14 1.2 13 14 Introduction The nature and definition of Operations Research Objectives of Operations Research Characteristics of Operations Research 4.8 Modelling in Operations Research 1.6 Types of models 4.7 General methods for solving O.R. models 18 19 Phases of Operations Research study Scope of Operations Research 1.40 Development of Operations Research in India Bienes Introduction Matrix Type of matrices Operation on matrices Properties of matrix addition Properties of multiplication of matrix by a scalar Multiplication of matrices Determinant of a square matrix Properties of determinants Evaluation of a determinant by Sarrus diagram Minors and cofactors Singular and non-singular matrix Transpose of a matrix Properties of transpose of a matrix Symmetric matrix Skew-symmetric matrix Rank of a matrixOPERATIONS RESEARcy LINEAR PROGRAMMING PROBLEMS. FUMIO KCL aa er SOMOS) coer 3.1 Introduction 49 3.2 Basie terminology of linear programming 49 3.3 Basic requirements of LPP 50 Zz 3.4 Basic assumptions of LP model 50 3.5 Advantages of linear programming 50 3.6 Limitations of linear programming 51 ( 3.7 Application areas of linear programming 51 : 3.8 Standard form of linear programming problems 52 3.9 Matrix form of LPP 53 3-1= Mathematical formulation of linear programming problem 53 $.11 Extra variable needed 54 ee 3.12 Solution of linear programming problems 63 9.13. Graphical solution methods of an LPP 64 109 EE i Introduction a 42 Terminology and notations 43 Fundamental theorem of linear programming 110 4.4 Reduction of feasible solution to basic feasible solution 112 45 ‘To find the improved basic feasible solution (BFS) from a given BFS 124 46 Conditions for the existence of unbounded solutions 26 47 Conditions for improved basic feasible solution to become optimal 127 48 Alternative optimal solutions 128 + 49 Inconsistency and redundancy 128 E. 410 Procedure to obtain initial basic feasible solution 28 11 Simplex algorithm 130 Simplex method : Case of minimization > Artificial variable technique | Two-Phase method e special linear programming problems in of simultaneous linear equations by Simplex method ES |of a matrix by Simplex method Naas U ons Ree) linear programming and sufficient condition for the existence of non-degeneracy jeracy in linear programming ee)6.3. Revised Simplex method for standard Form-1 6.4 Revised Simplex method for standard Form-I1 6.5 Comparison of Simplex method and Revised Simplex method Tas Ty 71 introduction 2, Relationship between Primal and dual Symmetric Primal-Dual problems Dual of an LPP with mixed restrictions 75. Some results on duality 7.6 Method for obtaining the solution to the dual from the final simplex table of the primal and vice-versa ZL Summary of the correspondence between primal and its dual COUT 8.1 Introduction 8.2. Dual Simplex Algorithm 8.3. Primal-Dual Algorithm 8.4 Steps of Primal-Dual Algorith 9.1 Introduction 9.2 Change in the objective function coefficie 9.3. Variation in the requirement Vector, b 9.4 Variation in the elements, aj of the coef 9.5. Addition of a new variable 9.6 Addition of a new constraint 40.1 Introduction 40.2. Need of Integer linear programming 40.3. Types of Integer linear programming problems 40.4 Methods to solve an Integer linear programming problem 40.5 Problem on mixed integer linear programming 10.6 The Branch and Bound technique EO 11.1. Introduction al Programming model formulation jneral form of Goal programming problem Solution of a Goal programming (GP) problemOPERATIONS RESEARCH CON’ iKTEUCairu hee 423-484 Introduction 423 1.2, Mathematical formulation of Transportation problem 423 1 13.3 Solution of the Transportation problem 427 1 13.4 Methods of finding initial solution 427 1 13.5 Test for optimality 443 ‘ 43.6 Degeneracy in Transportation problems 456 7 143.7 Unbalanced Transportation problem 463 i 13.8 Some miscellaneous solved problems 465 : 7 Ec Tea eas 14.4 Introduction Mathematical representation of Assignment problem Difference between Transportation and Assignment problem 486 Theorems on Assignment problem 486 Solution of Assignment problem : Hungarian Method 487 The maximal Assignment problem Unbalanced Assignment problem 448 Some miscellaneous solved examples 448 Travelling Salesman problem DECISION FHEORY .f 1 Introduction 15.2 Decision models 15.3. Steps of decision making process "154 Decision making situations “45.5 Decision making under certainty " Decision making under risk Decision making under uncertainty ior probability and Bayesian analysis mixed strategies es without saddle points lution of 2 x nand m = 2 GamesCharacteristic of a network Error in Drawing a network Construction of the network dia Methods of network ana Critical path analysis agram lysis Slack and float in network analysis Critical Path Project evaluation and review techniques ( Resource levelling and time cost trad Limitation of PER’ Comparison of CPM / PI f ana Introduction es of invento Meaning of inventory control Need of the inventory control entory co Types of inventory models Economic lot size m: Deterministic Model Model-1 nomic lot size model with unifor infinite Production rate with no short Model-2 : Economic lot size with differe Model-3 : E0Q production model; Ecc rate of demand, Finite rate Having no shortage Model-4: Fixed time model Model-5 : Economic lot size mode! of production and with shc Model-6 : Economic lot size model with Of production and having shorta Probabilistic Model 48.14 Model-1 : Single period model with Demand and time indepe 48.15 Model-2 : No set-up cost model; Single 48.16 Model-3 : The general single period m independent cost assumptions 48.17 Model-4: Probabilistic order-Ievel system 48.18 Advantages of inventory control : aoc . BE 49.1. Introduction 49.2 Queuing process 49.3. States of the system 19.4 Poisson process oc 49.5 More distributions19.6 Distribution of service time 696 19.7. Classification of queue models 698 19.8 Model-1 [M/M/1 : (/FCFS)] 698 19.9 Model-2: General Erlang Queuing Model : M/M/1 (cofFCFS) 704 19.10 Model-3 : WIM/1 : (NIFCFS) a7 19.11 Model-4 :M/MIC : («/F CFS) 709 19.42 Model-5 MIMIC : (NIFCFS): N>¢ 743 19.13 Model-6 :M/E\/1 : (~/ECFS) 8 19.14 Model-7: M/E,/1 : (1/ECFS) ae Machine repair problem ey eS es ea ICeeers 20.1 Introduction 743 20.2 Replacement models and their solution ae 20:3 General approach to solve replacement problems 744 20.4 Concept of present value 745 @ 20:5 Replacement when maintenance cost increases with time and the value of money remains same 746 20.6 Replacement of the item that deteriorates with time 748 20.7 Best replacement age of the items whose maintenance cost increase and value of money change with the time 756 20.8 Interval of optimum replacement 762 20.9. Present value for comparing replacement alternatives 763 2010 Replacement of items that fail completely and are expensive to be replaced 765 Mortality and mortality table Staffing problem Life testing techniques Group replacement method SIMULATION » Introduction ‘Simulation a ages and disadvantages of simulation numbers and pseudo-random numbers stic simulation generating pseudo-random numbers EeeSolution of a multi-stage problem of dynamic programming with finite number of stages Types of problems Solution of L.P-P. by dynamic programming ee Bor ventory problem by a dynamic programming technique 2 een dynamic and Linear programming problem 23 INFORMATION THEORY Ee E 23.1 Introduction 23.2 Communication process 23.3 Fundamental theorem of information theory 23.4 Statistical nature of communication system 23.5 Mathematical definition of information 23.6 Axiomatic approach to information 23.7 Measures of information 23.8 Measure of uncertainity : Entropy 23.9 Relations between entropies 23.10 Axioms for an entropy function 23.11 Uniqueness theorem 23.12 Channel capacity, efficiency and redundancy 23.13 Encoding 23.14 Shannon-Fano encoding procedure 23.15 Necessary and sufficient condition for noiseless coding Ear 24.1 Introduction 24.2 General sequencing problem 24.3 Sequencing decision problem for n jobs and two machines 24.4 Johnson's method (for n-jobs 2-machines) 24.5 Sequencing decision problem of n-jobs and three mach 24.6 Sequencing decision problems for n-jobs and m-machines 24.7. Sequencing problem involving two jobs and m-machines 24.8 Graphical method Pie al) Introduction 2. Maxima and minima of a function of several independent variables 53 Necessary condition for the existence of maxima