Linear Programming ea 2.1 INTRODUCTION Linear programming is withthe optimization (masincohccae optimal allocation of limited resources. It deals ned BT e minimization) of a function of variables known as objective fe aaa eat equations and/or inequalities known as constraints. The objective funeton may be profi, cost, production capacity or any other measure of eetiveness, which to st possible or optimal manner. The constraints may be imposed by different resources such as raw material availability, market demand, production process and storage capacity, etc. By linearity is meant a math i ae BEG ae eases ines peer mathematical expression in which the expressions I linear e.g., the expression ajx + azt, + asx; + .. + ay is linear. Higher powers of the variables or their products donot appear in the expressions for the objective function as well as the constraints (they do not have expressions like x)°, 1°”, xjx2, a1x) + a log x2, etc.). The variables obey the properties of proportionality (eg. if a product requires 3 hours of machining time, 5 units of it will require 15 hours) and additivity (e.g. amount of a resource required for a certain number of products is equal to the sum of the resource required for each). It was in 1947 that George Dantzig and his associates found out a technique for solving nilitary planning problems while they were working on a project for U.S. Air Force. This technique consisted of representing the various activites of an organization as a linear programming (L-P.) model and arriving at the optimal programme by minimizing a linear objective function. Afterwards, Dantzig suggested this approach for solving business and industrial problems. He also developed the most powerful mathematical tool known as “simplex method to solve linear programming problems. 2.2 “REQUIREMENTS FOR A LINEAR PROGRAMMING PROBLEM {All organizations, big or small, have at their disposal, men, machines, money and materials, the supply of which may be limited. If the supply of these resources were unlimited, the need for management tools like linear programming would not arse at all: Supply of resources being limited, the management must find the best allocation of its resources in order to maximize the profit or minimize the cost/loss or utilize the production capacity to the maxim extent, However, Fete eenanan umber of problems which can be overcome by quantitative methods, particularly the linear programming. Generally speaking, following conditions are satisfied: 1. There must be a well defined objective function (profit, cost or quantities prosce which is to be either maximized or minimized and which can be expressed as a linear function of decision variables. linear programming can be used for optimization problems if the 4oi" © } 42% OPERATIONS RESEARCH oon the amount or extent of attainment of the objective and >. There must be constraints these constraints must be capable oF being expressed as linear equations or inequalities in terms of variables. 3. There must by two different machines and pr. machine. Mather necessary requirement is that d negative. The non-negativity condition shows that for which negative quantities are generally illogical the resources must be in 5. As stated earlier, t producing greater number of a particular product, it mus Pe the total production capacity is limited. 213° ASSUMPTIONS IN LINEAR PI A linear programming model is based on the following assumptions: 1, Proportionality: A basic assumption of linear programming is that proportionality-exists in the objective function and the constraints. This assumption implies that if a product yields a profit of @ 10, the profit eamed from the sale of 12 such products will be € (10 * 12) ~¥ 120. This counts, Further, even if the sale price is constant, may not always be true because of quantity dis the manufacturing cost may vary with the number of units produced and so may vary the profit per unit. Likewise, it is assumed that if one product requires processing time of 5 hours then per such products will require processing time of S * 10 = '50 hours. This may also not be true tthe processing time per unit often decreases with inerease in number of units produced. The real world situations may not be strictly linear. However, ‘assumed linearity represents their close approximations and provides very useful answers ey "additivity: It means that if we use f, hours on machine 4 °o make product | and fp hours to make product 2, the total time required to make products 1 and 2 on machine A is 1) + f» hours. 