved In solution obtained in Step 2. Ifthe current sok solution. Step 4 Updating the solution Repeat Step 3 until an optimal solution is reached, 9.4 METHODS FOR FINDING INITIAL SOLUTION There are several methods available to obtain an initial basic feasible solution, Here we shall discuss only three different methods: 9.4.1 North-West Corner Method (NWCM) It is a simple and efficient method to obtain an initial solution. This method does not take into account the cost of transportation on any route of transportation. The method can be summarised as follows: ‘Step 1 Start with the cell at the upper left (north-west) comer of the transportation matrix and allocate as much as possible equal to the minimum of the rim values for the first row and first column, i.e. min (ai, 61) ‘Step 2 (0) If allocation made in Step | is equal tothe supply available at fist source (ay, in fist row), then move vertically down to the cell (2, 1) in the second row and first column and apply Step 1 for next allocation, (&) Iallocation made in Step 1 is equal to the demand of the fist destination (by in first column), then move horizontally to the cell (I, 2) in the first row and second column and apply Step 1 again for next allocation © Wa ‘Step 3 Continue the procedure step by step till an allocaton is made in the south-east comer cell of the transportation table. 1, allocate xj) = ay or by and move diagonally to the cell (2, 2).a) Transportation Problem 289 R , Fea pga the’ process of making allocation ata particular cell, supply equals demand, then next ‘ation of magnitude zero can be made ina cell either in the next rw or column This condition is known as degeneracy. Mlustration Taking data of Example 9.1, the initial solution using NWCM and total transportation cost is caloulated as follows and is shown in Table 9. ‘The cell (Sy Dy isthe north west comer cell in the given transportation able, The rim values for row S; and column D; are compared. The smaller ofthe two, ie, 5, is assigned as the first allocation; otherwise it will violate the feasibility condition. This means that 5 units ofa commodity ae to be transported fom source S; to destination Dj, However, ths allocation leaves a Supply of 7 ~ $= 2 units of commodity wo. Move horizontally and allocate as much a possible to cell (Si, D;) The rim value for ro Sy is2 and for column D, is 8. The smaller of the two, i.e. 2, is placed in the cell. Proceeding to row S2, since demand of D, has been met, nothing further ean be allocated to Dy, The unfulfilled demand of D; is now 8 2= 6 units, This can be fulfilled by S; with capacity of 9 units. So 6 unils #6 allocated to cell (6, Dp. The demand of Dy is now satisfied and a balance of 9 - 6 = 3 units remains with Sp ‘Table 9.2 Initial Solution using NWCM Dy Di Ds Ds Supply 19 30 50 10 s | @ |e 70 30 a 0 © 9 © Ss ET g 7 - |20 By moving in the same manner horizontally and vertically as succes demand and supply are met, ye encure thatthe solution is feasible, that is, the numberof positive allocations (occupied cells) is equal tom+n-1=3+4-1=6." “The total transportation cost of the i the quantity xy in the occupied cells with the correspon transportation cost of this solution is 5x 1942 «3046 3043x4044 x 704 M4 * 20= Rs 1.015 initial solution derived by the NWWCM is obtained by multiplying wing unit cost cy and adding. Thus, the total Total cost = 942 Least Cost Method (LCM) ( mata ot ow) ise the total transportation cost, we must try to transport as much as Hs) where the unit transportation cost is lowest: This method takes into f transportation for obtaining inital solution and can be summarised as Since the objective is to minim! possible through those routes (cel account the minimum unit cost o! follows:290 Quantitative Techniques le and allocate as much as ly or demand is exhausted sed out. ‘maximum allocation can Step 1 Select the cell with the lowest unit cost in the entre transportation eee possible to this cell and eliminate (Line out thatrow or column which either supPl If both a row and a column are satisfied simultaneously, ‘only one may be cross se cass the arsalles unit cost cel isnot sigue, thea select the cell coe be made, ly Step 2. After adjusting the supply and demand forall unerossed-out ro" and coun repeat the Beco ‘withthe next lowest unit cost among the remaining rows and columns of the transportation table an ; as much as possible to this cell and eliminate (lin ‘d column in which either supply OF demand is exhausted. Step 3 Repeat the procedure until the entire a destinations is satisfied. The solution so obtain sailable supply at various sources and demand at various ed need not be non-degenerate. tion using LCM and total transportation cost is Tlustration Taking data of Example 9.1, the initial sol caleulated as follows: “The cell with lowest unit cost (ie. 8) is (Ss, D3). The maximum which we can allocate 10 this cell is 8. This meets the complete demand of D3 and leave 10 units with Sy. as sbewn in Table 9.3. In the reduced table without column D2, the next smallest unit transportation cost is 10 in cell (5, Do, The maximum which can be allocated to this cell is 7. This exhausts the capacity of Si and leaves 7 units with Dy as unsatisfied demand as shown in Table 9.3. Table 93 Demand In Table 9.3, the next sinallest cost is 20 in cell (Ss, D,). The maximum allocation in this cell could be 7 units, This satisfies the entire demand of Dy and leaves 3 units with Sy as remaining supply as shown Table 9.4. In Table 9.4, the next smallest unit cost cell is not unique. That is, there are two cells (S2, Ds) and (Ss, D)) having same unit transportation cost 40. Allocate 7 units in cell (S;, D3) first because it can accommodate more units as compared to"cell (Ss, Dj), and allocate 3 units (only supply left with 55) in cell (S;, Di). The remaining demand of 2 units of Dy is fulfilled from $2, Since supply and — a each origin and destination is exhausted, the initial solution is arrived at, and is shown in ableTransportation Problem 294 D Dy Dy Dy ] Supply 19] : Ss, 30 50) - ak 40 ae ® Sy ‘ 0 Demand , 8 ‘The total transportation cost of the ial solution by LCM is calculated as given below: 7 20= Rs. 814 Total cost = 7x 10+2* 7047x4043 x40 48x 8+ obtained by NWCM. The optimal “The total transportation cost obtained by LCM is less than the cost solution can also be obtained faster by using LCM. 9.4.3 Vogel’s Approximation Method (VAM) ristie method and is preferred to the other two tod, each allocation is made on the basis of the opportunity (or penalty sired if allocations in certain cells wth minimum unit transportation cr were missed. In this method allocations are made so thatthe penaly cot is minimised, The advantage of this method is that it gives an ‘tial solution which is nearer to an ‘optimal solution or is the optimal solution itself. The steps in VAM are as follows: ing the difference between the smallest and next enalties for each row (column) by tak » This difference indicates the penalty or extra tation cost in the same row (column). the cell with the minimum unit transportation cost. Vogel's approximation (penalty or regret) method is heu methods described above. In this me or extra) cost that would have beer ‘Step 1 Calculate p smallest unit transpo Cost which has to be paid if one fails to allocate to 1 the row or column with the largest penalty and allocat