or minima Necessary condition for maxima and minima ifficient condition for maxima and minima; The Lagrange’s condition ma and minima of the function of three independent variables ind minima for a function of three independent variables inge’s condition method of undetermined multipliers an multipliers method in non-Linear Programming tive function 22.725.11 Solution of non-Linear programming problems when constraints are Not equality constraints : Kuhn-Tucker conditions 923 aoe are 26.1 Introduction 937 26.2 Unconstrained minimization methods 937 26.3. Some direct method 938 26.4 Some more unconstrained direct search methods 26.5 Some Descent (or indirect) methods 26.6 Steepest Descent or Cauchy method 26.7 Fleicher-Reeves method 26.8 Constrained optimization problems 26.9 Characteristics of a constrained problem F-26.40 Constrained minimization methods 26.11 Direct methods © 26.12. Indirect methods A Eo ENP Resse ce ROO 27.1 Introduction 3 27.2 General non-linear programming problem PP) 27.3 Canonical form of non-linear programming problem 274 Mathematical formulation of non-lin Programming problem: 27.5. Graphical solution of a non-linear programming problem 28 NON-LINEAR PROGRAMMING TECHNIQUES: eee eM esl atialital) ahaa saa W Lis 28.1. Introduction 1001 28.2 Quadratic programming 28.3 Kuhn-Tucker conditions for quadratic programming | 28.4 Wolfe's modified simplex method 28.5 Beale’s Method 28.6 Separable programming {9 Mixed integer approximation of Separable problems 40 Validity of mixed integer approximation ‘Method of solution of Separable programming problem Brees eee eres ecrheoes OPERATIONS RESEARCH of isd felt son rela| Hone . | P ! . < | | i 7 a INTRODUCTION ): The term Operations Research was first coined in 1940 by McClosky and Trefthen in a sn town Bowdsey of the U.K. This new science came'into existence in military context. During World War Il, Britain was having very limited military resources, therefore there was an urgent need to allocate resources to the various military operations, and to the operation in an effective manner. Therefore, the Brit ctivities within each in military executives and manage called upon a team of scientists to improve the execution of various military proiects. Because the team was dealing with research on military operations, this new scientific approach ‘was called Operational Research and later it was adopted by U.S.A. as Operation: t (abbreviated as O.R.). The success of this team of scientists in Britain encouraged Unitec Siates, Canada and France to start with such teams. The apparent success of OR in the military, attracted the attention of industrial management inthe new field, who were seeing solutions to the complex problems. In this way, OR began fo creep into industry, business and governmental organisations. The growth of OR has not only been limited to U.S.A. and the U.K. but has reached many (bounties of the world. Availability of faster and flexible computing facilities and the number {qualified professionals enhanced the acceptance and popularity of subject. Indicative of this hat the International Federation of Operations Research Societies, founded in 1959 now Seomprises member societies from many countries of the world. Thus the impact of OR can be dt in almost all walks of life. 3 [ROHILKHAND-1995, 98, 2000, 2008, 08,08, 15] TAS the name implies, Operations Research involves ‘Research on Operations. This says ething about both the approach and the area of application of the field. Salient aspects Slated to definition stressed by various experts on the subject are as follows 4. OR is a scientific method for providing executive departments with a quantitative ‘basis for decision regarding the operations under their control ‘ abal (1946) entific method of providing executive with an analytical and objective basis (1S. Blackett (1948) proach (updated scientific method) and an inter disciplinary t complex functional relationships as mathematical modelszl OPERATIONS RESEARCH for purpose of providing a quantitative basis for decision making and uncovering new __ problems for quantitative analysis. - Thieanf and Klehamp (1975) @ ORs the art of giving bad answers to problems to which otherwise worse answers are given. -T, L. Saaty (1958) 5. OR is the application of scientific method by interdisciplinary teams to problems involving the controls of organised (man-machine) systems so as to provide solutions which best serve the purpose of the organisation as a whole. - Ackoff and Saseini (1968) G. The term OR has hitherto been used to connate various attempts to study operations ‘oF war by scientific methods. From a more general point of view, OR can be considered to be an attempt to study those operations of modern society which involved Organizations of men or of men and machines P.M. Morse (1948) 7. ORisthe application of scientific methods, techniques and tools to problems involving the operations of systems so as to provide these in control of the operations with ‘optimum solutions to the problem Churchman, Ackoff, Amoff (1958) #8. OR is a management activity pursued in two complementary ways - one half by the fee and bold exercise of common sense untrammelled by any routine and other half by the application of a repertoire of well established precreated methods ani ; techniques. Jagjit Singh (1968) ®. OR is an applied decision theory. It uses many scientific mathematical or logical means to attempt to cope with the problems that confront the executive when hi ities to achieve a thorough going rationality in dealing with his decision problems. - Miller and Starr NOR is the systematic method oriented study of the basic structure characteristics, finctions and relationships of an organization to provide the executive with a sound, Scientific and quantitative basis for decision making E. L. Amoff and M. J. Netzorg dem definition of Operations Research can be summarized as below Research is the applications of the methods of mathematical sele mplex farising in the direction and management of system of me cig IS and money in industry, business t and defence. T Ghis to develop a scientific model of systems i ing me inane and risk with which to predict and compare the outcomes of iStrategies or controls. The me! ts of factors alternatives, determine is to provide a scientific basis to the decision makers for solving the ions of system to give a solution which is in best interest of the ‘called the optimum solution to the problem. [ROHILKHAND-1996, 2000, 10) opment of Operations Research, the essential characteristics ch to find out optimum solution. ‘approach to the system. wallOR is the inter-disciplinary team approach to find out the optimum return: OR tries to optimize the total output by maximizing the profit and minimizing the cost. OR uses scientific methods to arrive at the optimum solution. OR takes into account all significant factors and evaluates them as a whole, i.e., OR considers the wholistic approach. . By OR techniques, we can not obtain perfect answers to our problems, but only the: quality of the solution is improved from worse to bad answers, The task of Operations Researcher may be said to consist of studying an operation, Selecting variables to describe the operation, investigating the mathematical relationships of the variables, selecting objectives, measures and criteria of system performance and finding means to improve or optimize the system's functioning according to the selected measures and criteria FES MobeLLING In oP ERATIONS RESEARCH [MEERUT-1998, 2008; ROHILKHAND Model plays very important role in OR. They are representations of reality. Mod distilled and economic descriptions and explanations of the operations of the sy Tepresent. By experimenting on them, we can determine how the chan System will effect its performance. Models enable us to experiment mor the system itself which is either impossible or too costl #4 model in the sense used in OR is defined as a representation « Situation. It shows the relationships (direct or indirect) and inter-rela Teaction in terms of cause and eff Since the model is an abstraction of reality, it thus appears to be le itself. For a model to be complete, it must be representative of those as being investigated. An OR me Tepresenti enables to c to find so [fe find some optimal solution of the problem (Models have many advantages over a verbal description of a problem. Some of them are as follows : 1, AnOR model provides a good grip over the problem. 