2 is negligible. Some This, however, is true only if the change-over time from product 1 to product processes may not behave in this way. For ‘example, when several liquids of different chemical Compositions are mixed, the resulting volume may not be equal to the sum of the volumes of the individual liquids. + Continuity: Another assumption underlying the linear programming model is that the decision variables are continuous i, they are permitted to take any non-negali® values that satisfy the constraints, However, there are problems wherein variables are restricted to have integral values only. Though such problems, strictly speaking, are not linear programming problems, they are frequently solved by linear programming techniques and the values are then rounded off to nearest integers to satisfy the constraints. This approximation, however, is valid only if the variables have large optimal values. Further, it must ‘be ascertained whether the solution represented by the rounded values is 2 feasible solution and also whether the solution is the best integer solution. “| Certainty: Another assumption underlying @ linear programming model is that the various parameters, namely, the objective function coefficients, R.H.S. coefficients of the constraints and revource values in the constraints are certainly and precisely known and that their values do not change with time. Thus the profit or cost per unit of the product, labour and materials requ per unit, availability of labour and materials, market demand of the product produced, ete. a es re isan vie certainty, The linear programming problem is, therefore, assumed '0 5, Finite Choices: A linear programming model also assumes that a finite (limited) number of choices (alternatives) are available to the decision-maker and that the aoe estes we be alternative courses of action. For example, a given product may be processed lem may be as to how much ofthe product to allocate 10 which lecision variables should be interrelated and non- Tinear programming deals with real life situations limited supply. For example, if a firm starts ‘st make smaller number of other products ROGRAMMING MODELSLINEAR PROGRAMMING “ 43 interrelated and non-negative. The non-negativity condition shows that linear programming deals with real-life situations as it is not possible to produce/use negative quantities Mathematically these non-negativity conditions do not differ from other constraints. However, since while solving the problems they are handled differently from the other constraints, they are termed as non-negativity restrictions and the term constraints is used to represent constraints other than non-negativity restrictions and this terminology has been followed throughout the book. 2.4 APPLICATIONS OF LINEAR PROGRAMMING METHOD Though, in the world we live, most of the events are non-linear, yet there are many instances of linear events that occur in day-to-day life. Therefore, an understanding of linear programming and its application in solving problems is utmost essential for today’s managers. Linear programming techniques are widely used to solve a number of business, industrial, military, economic, marketing, distribution and advertising problems. Three primary reasons for its wide use are: 1. A large number of problems from different fields can be represented or at least approximated to linear programming problems. 2. Powerful and efficient techniques for solving L.P. problems are available. 3. L.P. models can handle data variation (sensitivity analysis) easily. However, solution procedures are generally iterative and even medium size problems require manipulation of large amount of data, But with the development of digital computers, this disadvantage has been completely overcome as these computers can handle even large L.P. problems in comparatively very little time at a low cost. 2.5 AREAS OF APPLICATION OF LINEAR PROGRAMMING Linear programming is one of the most widely applied techniques of operations research in business, industry and numerous other fields. A few areas of its application are given below. 1, INDUSTRIAL APPLICATIONS (a) Product mix problems: An industrial organisation has available a certain production capacity (men, machines, money, materials, market, etc.) on various manufacturing processes to manufacture various products. Typically, differents products will have different selling prices, will require different amounts of production capacity at the several processes and will therefore, have different unit profits; there may also be stipulations (conditions) on maximum and/or minimum product levels. The problem is to determine the product mix that will maximize the total profit. (b) Blending problems: These problems are likely to arise when a product can be made from a variety of available raw materials of various compositions and prices. The manufacturing process involves blending (mixing) some of these materials in varying quantities to make a product of the desired specifications. For instance, different grades of gasoline are required for aviation purposes. Prices and specifications such as octane ratings, tetra ethyl lead concentrations, maximum vapour pressure, etc of input ingredients are given and the problem is to decide the proportions of these ingredients to make the desired grades of gasoline so that (i) maximum output is obtained and (ii) storage capacity restrictions are satisfied. Many similar situations such as preparation of different kinds of whisky, chemicals, fertilisers and alloys, etc. have been handled by this technique of linear programming. (c) Production scheduling problems: They involve the determination of optimum production schedule to meet fluctuating demand. The objective is to meet demand, keeping inventory and employment at reasonable minimum levels, while minimizing the total cost of production and inventory.OPERATIONS RESEARCH (d) Trim loss problem: industries where b They are applicable to paper, sheet m items of standard sizes have to be with the objective of minimizing the waste produ (e) Assembly-line balancing: al and