he selected row or column satisfying the rim conditions. If there Is ir can be broken by selecting the cell where maximum allocation can be made. Step 2 Select te as much as possible in the cell having, the least cos a tie in the values of penalties, ‘step 3. Adjust the supply and demand and ero Sve ised simultaneously, only one of them js erossed out and | eo supply (demand), Any row or column with 2er0 supply or demand should not be us future penalties. ‘Step 4 Repeat Steps | to 3 until the e destinations are satisfied. Illustration Taking data of Example 9.1, the Table 9.5. ‘As shown in the Table 9.5, differences (penalty costs) for each row and column have been calculated. In i Fat round, the maximum penalty, 22 occurs in column Da, Ths (he cell (Ss, D3) having the least transportation cost 8s chosen for allocation. The maximian possible allocation inthis ells 8 and it satisfies seoeref in column Dz. Adjust the supply of 55 from 18 to 10 (18 ~ 8 = 10) és out the satisfied row or column, Ifa row and a column the remaining row (column) is assigned sed in computing tire available supply at various sources and demand at various use of VAM to obtain initial solution is illustrated in2 Quantitative Techniqu ‘Yable9.5 Initial Solution Using VAM D, Supply | Row differences rT S [9] 7 9 9 40 S10 9 10 20 20 40 5s. {40 8, | 12 20 50 - Demand 5 y Column q liferences| 21 ‘The new row and column penalties are calculated except column Dz because its demand has been satisfied. The second round allocation is made in column D; with target penalty 21 in the same way as in the first round as shown in cell (S), Di) of Table 9.5. In the third round, the maximum penalty 50 occurs at S, The maximum possible allocation of 10 units is made in cell (Ss, D,) having least transportation cost 20 as shown in Table 9.5. ‘The process is continued with new allocations tll a complete solution is obtained. The initial solution using VAM is shown in Table 9.5. The total transportation cost associated with this method is calculated as: Total cost = 5 x 19+ 2x 10+ 7x 40 +2 x 604.8 x 8 + 10 x 20 = Rs 779 Example 9.2 A dairy firm has three plants locaied in a state, The daily milk production at each plant is as follows: Plant 1: 6 million litres, Plant 2: 1 million lites, and Plant 3: 10 million litres Each day, the firm must fulfil the needs of its four distribution centres. Minimum requirement at cach centre is as follows: Distribution centre 1: 7 million litres, Distribution centre 2: 5 million litres, Distribution centre 3: 3 million litres, and Distribution centre 4: 2 million litres Cost in hundreds of rupees of shipping one million litre from each plant to each distribution centre is given in the following table: : Distribution Centre Dy D; Dy Dy Py 2 3 i 7 Plant Py 1 0 q | PB 5 8 15 9‘Transportation Problem 293 Find initial basic feasible solution for given pr (a) North-west comer rule aan ie (b) Least cost method (©) Vogel's approximation method ; if the object is to minimise the total transportation cost. Solution (a) North-West Corner Rule Table 9.6 Distribution Centre Dy Ds Dd Db Supply 2 3 I 7 al” © 6-4 ey 0 6 |! Ma Plot P|! @ l= P; 5 8 1s 9 10 =a ' 3 7 i @ @ | @ ’ Demand T= bh Sah 3=by 2h @) Comparing ay and by, since a, < by; allocate xy = 6. This exhausts the supply at Pr and leaves T unit as unsatisfied demand at Dy. ii) Move to cell (Ps, Dy). Compare a; and by (ie. 1 and 1). Since ap = by, allocate sat = i) Move to cell (Ps, D3). Since supply at Ps, is equal tothe demand at Dy, Ds and Da, therefore, allocate x32 = 5, 333 = 3 and x54 = 2 that the number of allocated cells (also called basic cells) are 5 which is one less than It may be noted 6), Thus, this