2. It indicates the limitations and domain of the activity. &. An OR model provides a logical, scientific and systematic approach to the problem. 4, Through OR model, one can incorporate useful tools which help in eliminating duplication of methods applied to solve specific problems. 5. OR models provide avenues for new research and improvement in the system. pyantaces oF A Mobel few disadvantages also. Some of them are as follows are only an attempt in understanding an operation and should never be d an absolute in any sense. ‘of any model with regard to the corresponding operations can only be ng an experiment and relevant data characteristics.(a Operations RESEARCH OPERA ow Rewance © An OR model should take into account new formulations without having any significant change fh ints frame. * It should make minimum possible assumptions, * The model should accommodate a parametric type of treatment. * The model should be simple and coherent. The number of variables utilized by it should be less. Modelling is the essence of Operations Research approach. By Building Model, the 9. complexities and uncertainties of a decision making problem can be changed to a logical structure that is amenable to formal analysis » Ei tvees oF movers [MEERUT 2006, 09] ie *here are several types of models that are commonly used in OR as well as in most sciences 1. Iconic Models. Iconic models represent the system as it is but in different size Thus, iconic models are obtained by enlarging or reducing the size of the system. In ac ‘Other words, they are images. Some common examples are photographs, drawing, model aeroplanes, ships, engines, Plobes, maps, etc. A toy aeroplane is an iconic model of a real one. Solvi Analogue Model. it represents a system or an object of an inquiry by utilizing a set the mea lof properties different from what the original system processes. For example grapl solved b are very simple analogues because distance is used to represent the properties such. a {8 time, number, percent, age, weight, and many other properties. Contour lines on a map represent the rise and fall of the heights. In general, analogues are less speci less concrete but easier to manipulate than the iconic models, 2. Mathematical (Symbolic) Model. The symbolic or mathematical mode} is fone which employs a set of mathematical symbols (ie., letters, numbers, etc.) to represent decision variables of the system. These variables are related together by | means of a mathematical equation or a set of equations to describe the behaviour of D the system. The solution of the problem is then obtained by applying well-developed athematical techniques to the model. he symbolic model is usually the easiest to manipulate experimentally most general and abstract. Its function is more often explanatory rather than criptive. criptive Models. A descriptive model simply describes some aspects of a uation based on observation, survey, questionnaire results, or other available data. result of an opinion poll represents a descriptive model: ‘Model. Such models can answer ‘what if’ type of question, i-e., they can ions regarding certain event. For example, based on the survey results, works such models attempt to explain and predict the election results are actually counted. Normalative Models. Finally when a predi uccessful, it can be used to prescribe a source of action. Linear lative or prescriptive model because it prescribes what the ‘Such models assume conditions of complete certainty and ple, inear programming, transportation and assignment — jodels.Sol decisions can be made. For example, insurance companies are willing to insur against risk of fire, accidents, sickness and so on because the patt been compiled in the form of prc ee os ae Dyn Models. 1 ability distributions D these models, time is considered as one of the important variables and admit the impact of changes generated by time. rte models, not only one, but a series of interdependent required dug the planning horizon Static Models. These models do not consid adels do not consider the impact of changes that take SEER aning horizon, ic, chey areindependentof eae only one decision is needed for the duration of a given time period : GENERAL METHODS FOR SOLVING 0.R. MODELS IMEERUT-2009, 10; GARHWAL-2010; KANPUF Iwing 4 model consists of finding the values of the controlle dependent decisions is required during t the imeasure of performance, or of estimating them approxima solved by the following three models Analytic Method. If the OR model is solved by usin mathematics such as differential calculus and finite dif task, then such type of solutions are called analytic solutic inventory models are obtained by adoptir alled a Numerical Numerical ihe iterative or trial and error methods. Whenever the cl iterative procedure. The classical methods may fail beca onstraints or of the number of variable Inithis procedure, we start with atrial solution and a set of rules for impr Ieial solution is improved by the given rules and is then replaced Solution. This process of improvement is repeated Gsipossible or when the cost of further calculation can not be jus procedures are divided into three groups (@ After finite number of repetitions, no furth Gi) Although successive iterations improve the solutions, we are d the {solution asa limit of an infinite process. NGaD Finally, we include the trial and error method which how Jengthy, tedious and costly even if electronic computers are us 4. The basis of Monte Carlo technique is random sampling, Gable’ possible values. For this technique, some random numbers are required fay be converted into random variates whose behaviour is known from past fhe main steps of Monte Carlo method are as follow Pet tohhave a general dea ofthe system, we first draw a low di Ge take correct sample observations to select some suitable model forthe system. ep, we compute the probability distributions forthe variables of our interest Wvert the probability distributions to a cumulative distribution function. lof random number is now selected with the help of random number tables er is likely to be iagram of the system.1s | OPeRations Restancu STEP 5. Next, we determine the sequence of values of variables of interest with the sequence of random numbers obtained in step-4 STEP 6. Finally, we construct some standard mathematical function to the values obtained in step-5. ApvantAces oF Monre-Carto METHOD 1. These are helpful in finding solutions of complicated mathematical expressions which is not possible otherwise. 2. By these methods, difficulties of trial and error experimentation are avoided DisabvanTages oF Monre-CarLo METHOD 1. These are costly way of getting a solution of any problem. 2. These methods do not provide optimal answers to the problems. The answers are good only when the size of the sample is sufficiently large. (BRED Puases oF operations researc stupy [ROHILKHAND-1997 The major phases of OR study and their importance in solving the problem are described 1. Identification of the Problem. Identifying the problem correctly for the OR Study is main concern of most of the organisations. It is almost impossible to obtain the ‘right answer from the wrong problem’ Formulation of the Problem. To find the solution of a problem, one must be able to find the problem and formulate it so that it is susceptible to research. Formulation ‘of a problem for research consists in identifying, defining and specifying the measure ‘of the components of a decision model. For the formulation of a problem, the following information are needed : ) What are the objectives? Controllable variables and their ranges. ) The uncontrolled variables that may effect the outcomes of the available choices. Since, itis difficult to extract a right answer from the wrong problem, therefore. this d with the reformulation of the problem in an appropriete form which is enient for analysis, The most suitable form of this purpose is to construct a ical model representing the system under study. It requires tie identification and dynamic structural elements. A mathematical model should include ‘three important basic factors : input data. Once an appropriate model has been ita required by the model is to be collected. This information I kept records, from current tests and experiments or even fromAfter selecting input data, the next phas is to derive a sol a solution from t Inthe search fi