glass manufiactyy Cut to smaller sizes as per customer requite ced, It relates to a category of - of problems wherein the final rod fas a number of different components assembled together. These componene ae beminnbey in a specific sequence or set of sequences. Each assembly combination of tasks so that his task time (f) Make-or-buy (sub-contractin, ny Ment Assembled Perator is to be assigned the tash ts less than or equal to the cycle time. 8) problems: They arise in an organisation in the f Production capacity limitations and sudden Spurt in demand of its products. The manufue being sure of the demand pattem, is usually reluctant to add a decision regarding the products to be manufacured with hi be subcontracted so that the total cost is minimized 2. MANAGEMENT APPLICATIONS (a) Media selection problems: They involve the different advertising media such as T-V., radio, magazines and newspapers that will maxims Public exposure to company's product. The constraints may be on the total advertising budget, ‘maximum expenditure in each media, maximum number of insertions in each media and the (6) Portfolio selection problems: They are frequently encountered by banks, financial companies, insurance companies, investment services, etc. A given amount is to be allocated among several investment altematives such as bonds, saving certificates, common stock, mu fund, real estate, etc. to maximize the expected return or minimize the expected risk (c) Profit planning problems: They involve planning profits on fiscal year basis to maximize Profit margin from investment in plant facilities, machinery, inventory and cash on hand (d) Transportation problems: They involve transportation of products from, say, 1 sources situated at different locations to, say, m different destinations. Supply position at the sources, demand at destinations, freight charges and storage costs, etc. are known and the problem is to design the optimum transportation plan that minimizes the total transportation cost (or distance or time). er, not additional capacity and has to make s OWN resources and the products to selection of advertising mix among utual (e) Assignment problems: They are concerned with allocation of facilities (men or machines) to jobs. Time required by each facility to perform each job is given and the problem is to find the optimum allocation (one job to one facility) so that the total time to perform the jobs is minimized. () Manpower scheduling problems: They are faced by big hospitals, restaurants and companies operating in a number of shifts. The problem is to allocate optimum manpower in each shift so that the overtime cost is minimized. 3. MISCELLANEOUS APPLICATIONS (a) Diet problems: They form another important category to which linear programming has been applied. Nutrient contents such as vitamins, proteins, fats, carbohydrates, starch, etc. in each of a number of food stuffs is known. Also the minimum daily requirement of each nutrient in the diet as well as the cost of each type of food stuff is given and the problem is to determine the minimum cost diet that satisfies the minimum daily requirement of nutrients. (6) Agriculture problems: These problems are concerned with the allocation of input resources such as acreage of land, water, labour, fertilisers and capital to various crops so as 10 maximize net revenue. (c) Flight scheduling problems: They are devoted to the determination of the most economical patterns and timings of flights that result in the most efficient use of aircrafts and crews- (d) Environment protection: They involve analysis of different alternatives for efficient waste disposal, paper recycling and energy policies.LINEAR PROGRAMMING * 45 (c) Facilities location: These problems are concerned with the determination of best location of public parks, libraries and recreation areas, hospital ambulance depots, telephone exchanges, nuclear power plants, ete. Oil refineries have used linear programming with considerable success. Similar trends are developing in chemical industries, iron and stee! industries, aluminium industry, food processing industry, wood products manufacture and many others. Other areas where linear programming has been applied include quality control inspection, determination of optimal bombing patterns, searching of submarines, design of war weapons, vendor quotation analysis, structural design, scheduling military tanker fleet, fabrication scheduling, steel production scheduling, balancing of assembly lines and computations of maximum flows in networks. In fact linear programming may be used for any general situation where a linear objective function has to be optimised subject to constraints expressed as linear equations/inequalities. 