solution is the degenerate solution. The transportation the required number m+n-1(3+4— cost associated with this solution is: Total cost = Rs (2*6+1x1+8xS+ 15 3 +9 2) x 100 = Rs 11,600 (b) Least Cost Method Table 9.7 Distribution Centre D Di Ds Ds Supply 2 3 ul 7 nl @ 6 ‘ pint 72]! § ' 1 lant Pa oO nls 8 15 9 "9 ‘| oO @ @_L© 4 Demand 7 5 3 2 |284 Quantitative Techniques (]) The lowest unit cost in Table 9.7 is 0 in cell (P, Da), therefore maximum possible allocation whicy can be made here is 1. This exhausts the supply at plant P2, therefore row 2 is crossed out. Gi) The next lowest unit cost is 2 in cell (Pt, Di). The maximum possible allocation which can be made here is 6. This exhausts the supply at plant Py, therefore, row P, is crossed out, ) ii) Since the total supply at plant Ps is now equal tothe unsatisfied demand at all the four distribution centres, therefore, maximum possible allocations s ells (P3, Di), (Ps, D3), (Ps, Ds) and (P3, Di). The number of allocated cells inthis case are six whichis equal tothe required number m +n ~ 1 -@ +4~ 1 = 6). Thus, this solution is non-degenerate, : The transportation cost associated with this solution is Total cost = Rs (2x 645x148 x44 15 x3 49% 2) x 100 Rs 11,200 (©) Vogels Approximation Method ‘The method is self-explanatory as shown in Table 9.8 ing the supply and demand conditions are made in Table 9.8, Distribution Centre ‘ D> Dr Ds D Supply Row penality rag J u 7 Gedy ir 5 ‘| ® ® ! 1 0 «| 6 1 i 0 2° « Plant P2 O PY ic 8 15 9 tes) a) | © ®1o Demand 7 5 3 2 Column 1 3 5 6 penalty 3 5 4 2 3 - 4 2 The number of allocated cells in Table 9.8 are six, which is equal to the required number m+n—1 (3 +4~1 = 6), therefore this solution is non-degenerate. The transportation cost associated with this solution is Total cost = Rs(2* 1+ 3*S+1 x 145% 6415 *3+9% 1) * 100=Rs 10,200 It can be seen thatthe total transportation cost found by VAM is lower than the costs of transportation determined by the other two methods. Therefore, itis of advantage to use this method in order to reduce computational time required to obtain optimum solution REVIEW QUESTIONS 9.1 1. Show that all the bases for a transportation problem are triangular 2. With reference to a transportation problem define the following terms: (i Feasible solution (ii) Basic feasible solution (iv) Non-degenerate basic feasible solutionwe vive Sut) ie 3) Gey (8) | 4owe the following _tyamspp stubiom Pxoblean. bu NWiUCM (Nosh West Cosmet Method.) L oe Plants jos Chouses —— Se FG ‘lable A me eT a 2 a é tes ( G uo—+ -F— f UBC] Eh a Co A Mee Le] edge: 5 fer ( So-5 te 200 14 ¢S0) + 30 (60) + Un (30) + to (20) + 20 uo’ Gso + 1%00 +)900 + I4on + “Oo 61$0 » Be p/ N cuo) + ¥ (60) + Ho (So) +)ocuo +uko + 200 +l(Baer 15 44a polation CB) | fowe tne —_foliauainng A =n vam. (‘Vogel's Sane coating methodl-) Dlamts Gimehouses ——[frmownt JP, Py Pa 0 a E e ai leblel A eee eee 30 © f ie Se ee c C60) ob ' - ae Amount} ber 6, | 5e,, | ton | Zoo Reqused| . Ga or lo ite} qo ® = lo __5O Ga = = — aso 5430 19 (50) _* % (609 ~_LkO + _Jooo + Lo ($0) _ + 60 (uo? + 9400oO Pxob\eon 4 7 lean adh. ON 92 | 93 Dy Supply OL aba es Oo ine ug palates ge 03 | ag ge apa 8 | ate Ze. Demand xe-_ [Gen | ne 4 Bae ae eS +35 (Go) + lo ¢ 90) 34 Coo) + 0 (60d +59 (qo) x 5) (600+ 34 (KO) 3400 +o + U6KO +3060 +2190 + 3U90 + ¥00 yD 1SFEKO(ex 91) co 1 ally wal ta ‘o Pxoh\enn by Nwem __ Method. Dy Do [Oa 5} De —Lsupely 1 Up a Oty fay Loe | ( { | il al a i 10" are se 4230 I 03 | a¢— =f Hto fg Wyo Demin) 2-0 Ge 4e ern |e 8?! be 00 __ ” U9 CoD + Ux (40) + HG Wor + $2 (309 +43 C130) + 0660 9960 + 3nuo + UGo + S120 + JS30 + L5go +o 20530(13) | Sowe the kolo ing jnomspostakom > S%0h\eam by vam