om this model. Herein Operations Rescarch, we re al inthe sea an optimal solution. An optin maize o nimize the objective function inthe model inti est interest ofthe pb consideration. Many procedures have heen develope interest of the problem under have been developed and will be discussed in detail again tested as a whole for t eels acre or the errors if any. A model may be can provide a reliable predicti h Operations Research realises that this model k he updates the mod. hee model time to time by takin ceroraherokea Onee placed on ¢ the model is no longer an ai controls should be any significant changes in the conditions change so much that the system, the model is invalidated. C be established to detect the hanges ii revised to reflect these changes. Research study is to implement the optimum solutic conditions are constantly changing in the world remain valid for a long time. Therefore soon as possible so that the model, its solu __be modified accordingly. BR} means th m ‘at each constraints may assume only one of three possible forms, (vi) The restriction, *i= 0 implies that x's must be non-negative In compact form the general LPP can be written as follows. optimize (maximize or minimize) z= (objective function) subject to ¥ a, 2b i iA (constraints) such thatx) = 0,j = 1,2,...,n (non-negative restrictions) © The input-output coefficients aj; are also known as structural coefficients or technological coefficients Which represent the exchange coefficients of the i decision variable in the i constraints, ® The constraints are generally expressed by "less than or equals (<)" sign in maximization problem. ‘Where as in case of minimization problem, it can be expressed by ‘greater than or equal to (2) sign. 1%. Aq 4g) Gy2 + Ay to the constraints Gm Ama mn NL bm aPh ) The objective of a diet problem is to ascertain the quantities of certain foods that should be eaten to meet certain nutritional requirement at a minimum cost. The consideration is limited to milk, green vegetables and eggs and to vitamins A, B, G, The number of milligrams of each of these vitamin contained within a unit of each food and their daily minimum requirement along with the cost of each food is given as belo eisai a 1 10 Img B 10 10 50 mg c 100 10 10 mg 10 8 Formulate a linear programming problem for this diet problem. [MEERUT-2004, 10, GARHWAL. SOLUTION. Let the diet contain x, litres of milk, xp kg of vegetables and x3 dozens of egg Then total cost (2) per day in rupees is Z = 20x; + 10x, + 8x3 (objective function) Further, total amount of vitamin A in daily diet is (& + xp + 10x3)mg According to question Xt Xp + 10x21 Similarly for vitamin B and C, we must have 100x, + 10x, + 10x3 > 50 10x; + 100x, + 10x3>10 | @ the quantities of different food items to be consumed can not be negative * x1 2.0, x) 20,320 (Non-negative restrictions) “Hence, the mathematical model of given LPP is given below Minimize Z = 20x, + 10x, + 8x3 (objective function) subject to Xj +X2 +10x3 21 100x; +10x2 +10x3 250 (constraints) 10x; +100x2 +10x3 > 10 X 2 0,x) 2 0,320 (Non-negative restrictions) (Constraints) ®LPP: FORMULATION AND GRapuicay, SOLUTION: is Vitamin |itre of mitk| Vegetables | Eggs | Minimum daily (in kg) | (ozen) | _ requirement A 1 1 10 1lmg B 100 10 10 50 mg Cc 10 100 10 10 mg Cost in® 20 10 8 Formulate a linear programming problem for this diet problem. [MEERUT-2004, 10, GARHWAL-2008) Let the diet contain x; litres of milk, x, kg of vegetables and x, dozens of eggs. Then total cost (Z) per day in rupees is Z = 20x, + 10x) + & (objective function) ...(1) Further, total amount of vitamin A in daily diet is (xy +2 + 10x3 i ae i. Similarly for vitamin B and C, we must have | 100x; + 10x, + 10x, 250 | (Constraints) 0°) and 10x; + 100x3 + 10x5 > 10 4) + Also, the quantities of different food items to be consumed can not be negative, 30 x1 2 0, x2 2 0, x3 20 (Non-negative restrictions) Hence, the mathematical model of given LPP is given below: Minimize Z = 20x, + 10x + 8x5 (objective function) eat ee (constraints) 100x, + 10x, +10x3 250 10x, +100x2 +10x3 2 10 x1 = 0,2 = 0,x320 (Won-negative restrictions) ‘such that dealer deals in two items vis, tables and chairs. He has ‘to invest and a space to store almost 60 pieces. A table % 500 and a chair of % 200. He can sell a table at profit Sand « chair at a profit of = 15. Assume that he can sell all 3 at he buys. Formulate the problem as an LPB, so that he the profit. ber of tables and x, be the chairs 50x;OPERATIONS RESEARCH and, the profit on x) chairs = © 15x. % = total profit Z = 50x, + 15x2 Here, we have to maximize the profit Z. Now, the cost of x; tables = 500x, and the cost of x3 chairs = 200x = total cost of x; tables and x» chairs = 500x, + 200x2 Since, the dealer invest & 10,000, so 00x, + 200x) < 10000 (2) Also, ota tenis 3G ‘EXAMP which can be atmost 60 therefore X1 + X2 5 60 (3) Further, since number of tables and chairs cannot be negative Therefore, x1 20,7 20 Thus, the mathematical model of the given LPP is as follows: Maximize Z = 50x, + 15x2 (objective function) : subject to 5x, + 2x9 $100 (constraints) xy +x2 $60 pie such that 20 20 (non-negative restrictions) ae EXAMPLES. A goldsmith manufactures necklaces and bracelets. The total number of necklaces and bracelets that he can handel per day is atmost 24. It takes one hour to make a bracelets and half an hour to make a necklace. It is assumed that he can work for a maximum of 16 hours a day. Further the profit on a bracelet is 300 and the profit on a necklace is 2100. Formulate this problem as an LPP so as to maximize the profit. Let us suppose Total number of necklaces manufactured = x1 and Total number of bracelets manufactured = xp Since the profit on a necklace is $100 = Profit on.x necklaces = 100x; Similarly, the profit on a bracelet is €300 => Profit on xp bracelets = 300x> B Total profit Z = 100x, + 300x. =) To maximize the profit, we have to maximize Z. Further, it takes one hour to make one bracelet, +. Total time required to make x2 bracelets = 1 - x» = x5 hours Also, it takes half an hour to make one necklace, # Total time required to make x, necklace = (1/2)x, SOLUTIO? So, total time required to make x; necklaces and x9 bracelet XL Aly x, soe Here, total time available per day is 16 hours, so we can write ~L +x, <16 2 1 + xy $32 @ EXAME tis also given that the total number of necklaces and bracelets that the goldsmith eae manufacture in a day is atmost 24, xy +2524 --@)LRP:| FORMULATION AND GRAPHi¢Ay SOLUTION: s Also, al model of the LPP j s given by Maximize Z = 100%, + 300%, tL BEN areas (objective function) x +x) <24 (constraints) such th = ) P fat X12 0,x)20 (non-negative restrictions) EXAMPLE 4. A dietician decide, ‘i P « r. The first one gives 7, 5, 2 Units per gram of vitamin A, B, C respectively. The first food costs 2 per gram and seco grams of each food stuff the food expense at a min problem for this problem. Tt is often convenient to construct the following table problem carefully. nd 71 per gram. How many should the family buy everyday to keep imum? Formulate a linear programming SOLUTION. after understanding the Foods Vitamin A Vitamin B | Vitamin c Xen 7 5 2 2 ¥(x2) 2 4 s = Minimum BS 20 16 A daily needs Glearly, — Z = 2x +x Since, we have to Keep the expenses minimum, So we have to minimize Z = 2x; + x2 @ Also, we observe that Tx + 2x, 230 Sxy + 4p > 20 © Dey + Bx) > 16 a 33 2 ox eat model ofthe LPP is given below Gime 2-2 + ene race) in = subject t0 7x, + 2x, 230 : 5x, +4x2 220 (constraints) 2x, + 8x, 216 such that x, 20,220 (non-negative restrictions) s 20, pany inary doll ictures two type of dolls; an ordinary ie 4 Ber eeean type of doll B takes twice as tans to a ee of the type A. It is given that the company ‘to make a maximum of 2000 dolls per day if it producesOPERATIONS RESEARCH LPP o i ion. The supply of Plastic is sufficient to ea a Petit per Wa) (both pend B combined). The deluxe Profion requires a fancy dress of which there are only eo pieces per day available. If the company makes profit of 30 and 750 a Poll respectively on doll A and B, "yormulate the problem as an to maximize the profit. [MEERUT 2007, 09, 1] SOLUTION, Let the no. of dolls of type A = %1 ‘and the no. of dolls of type B = x2 Ife hours are required to produce one doll of type produce one doll of type B will be 2t hours. Now, since the time available per day is 2000¢ hours Therefore, yt + Xp“ 2S 2000¢ > xy + 2x, < 2000 Further, since the supply of