2.6 FORMULATION OF LINEAR PROGRAMMING PROBLEMS First, the given problem must be presented in linear programming form. This requires defining the variables of the problem, establishing interrelationships between them and formulating the objective function and constraints. A model, which approximates as closely as possible to the given problem, is then to be developed. If some constraints happen to be non-linear, they are approximated to appropriate linear functions to fit the linear programming format. In case it is not possible, other techniques may be used to formulate and then solve the model. EXAMPLE 2.6-1 (Production Allocation Problem) A firm produces three products. These products are processed on three different machines. The time required to manufacture one unit of each of the three products and the daily capacity of the three machines are given in the table below. TABLE 2.1 ‘Machine Time per unit (minutes) Machine Product! | Product2 | Product 3 capacity (minutes/day) M, 2 3 2 440 My 4 470 My 2 is = 430 Tris required to determine the daily number of units to be manufactured for each product. The profit per unit for product 1, 2 and 3 is & 4, © 3 and 6 respectively. It is assumed that all the amounts produced are consumed in the market. Formulate the mathematical (L.P) model that will maximize the daily profi. [G.N.D.U. BBA Sept., 1998; H.P.U. MCA, 1999] Formulation of Linear Programming Model Step I: From the study of the situation find the key decision to be made. It this connection, looking for variables helps considerably. In the given situation key decision is to decide the number of units of products 1, 2 and 3 to be produced daily. Step 2: ‘Assume symbols for variable quantities noticed in step 1. Let the number of units of products, 1,2 and 3 manufactured daily be x), xp and x3. Step 3: Express the feasible alternatives mathematically in terms of variables. Feasible alternatives are those which are physically, economically and financially possible. In the given situation feasible alternatives are sets of values of x), x3 and x3,46 OPERATIONS RESEARCH _ 1 me yee ince negative production has no meaning and is not feasible Step 4: Mention the objective quantitatively and express it as a linear function of varia present situation, objective is to maximize the profit ng, ie. maximize Z= 4x, + 34 + 68 Step 5: Put into words the influencing factors or constraints. These occur generally bee constraints on availability (resources) or requirements (demands). Express these cone as linear equations/inequalities in terms of variab) IS ay Here constraints are on the machine capacities and can be mathematically expresseg Duy + 3xy + xy $ 440, s Ax, + Oxy + 3x; $470, ny + Sty + Oxy $430. Therefore, the complete mathematical (L.P.) model for the problem can be written as maximize = + 6x. subject to constraints, 2x, + where xy.%X EXAMPLE 2.6-2 (Diet Problem) A person wants to decide the constituents ofa diet which will fulfill his daily requiremens proteins, fats and carbohydrates atthe minimum cost. The choice is 10 be made from four dif types of foods. The yields per unit ofthese foods are given in table 22. 320. TABLE 2.2 5 Wield per unit Cost per unit Food oe | Proteins Fats__ | Carbohydrates ® 7 3 2 6 45 > 4 4 0 3 3 7 35 4 6 3 4 65 Minimum requirement |__800 200 700 Formulate linear programming model for the problem. Formulation of L.P. Model Let xj.X3.%) and x, denote the number of units of food of type 1, 2.3 and 4 respect Objective is to minimize the cost ie. Minimize Z = & (45x, + 40x; + 85xy + 65x.)- Constraints are on the fulfilment of the daily requirements of the various constitues® ie. _ for proteins, Bay + dry + Sry + 6x4 2 800, for fats, 2x, + Quy + xy + Sty 2 200, and for carbohydrates, 6) + 4xz + 7xy + 4xy 2 700, where 2), 5, %5, "4, each 2 0. og__————— — EXAMPLE 2.6-3 (Blending Problem) A firm produces an alloy having the following specifications. (specific gravity < 0.98, (ii) chromium > 8%, (iii), melting point > 450°C. LINEAR PROGRAMMING “ 47 ——————n © Raw materials A, B and C having the properties shown in the table can be used to make the alloy. TABLE 2,3 aay elt Properties of raw material A B Cc Specific gravity 0.92 097 1.04 Chromium ™% 13% 16% Melting point 440°C. 490°C 480°C Costs of the various raw materials er ton are: % 90 for A, = 280 for B and ® 40 for C. Formulate the L.P. model to find the proportions in which A, B and C be used to obtain an alloy of desired properties while the cost of raw materials is minimum. (PU. BE. (E. & Comm.) De., 2011; G.N.D.U. BBA April, 2006) Formulation of Linear Programming Model Let the proportions of raw materials A, B and C to be used for making the alloy be x), x7 and x; respectively. Objective is to minimize the cost ie, minimize Z = 90x, + 280x) + 40x; Constraints are imposed by the specifications required for the alloy. They are 0.92x, + 0.97x + 1.04x; < 0.98, Tx, + 13x + 16x; 2 8, 440x; + 490x, + 480x3 2 450, and ntmtay=l, as x, x2 and x3 are the proportions of materials A, B and C in making the alloy. Also is X2y.%3, each 2 0, EXAMPLE 2.6-4 (Advertising Media Selection Problem) An advertising company wishes to plan its advertising strategy in three different media— television, radio and magazines. The purpose of advertising is to reach as large a number of Potential customers as possible. Following data have been obtained from market survey: TABLE 24 | Rlevision | Radio | Magazine I | Magazine IT Cost of an advertising unit & 30,000 % 20,000 215,000 @ 10,000 No. of potential customers reached per unit 2,00,000 6,00,000 1,50,000 1,00,000 No. of female customers reached per unit 1,50,000 4,00,000 70,000 50,000 The company wants to spend not more than © 4,50,000 on advertising. Following are the + Further requirements that must be met:() at least 1 million exposures take place among, (ii) advertising on magazines be (ii) at least 3 advertising (iv) the