plastic is s (both types) So, xy + x2 < 1500 (2) A then the time required to (1) sufficient to produce 1500 dolls per day dress for deluxe version of doll B is available for 600 pieces per ‘Again, the fancy day only So, x2 $ 600 @ oan ‘Also, total profit on both the types of dolls per day is Z = 30x, + 50x2 4) We have to maximize Z. Also, no. of dolls can't be negative, ie., x; 20,220. Hence, the mathematical model of LPP is given as below: Maximize a 30x, + 50x, aeuN ry drSLPP: FORMULATION AND Gr. ‘APHICAL SOLU she available capacities o Betas for the given ume posts °F grinding, 60 for turning Ba ft And 200 for testing. Profit on each £30and that for each unit of Bi heproblem as a linear progr to maximize the profit assuming thar yet HrgAll all the items that it pens Ppevaling market price, | "USE At the _ A resourceful home decor two types of lamps say A and go through iets a Bebe lamps. econdia finisher. Lamp A requires 2 hours fhe enters time and 1 hour of the frst ate os ees 2 hours of finisher’s time. The cutter has 104 Biers ie i avers available each month, Profit on the lamp A Bfe60 and on the lamp Bis € 110. Assuming MMT can geil alll that ‘he produces hhow many of each type of lamps should be manufacture per month to obtain the best returns? Formulate a LPP for this problem. 20 for assembling Ach unio Aig 20, Formulate ‘or manufactures ‘The manager of an oil refinery must decide on the optimal mix of 2 possible blendin; process of which the input and output per production run are as follows 1 a ‘The maximum amount available of crudes A and B are 500 units and 400 units Tespectively, Market demand shows that * fers300 units of gasoline X and 260 units of gasoline Y must be produced. The profi fx) production run from process 1 and ?.aV% 400 ‘and 2500 respectively. Formulate ! maximizing the profit. "6, Afactory produces two products ‘Aand B. To er etceire one unit of product A 2 mache a “has to work for u hours and a craftsman to work for 2 hours. TO manufacture st Fproduct B, the machine has Wt work for Bet aroan for one bow: 18% factory can avail fof 80 hours of SY 7p hours of erafesman Ee nit of A and the sale of each UPN if the wan nc nal om frat oe Nee gue ened nate at oe tees neha oe rd proces tect Reet Products A B G 4 3 H 2 2 Machines Gand HT have 2000 and 2500 machines ‘minutes respectively, The firm must manufactures 100 A's, 200 Bs and 50 's but not more than 150 4's, Setup a linear programming problem to maximize profit A manufacturer produces three models 1, Il and Ill of a certain product. He used two types of raw material A and B of which 4000 and 6000 units respectively are available The raw material requirements per unit of the three models are given below [maw] Requirement per unit |material] of given model a lt Machine A 2 3 B 2 The labour ime for each unit of model I is swice that of model Il and three times that Gf model Il, The entire labour force of the factory can produce the equivalent of 2500 mits of model I. A market survey indicates the minimum demand of the three models tre 500, 500 and 345 units respectively formulate the problem as a LPP in order fo determine the number of units of each product which will maximize profit, complete unit of a certain product const a four units of component A and three Unt or resmponent B. The two components (4 of fg) are manufactured from swo differoy sials of which 100 units and 200 raw materials ee tinits respectively are departments are engaged in process with each, departing i Bimerent method for manuleeore, iy aieponents, The following table Bo cow material requirements PE" ction. raw nd resulting units of each ‘component.OPERATIONS RESEARCH The objective is to determine the number of more than 6 advertisement and that B ang 1 objective is to determi we at least two advertisement, ition runs for each department which C each have at arenienigethertalnumberofeomponent Formulate the LPP model. (CA. NOV- se units ofthe final product. 11. A city hospital has the following minimal CES: daily requirement for nurses Input per Depart-jereni(unite) “| _run (units) Period | “Hock Time ca oe ment | Raw Material | Components (24 hr. day) vom a iene sae ure 1 |6am-10am 2 1 TiS 6 4 2 |10aM-2PM Z B 4 8 5 8 3 | 2PM-6PM 1s EB 2 Ea 7 3 4 |6PM-10PM 8 Formulate a linear programming model to the above problem. s |10PM-2AM 20 The owner of the Metro sports wishes to 6 | 2aM-6aM 6 determine how many advertisement to place in the selected three monthly magazines A, Band C. His objective is to advertise in such @ way that total exposure to principal buyers ‘of expensive sports good is maximized Percentage of readers for each magzine are Known. Exposure in any particular magazine is the number of advertisements placed Nurses report to the hospital at the begining of each period and work for 8 consecutive hours. The hospital wants to determine the minimum number of nurses available for each period. Formulate this LPP [MEERUT-2015) 12. A tyre factory produces three types of tyres milkiplied by the number of principal buyers. T), T, Ta. Three different type of chemicals SME Srie ata may be used say Cy, Cz, C3 are required for production. One 7; tyre needs 2 units of Cy, 3 units of . Magazine Cs, one Tp tyre needs 3 units of C,, 2 units of ad ae G; and 2 units of C, and one 7, tyre needs 5 units of Cp and 4 units of C. The factory has ‘ mGnlakhs) | 1 | 0.6 | 0.4 only a stock of 20 units of C), 25 units of Cp “Principal buyers | 10% | 159 | 79 and 30 units of Cj, Further the profit from IGost'per the sale of one tyre 7; is 260, one tyre Tp is ieut (in 2) 2100 and one tyre of type T is €80. Assuming that the factory can sell all that its produces. "The budgeted amount is atmost 7 1 lakh for Formulate a linear programming to maximize 5000 | 4500 | 4250 the advertisement. The owner has already ee Meee that magazine A should have ng © B*PPOFE mize Z = 30x + 40%) + 20x + 15x, 2. Maximize Z = 20r, + 30, ct to the constraints subject to the constraints x #22400 2x; + x7< 600 and x, 20; x)>0 4. Maximize Z = 60x, + 110x, subject to the constraints xy + x25 104 2 + 2x<76 and xy 20;x220ORMULATION AN ID GI "RAPHICAL SOLUTIONS Maximize Z = 400x, ss Beet tem? eee amine 6. Maximize Z = 100x, + 80x) ea Se rn a <400 subjets ws che comnts i542 80; 90, 7 Ses 1LSx + 34 <80 i. 1205x220 2x, +2570 laximize Z = 30x, + 2 and x; 2 0}x22 ‘subject to the a eee nie 2 6 eee . Maximize Z= 60s, + 40x + 100% Bees, + s,.< 2000 subject to the constraints 2x, + 2xp + 4x < 2500 Seles ee fry + 2xg + Pay < 6000 Gx, + Bxy + 2xy < 15000 2a, Sea 29 and x2 $00; x22 500; x52 375 Maximize Z = V, where V = min £6 + 5x, +79), (4 +849 +39) ‘subject to the constraints . a 4 i Gx + Sz + 7x5 ~4V 20; Ary + Bx + 3x -3V20 xy + 4x9 + 2xg < 100; Sxy + Bxy + 7x5< 200 and x1, x2, x3, V20 ‘i Maximize 2 = 10000r, + 9000x, + econ: eee yee S000x + 4500x + 4250: < 100000 nae [ee ele ee, 22x32 2 tna ee rit x20 Sie Se eae Maximize Z = 60x; + 100x, + 80x; i soi oR Subject to the constraints Bry + 3x, <20 Dx; + 5x3 < 25 Bx; + 2x2 + 4x3 < 30 and x, 2 0, x2 20,x320 100- m), a solution obtained by setting variables is calledOPERATIONS RESEARCH Fa ates whowe value did not appear inthe above solution are called non-basie z ee sles and the remaining m variables are kriown as basic variables 7 Basie Feacraie SowuTION aeons: also basic solution, ie., all basic variables z A feasible solution to a general LPP which is assume non-negative values is called basic feasible solution Generally, basic feasible solutions are of two types @ Degenerate Basic Feasible solution: A basic solution to the system of equations is called degenerate if one or more of the basic variables become equal to zero. Gi) Non-degenerate Basic Feasible solution: A basic solution is called non degenerate if values of m basic variables are non-zero and positive. 3.12.6 Optima: Basic Feastsue So.TION ‘Any basic feasible solution which optimize (maximize or minimize) the objective function of a general LPP is called optimal basic feasible solution. 3.12.7 Uneounoe SouuTon A solution which can increase or decrease the value of the objective function of an LPP s indefinitely is said to be an unbounded solution. 3 BUZ8 Frasier Resion Es ‘The common region formed by all the constraints and non-negative restrictions of is called feasible region. = BAZ. Incase Resion The region common to all constraints in which all the decision variables are negative is talled infeasible region E 3.12.10 Convex Resion s IW the line segment joining any two arbitrary points of the region lies entirely within the ‘region, then this region is said to be convex region ERE RcraPuical souution meTHoDs oF AN LPP s To obtain the solution of an LPP by graphical method, hi e fi odin grap! l, we shall recall the following a (Q) An optimum solution of an LPR if it exists, occurs at one of the extreme points, ie. comer points of the convex polygon of the set of feasible solutions (Fundamental ‘extreme point theorem) {Q) Ifthe optimum solution occurs at more than one e ject fnction wil be the same forall Convex Combinations of tose one Soe (3) The collection of all feasible solutions to an LPP constitutes a convex set whose ‘extreme points correspond to the Basic feasible solution, | There are a finite number of basic feasible solutions within the feasible sokution space. ) If the convex set of all feasible solutions of the system of simultaneous equations mS BX 2° is a conven polyhedron, then at least one of the extreme points gives point ofthe feasible region fall atthe intersection of two constraints equalities. point of the convex set give the basic feasible solution toan LPR LPP es aform, then formulate the probe e the, Mien problem is and an objective function, ; : Each inequality in t Rattable by solving ihe. equate’ etme ee s arity equation. Similarly, give ariable and find the corresponding value of the of ; of values. Connect these points by a straight constraints and get as many straight line as there are i line representing one constraint (aie STEP 3. We have to identif ‘i constraints simultaneously. For "greater than will be the area which lie above the constraint this area is generally the region below these lines. Or region includes the points on the constraint STEP 4. Select one of the following two techniques o Since, we know that an optimal solution of an LPP of the feasible solution space. Thus, first we determ the feasible region and then compute the value then compared. The steps of the solutions will be clear by the Identify each of the corner p’ of simultaneous equatior Determine the coordinate of « é STEP 3. Compute the value of object a . STEP 4, Identify the extreme point that gives 01 The point where the objective function of the LPP 7 Rows. * If two vertices of the Con then all points on the line objective function. In this a gment joinin the LPP is said to have infin To determine which satisfies the constraint {are feasible points. if it Jare feasible points MerHo0 “method, the optimal solution ofit line is 4 collection of point Bicening different valucs '0 7, fe method are as follows: he objective noes ‘of the objective et pe spe of is found by using the slope of Ul ts which designate with same ¥# iget different profit lines. The *OPERATIONS RESEARCH [ne dae ee ee ere é Identify the feasible region and its extreme points ‘Assign a constant value say Z; to the objective function and draw the corresponding line of objective function (called iso-profit line). say Z, to the objective function and draw the STEP 1. STEP 2. STEP 3, Assign another constant val corresponding line of the objective function, STEP 4, If Z; > Zp and objective function Z is to be maximized then we move the line corresponding to Z; to the line corresponding to Z» parallel to itself as farthest point within the feasible region is touched by this line and any further displacement of this line takes it out of the feasible region. The coordinates of the farthest point so obtained will give the maximum value of Z. But if we have to minimize the objective function Z, we have the line corresponding to Zp to the line corresponding to Z; and find the nearest point in the same way as we had find the farthest point. Then, coordinate of the nearest point will give minimum value of the objective function. Since every point in the feasible region satisfies all the constraints of an LPP and there are infinitely Iman points so itis not easy to anyone to find a point that gives a maximum or minimum value of fhe objective function. To handle this situation, following results TEER be the feasible region for an LPP and let Z = ax + by be the objective function @ When Zhas an optimal value, where x and y are subject to constraints describe by the line or eems very useful inequation, this optimal value must occur at a corner point of the feasible region, (i) IFRis bounded, then the objective function Z has both a maximum or minimum value at that = point on R. ow Reve ® The iso profit line method is also applicable if the problem has more constraints See Type |. Base oN Finoinc THE SOLUTION set oF SimuTANEoUs LiNEaR INEQUATIONS EXAMPLE! Draw the graph of the solution set of the inequations 2x + 3y > 6 x + 4y<4,x>0andy>0. The corresponding equations of given inequations are: 2x + By = 6x4 4y=4,x=0,y=0 Region represented by 2x + 3y > 6. The line 2x + 3y > 6, meets x-axis at AG, 0) (puty = 0) and y-axis at B(0, 2) (put x = 0 in it). We find that (0, 0) does not Satisfy the inequation 2x + 3y > 6, the portion not containing the origin represented by the given inequation represents the solution set.Notations line @ 2 + 3 line @ x+y = 4 | Point A= 4 (3,0) Point Ay = A, (4,0) Fig. 1 Region represented by x + 4y < 4. The line x + 4y = 4 meets xaxis at AiG, 0) (puty = 0 in it) and y-axis at B,(0, 1) (put x = 0 in it). We find that (0, 0) satisfy the inequation x + 4y < 4, So, the portion containing the origin represented by the given inequation represents the solution set. Region represented by x > 0, y > 0, represent the first quadrant. Hence, the shaded region given in the figure represents the solution set of the given linear inequations. Draw the graph of the solution set of the linear inequations. X+Y<5, 4x +y24,x + 5y25,x<4,y<3. The corresponding linear equations of the given linear inequations are x + y ax ty =4,x 4 Sy =5,x=4y=3. ‘Notations| line @ x+y line @ : 4x +y line @ x +5) line :x=4 line © :yLPP: FORMULATION AND Graricai souatons i Region represen: a ted by the 3x + 4y > 12. The line 3x + 4y = 12 meets the -axis at B,(0, 3) (put x = 0 in it). Join these é Clearly, sfy the given inequation. So, porion ie Bare the origin is represented by the inequation 3x + 4y > 12. represented by x = 0, ight y- Reais) 'Y X > 0. Clearly, x > 0 represents the right y-axis Region represented by y > 1. Clearly, Gistance of 1 unit from it. Since, (0, 0) di containing the origin is represented by the Hence, the shaded region represents the sol Y = Lisa line parallel to x-axis at a loes not satisfy y > 1. So, portion not given inequation lution set of the given inequations. Solve the following system of inequations graphically, ax+y-350 x-2y+1<0 (20, y>0). The corresponding linear equations of the given linear inequations are: x+y-3= 0, x-2y + 1=0,x=0,y =0. Region represented by 2x + y - 3 > 0. The line 2c + y - 3 = 0 meets the 2caxis at 43.0) (puty = O init) and y-axis at B(O, 3) (put x = 0 in it). Join these points by a thick line. Clearly, (0, 0) does not satisfy the inequation 2x + y - 3 > 0. So, portion not containing the origin is represented by the given inequation 2+ y-320. ay: line @ : 2x +y— line @ : x-2y + Point A=A (3, 0) Point B=3 (3) Point Ay = (1,0) Point By = ( Jd by x= 2y + 1.