number of advertising units on television an Sand 10. Formulate an L.P. model for the problem [H.PU.B. Tech. (Mec th) June, 2010; Formulation of Linear Programming Let xj,.x3, 4) and x, denote the number 0 magazine 1 and magazine 11 respectively. the total number of poten! The objective is to maximize hen maximize Z Constraints are on the advertising budget: ‘on number of female customers: reached by the advertising campaign: oon expenses on magazine advertising: on no. of units on magazines: ‘on no. of units on television: on no. of units on radio: where x15 2, %3 x4, each 2 0. EXAMPLE 2.6-5 (Inspection Problem) inspectors, Land Il to ‘A company has two grades of east 1,500 pieces must be inspected in an %, Grade II inspector c ‘an hour with an accuracy of 96 of 92%. Any error made by an inspe dnd 15 grade IT inspectors in the company, find the Com. the daily inspection cost. (PU: B. Formulation of L.P. Model Let x, and x; denote the job of quality control inspection. The objective is to minim' two types of costs: wages paid of a grade | inspector/hour is 10° ( Wages of grade I inspector are ® 5 per hour sector costs & 3 to the company. ize the daily cost of inspection. Now the company has to the inspectors and the cost of their inspection errors. The cost Jemale customers, limited to & 1,50,000, ine and 2 units On magazine 1 units be bought on magaz Mf radio should each be between P-T.U. MBA May, 2007) Model ‘advertising units to be bought of television, radi tial customers: reached. Dry + Or + 1Sty + Xa) + 15,000x, + 10,000x, 4,50,000 < 450, 30,000x, + 20,000x2 on 30x, + 20x + 15x, + 10%. 470,000, + 50,000x, 2 10,00,000 2 100, 1,50,000x; + 4,00,000%2 on 15x; + 40x, + Tes + 54 + 10,000x, ¢ 1,50,000 15,000x; or 15x; + 104 S150, 523, 422, $x, 51007, 25,41 S10, < 10, 5 xy $ 10 or x2 2 55% undertake quality control inspection. At "s-hour day. Grade I inspector can check 20 pieces in necks 14 pieces an hour with an accuracy while those of grade II inspector are® 4 per how. If there are, in all, 10 grade I inspectors optimal assignment of inspectors that minimizes April, 2008; D.U. MBA, 2003, 02, 00; NIFT Mohali, 2000) number of grade I and grade II inspectors that may be assigned the to incut %(5 +3 * 0.04 * 20) = % 7.40. Similarly, cost of a grade II inspector/hour is 2 (443 * 0.08 * 14) = & 7.36, ~. The objective function is minimize Z = 8(7.40x, + 7.36x;) = $9.20x, + $8.88xpLINEAR PROGRAMMING * 49 Constraints are on the number of grade I inspectors : x, < 10, on the number of grade II inspectors : x» < 15, on the number of pieces to be inspected daily: 20 x 8x, + 14 x Rr > 1,500 or 160x; + 112x > 1,500, where x1, x7 2 0. EXAMPLE 2.6-6 (Product Mix Problem) A chemical company produces two products, X and ¥. Each unit of product X requires 3 hours on operation I and 4 hours on operation Il, while each unit of product ¥ requires 4 hours on operation I and 5 hours on operation II. Total available time for operations I and II is 20 hours and 26 hours respectively. The production of each unit of product ¥ also results in two units of a by-product Z at no extra cost. Product X sells at profit of € 10/unit, while ¥ sells at profit of & 20/unit. By-product Z brings a unit profit of & 6 if sold; in case it cannot be sold, the destruction cost is € 4hmit. Forecasts indicate that not more than $ units of Z can be sold. Formulate the L.P. model to determine the quantities of X and Y to be produced, keeping Z in mind, so that the profit earned is maximum. [B.PU.T, 2012; PU.B. Com. April, 2006; Jammu U.B.E. (Mech.) 2004; PT.U.B.Tech., 2000] Formulation of L.P. Model Let the number of units of products X, Y and Z produced be x), x2, x=, where umber of units of Z produced jumber of units of Z sold + number of units of Z destroyed 3 + Xq (Say). Objective is to maximize the profit. Objective function (profit function) for products X and Y is linear because their profits (© 10/unit and & 20/unit) are constants irrespective of the number of units produced. A graph between the total profit and quantity produced will be a straight line. However, a similar graph for product Z is non-linear since it has slope +6 for first part, while a slope of —4 for the second. However, it is piecewise linear, since it is linear in the regions (0 to 5) and (5 to 2Y). Thus splitting x, into two parts, viz. the number of units of Z sold (x;) and number of units of Z destroyed (x) makes the objective function for product Z also linear. Thus the objective function is maximize Z = 10x, + 20x + 6x3 ~ 4x4. Constraints are on the time available on operation I: 3x; + 4x, $ 20, on the time available on operation I: 4x, + 5x3 < 26, on the number of units of product Z sold: x; < 5, on the number of units of product Z produced: 2Y = Z or 2x) = x3 +44 OF — 2x +43 +14 =0, where x), X2. X55 Xs, each 2 0. EXAMPLE 2.6-7 (Product Mix Problem) A firm manufactures three products A, B and C. Time to manufacture product A is twice that Sor B and thrice that for C and if the emire labour is engaged in making product A, 1,600 units of this product can be produced. These products are to be produced in the ratio 3: 4 : $. There is demand for at least 300, 250 and 200 units of products A, B and C and the profit earned per unit is % 90, & 40 and & 30 respectively Formulate the problem as a linear programming problem.