< 0. The line x-2y + 1 = Omeets the 0) andy-axis at a(og -Join these points by a tick line, Clears : satisfy the inequation x ~ 2y + 1 < 0, So, portion not containingOPERATIONS RESEARCH i Deract ty the constraints 3x, < 180 and 5x2 < 200 are plorted on the graph Similarly, the constraints 3%) < Notations 4s Tine @: 4x; + 6x2 = line @: 3x; = 180 | 100 line @ : 5x2 = 200 oe Point A = A(60, 0) 2 Point B = B(60, 20) ~~ 2 2 Point C= C(30, 40) = 60 © 20 40 60 80 90™100 *! Clearly the above shaded area OABCDO is the feasible region. We know that the optimal value of the objective function occurs at one of the extreme points of the feasible region. So, we have to evaluate the value of the objective function at each extreme point of the feasible region as shown below Coordinates | Value of the objective function SO er ae Ze ae + 10x. oO (0, 0) 15x0+10x0=0 A (60, 0) 15 x 60 + 10 x 0 = 900 B (60, 20) 15 x 60 + 10 x 20 = 1100 ic (30, 40) 15 x 30 + 10 x 40 = 850 D (0, 40) 15 x 0 + 10 x 40 = 400 Itis clear from the above table that Z is maximum at B(60, 20) and its maximum value is 1100. Hence, optimal solution to the given LPP is x, — 60, x» — 20 and Max Z = 1100. EXAMPLE2, Solve the following LPP by graphical method Maximize Z = 4x, + x subject to the constraints 1 +x2<50 3x1 + x2 < 90 and Xy x2 20 SOLUTION.STEP 1, First consider the constraints as equations, ie., X1 + x) = 50 Bx, + x) = 90 STEP 2, In order to draw x, + x» = 50, put x, = O we get x; = 50 and by putting 7 X1 = 0, we get x, = 50 = The linex; + x2 = 50 passes through the point (50, 0) and (0, 50). Now, plot the points (50, 0) and (0, 50) on coordinate axes and joining them by a straight line. SeamLPP: FORMULATION AND Grapinicay SOLUTIONS Similarh | 73 | imilarly, we dr: STEP 3. Since, x; + x, eee line 3x, + xy = 90, $50 an x pices xy + % 3g = 50 ine ees 5 20 therefore, the region below the line ee is the possible region and Tegion. So the common bea’, 1,00 ithe Positive quadrant is the ae feasible region, gion denoted by the shaded region OABC is the Ba) Notations] Hine @ =x, + = 50 tine @ :2 + 3 = 90 Point A A(30,0) Polat B = 8(20, 30) Point J area OABCO is the feasible region. We know that the optimal value of the objective function occurs at one of the extreme point of the feasible region. So, we have to evaluate the value of the objective function at each extreme point of the feasible region as shown below: Coordinates ‘Value of the objective function Extreme point Gp 2). 4x; + Xo oO (0, 0) 4x0+0 A (30, 0) 4x 30 +0=120 B (20, 30) 4x 20+30=110 (0, 50), 4x0+50=50 ‘ble that Z is maximum at A(S0, 0) and its maximum Clearly, the above shaded a c s clear from the above tal is 120. the optimal solution is given by *1 = 30. (0 and optimal value of Z is 120. following LPP by corner point method. ie Maximize Z = 60x; + 40%2LRP: FORMULATION /AND GRAPHICAL SOLUTIONS Extreme point | Coordinates | value of the objective function Gy, x) Z = 60x; + 40x. A (5, 0) 60 x 5 + 40 x 0 = 300 es (6,0) 60 x 6 + 40 x 0 = 360 c (4,4) 60 x 4 + 40 x 4 = 400 D (0,6) 60 x 0+ 40 x 6 = 240 E (0, 4) 60 x 0 + 40 x 4 = 160 Itis clear from the above table that Zs m aximum at C(4, 4). Hence, the optimal Solution is given by x, = 4, x» = 4 and optimal value of Z is 400. Tyre 4. Based on Corner Pont Metuop: Minimization Prost BAMPIES. Solve the following LPP by corner point method: Minimize Z = 4x, + 3x9 subject to the constraints 200x, + 100x, > 4000 X1 + 2x) >50 40x; + 40x, > 1400 and Xp X2>0 SOLUTION.STEP 1. Consider the constraints as equations, ic, 200x, + 10x, = 4000 x + 2x9 = 50 40x, + 40x2 = 1400 [Notations line @ : 200x, + 100x; = 4000 ine @ 2x; + 2a = 50 Hine @: 40x, + 4 Point A= A(0, 40) 1400 Point C= 6(20, 15) Point D = D(S0, 0)OPERATIONS RESEARCH fhe point (20, 0) and = 4000 passes through the p , sTeP2, The line 200s # 100%2 5." 50 passes through the point (50, 0) and Oe rticiae = 1400 passes through the points line 40x; + 40x. = is Pete 25) ‘Plot these points on the coordinate axes and join them fy gstraigh ine as shown in the adjoining graph. : Clearly, the above shaded area in the positive quadrant is the feasible region Now, we have to evaluate the value of the objective function at each extreme point of the feasible region. Coordinates | Value of the objective function Extreme point Gey, x2) Z = 4x, + 3x2 A (0, 40) 4% 0+3x 40 = 120 B (5, 30) 4x5+3x30=110 c (20, 15) 4x 20+3%x 15 = 125 D (50, 0) 4x50 +3 x0=200 Clearly the point B give the minimum value of the objective function. Hence, the optimal solution of the given LPP is x; = 5 and x2 = 30 and optimal value of Zz 10. EXAMPLES, Solve the following LPP by corner-point method. Minimize Z = 3x, + 2x2 subject to the constraints 5x, + x2 >10, x, + x2>6, X1 + 4x2 212, xy 4220 (MEERUT. 2008} Consider the constraints as equations 5x, +x) = 10 x $x =6 x + 4x = 12 X1%220 Then apply the usual procedure (as discussed earlier). Plot these equations on graph paper and by using the inequality condition of each constraints to mark the feasible region. Hine @: 5x, + x, line @ :x,'+ x, line @:x, + 4x Point A= 4.12, 0) Point B= 8 (3,2) Point C=C (1,5) Point D =D (0, 10) m i> ‘soLu’—1x0+2 -1x2+2x0 -1x4+2x2=0 1x2+2x4=6 10 _ 20 -1x0+2x>= 3 occur at the extreme Xp = O and te alue of the objective function Z = solution of the given LPP is : x1 Glearly, the minimum vi point A(2, 0). Hence, the optimal EXAMPLEL Using, iso-profit method, find the maximum value of the function Z = 2x, + 3xq subject to the constraints 2x, +x. <5 X1-%2<1 x2<2 and x, >0,x,20 [MEERUT-2008] SOLUTION. Converting the given condition into equations and drawing these lines, the permissible region is OPQRSO (shaded region) which is the set of all feasible solutions of the problem. We have to maximize the objective function Z= 2x, + 3x ‘iy soLu 5 line @ = x, ~ line @ = x, = 2 line @ Point P Point Q Point R = 0 is shown by dotted line through the origin 9LPP: FORMULATION AND GRapuicay SOLUTION: I : iS which is parallel to the j ee ss, i ‘So-profit line, Now draw dotted parallel lines away from ithest line from the origin passes through aS 2) > Since, the farthest | ine from the origin O passes through the vertex a(S 2) Bugs Hence # the required optimal solution is given by x =2\x,=2 end Max. Z Ix3+3x2=9 Solve Pp sollowing L.BP. graphically using Iso-profit method. Maximize z = 2x, +x subject to the constraints. 5x, +10x2 <50 X+x_21 x2 <4 x -x,<0 and X12 0,x2 20 To solve the above L.PP we proceed as follows: STEP 1. Making all the constraints as equations: xy-%2=0 STEP 2. Draw these lines on the x,x2- plane as follows: The line 5x, +10x, -50 passes through the two points (10, 0) and (0, 5) Plot these points on the co-ordinate axes i.e., x-axis and x,~ axis respectively and join them by a straight line, The line x +x)=1 passes through the points (1, 0) and (0, 1). Plot these point on the x-axis and x, axis respectively and join them by a straight line. Similarly, we draw the lines x, = 4 and x, -x2=0 Gince sx +10x) $50, x) +221. $4,m —%2 $0. So the regions below the 5x, 420% =50 in the positive regions, and the regions above the line land x,—x2 =0 in the positive quadrant is the possible region. jon common to all above regions is the feasible region, which is jn following figure by the shaded region ABCDEA.OPERATIONS RESEARCH LPP Fig. 14 STEP 4. Iso-profit Method: (a) Assign a constant value Z,=2 (LCM. of 2 and 1) to the objective function Z, we get a line 2x +x Now draw this line (1) which is as shown in figure by P,Q, . (b) Again assign a constant value Z,=4 to the objective function Z, we get aline (1) 2x +x) =4 2) Now draw this line (2) which is shown in figure by P,Q,, clearly both the lines given by eqns. (1) and (2) and parallel to each other. Since Z, <7 and the L.PP is of maximization, so we move the line Qs from AQ, to P,Q, parallel to itself as far as possible, until the farthest point @ 50 2(50 50 15'15 further displacement of the line takes it out of the feasible region ABCDEA Thus, the point (ie $e Jobrainea on solving the lines x,-x,=0 and 5x, +10x, =50, gives the maximum value of the objective function Z, and the line P,Q; is the line of maximum objective function. Hence, the optimal solution isx, = °° )sitnin the feasible region ABCDEA is touched by the line. Any the optimal value ofORMULATION/AND|GRAPHICAL SOLITONS DXAMPLES. Solve the Jollowing L.RP by graphical method: ; ____ Maximize z_ 3x 4 subject to the constraints 1+ xq %1 +X $1500 %1 + 2x9 < 2000 X2 < 600 and X20, x2 20 (GORAKHPUR-2010) SOLUTION.STEP 1. Making the given constraints as equations: xy +x) =1500 x + 2x, = 2000 x)= 600 STEP 2. The line x,+x,=1500passes through the two points (1500, 0) and (0, 1500). Plot the points (1500, 0) and (0, 1500) on x,- axis and x,- axis respectively and joining them by a straight line. The line x, + 2x, = 2000 passes through the points (2000, 0) and (0, 1000) Plot these point on x; axis and x, - axis and joining them by a straight line. The line x, =600is parallel to the x-axis and passing through the point (©, 600). STEP 3. Since we have, x1 +p $1500, x, +2, $ 2000, xp $600 and x; 20, x20 Therefore, the region below the line x, +x) -1500 in the positive quadrant is the possible region, the region below the line x, +2, =2000 in the possible region, the region below the line xj =600in the positive quadrant is the possible region x2, tine @): x, + tine @) x, + 2x2 tine @): x = 600 Point B = B(1000, 500)OPERATIONS RESEARCH LPP: FoR! in the positive quadrant three regions shaded region ommon to above hown in figure by the Thus, the region c hich is as s! is the feasible region, Ww OABCDO. rcchod 4, Iso-profit Method: . J STEP. Isoprofit MetnOr’ ._ssopto the objective function Z, we Bet aling 3xy +5x) =1500 () ‘as shown in figure by 7.1 Now, draw this line (1), which is ive function Z, we (©) Again assign a constant value Z, =3000+0 the object get a line 3x, +5x2 =3000 Q) ), which is as shown in figure by P22 Now draw this line (2 arallel to each other: | Clearly, both the lines AQ, and P,Q. are Pa Since 2, <2,and the LBP is of maximization, so we move the line FiQs from P,Q, to P,Q; parallel to itself as far 00) within the feasible region OABCDO is touched b} displacement ofthe line takes it out of the feasible region OABCDO Thus, the point B (1000, 500) is obtained on solving the equations xy +x, =1500and 1 + as possible, until the farthest point B (1000, yy the line. Any further 2000 , gives the maximum value of the objective function Z, and the line P,Q, is the line of maximum objective function. Hence, the optimal solution is x = 1000, x, =500 and the optimal value of Z = 3x1000+5x500=5500 Tyre 6. Basen on [SO-Prorit MeTHod: MinimizaTiON ProBleM graphically. Minimise z = 1.5x, +2.5x_ subject to the constraints x1 + 3x2 23 x, +x222 and x1 20, x2 20 bi 3 2 [KANPUR-2010) ‘SOLUTION.STEP 1. Changing all the given constraints into equi xy +3x) =3 xytx2=2 STEP 2. Draw these lines on x, x, -plane. Theline x, +3x Pes roush eae The line x, +x) = shown in figure. 2 passes through the points (2, 0) and (0, 2) which is 2 3. Since x, +3x, 23, x, +x,22and x, >0,x,20 The region common to both the re; %1+%2=2 in the first quadrani it the shaded region ABC. r ‘gions above the lines x, +3x,-3 and xy +3x2=3 amt the feasible region as shown in figure bY ~~STEP 4. Notations| Hine @ = x, + 3x; Iso-profit Method (a) Assign a consta ie function Z, we a) Now draw this i nown in figure by AQ (b) Again assign function Z, we get aline (2) Now draw this line (2) on x, xz-plane which is showa in Fig. 1-16 by ,Q2 Since, Z,>Z, so we move the line RQ; FQ, to P,Q, parallel to itself, until a point B of the feasible region ABC is touched by the line The point B tothe intersection ofthe Lines =, +x, =Zand x,+3%,=3, so onsolving these of Bas G 2 } ic co-ordinates equations, we obtain the co 35 Thus, the point (5 3 wi give the a ae ofZ. the optimum solution of LPP is 2=3,*2"and the optimum value of a a. jon Z=15x2+25x5=35 ‘function 2=15x5+25%>OPERATIONS RESEARCH (El Tree 7. Prosuns avn Unsure SOuToxs ing LPP ly the following EXAMPLE! Solve graphically tne ees subject to the constraints xy +%221, 2x, + x251, 4x1 - 2x2 <1, oe 5 g graph by See ec ene permissible region is shown in the following graph by SOLUTION. Proceeding as usual, 3 x shaded region. Here, Z = 4x, + 5x2 => a From the above figure, it is clear that dotted line through the origin representing Z =O can be moved parallel to itself in the direction of Z increasing and still has some point in the permissible region. Also, Z = 4x, + Sx) = 0 => ~L— 3 a. Hence, Z can be made arbitrarily large and so the problem has no fi value of Z. Thus problem has unbounded solution, EXAMPLE2. Solve the following LPP graphically Maximize Z = 0.75x, + x2 ‘subject to the constraints inite maximum X1~ X22 0,~ 0.x, + X2 <1 and x, x,>0 SOUITION. Proceeding as usual, we get the permissible region which shaded in the following figure. Hence, Z = 0 =, Xi __4 cesNotations] line x, -x) =0 line @ :-0.5x; +x) = 1 Fig. 18 From the above graph, it is clear that the dotted line Passes through the origin Tet ait REZ = 9 can be moved parallel to itself in the direction of Z increasing and will have some points in the feasible region. Therefore, Z can be made arbitrarily large and hence the problem has no finite maximum value of Z. Thus the problem has unbounded solution, Solve the following LPP graphically: Minimize Z ='5x, — 2x» subject to the constraints 2x, + 3x2 >1 and Xp x2 20 Proceeding as usual, we get the permissible region, which is shaded in the figure given below. The permissible region is the set of all the constraints and the non-negative restrictions, is unbounded. Solving simultaneously the equations of the 1 (4 corresponding intersecting lines we get two vertices Al 570} and (05)Fig. 22 2 ; Te ona amet an = feasible solution space $e unique s ues ee ae Fes all the constraints. therefore, there is no feasible variables x, and x2 t ion of this problem. .g LPP graphically. +2 soluti EXAMPLES. Solve the followin: Minimize Z = * subject to the constraints 1 x +m 6 tea Bx, +223 and xp x220 IUTON. Changing all the constraints to equations and then draw these lines on 22 plane by following the usual procedure. 2) line 3x, + xp = 3 passes through the point (1, 0) and (0, 3) - , 3). 1 Clearly, the line x; +5 = passes through the point ( i 0) and (04) 4 the 2 Now, since we have < My +X2 S53, +x, 23 and ak ind x1 20, x» 20, the region below issible region and the region ab, ove the line the line x; + xpquadrant. The solution = ax + 3 is the permissible region in the positive + Os = 16 spaces are shaded in the above figure, which shows that there is no point satisfying both the constraints. Hence the given LPP has infeasible solution. EXAMPLE4. Solve the following LPP graphically. Minimise Z = 5x; + 2X2 22 subject to the constraints xy tans2 Sx, + Bx, > 12 and Xp x22 0 and draw them on xyxq-plane. Clearly values gj SOLUTION. Changing all the constraints to equations (2, 0) and (0, 2) and the line no feasible the line x, + x) = 2 passes through the point 3x, + Bx = 12 passes through the point (4, 0) and (0 4). x Fig. 24 nes on XyXy_ Meat Pea x Clearly, the region below the line x, + x = 2 and above the line Sxj + the permissible region. The solution spaces are sha¢ ded in the above fig there is no common part to both the region, so there is no p constraints. Hence, the given LPP has infeasible solutions. EXAMPLES. Solve the following LPP graphically. Minimize Z = 3x, + 9X2 subject to the constraints x1 + 3x < 60, x1 + X22 10, x1 0 Converting the given constraints into equations, then plot these lines 01 mn 4X2" plane as shown below:tine@ | tine © Point A | point B=B15,15) | point C=C (0, 20) | point D =D (0, 10) Fig. 25 Se qs 20 25 30 35 40 45 50 55 60 85 *1 jos 1 Gxtiemepoint | Coordinates | Value of the ‘objective function p X2) Z= 3x; + 9X2 A (555), 3x5+9x5=60 B as, 15) 3x15+9x 15 = 180 c (0, 20) 3x0+9 x 20 = 180 D (0, 10) 3x0+9x10=90 Here, itis dear that Z is maximum at BIS, 15) and C(O, 20) and its maximum value is 180. Hence, the given LPP has multiple optimal solutions at B and ¢ Tree 4. More Posies on Moco ConsTRAINTS ‘A redundant constraint is one that does not affect the feasible solution 5 i XA a space and thus redundancy of any constraint does not cause any difficulty in solving LPP graphically 25 EXAMPLEL Solve the following LPP graphically. _ Minimize Z = x1 + x2 subject to the constraints 5x, + 10x, < 50 xy +x221 x24 and Xp x2 20 Converting the given constr ai aints into equations and drawing these lines, we haveLPP: FORMULATION AND GRAPHICAL SOLUTIONS ax Clearly, the permissible region is ABCDEOA (shaded region) which is the se feasible solutions of the given LPP Taking Z = x1 + *: Brine corresponding to z= 01s shown by dorted fink © 9 Drawing lines parallel to this line, we sec that @ through the vertices A(1, 0) and £(0, 1) of the nearest to the origin O- = Every point of the line segment AE gives the o Hence, the given LPP has infinite number of solut . Solve the following LPP graphically. Maximize zoxthx subject to the constraints 3x, + 2x25 12 x,<2 x, +x228 ax, +%224 and Xy 220 [MEERUT-2006] Proceeding stepwise as usual, we observe that there is no permissible region of the problem, whose points satisfy all the constraints, ie, there is no Point which will satisfy all the constraints.BAMPLES Solve the following LPP graphically. ‘Maximize Z = 7x; + 3x2 subject to the constraints ouTO 2X1 + 2xq > 3, x1 + x2 <4, X1 < 5/2, Xp < 3/2 and xy, x22 ‘Converting the given constraints into equations and then plot them by following the usual procedure and then use the inequality conditions to mark the feasible region as shown in the following figure. Beaaes 2) Point C=C/ 0,ORMULATION AND GRAPHICAL SOLUTIONS ll Here, the extreme points of the feasible region are A, B and C. Now, we compute the value of the objective function at each of these extreme points. Fatreme point | Coordinates | Value of the objective function (xy %2) Z = 7x, + 3x2 3) [pded-? 9) less c (02) 0x74+3x 2 Glearly, the maximum value of the objective function Z = 22 occurs extreme point (3.2) Hence, the optimal solution to the given LPF x $ and Max Z 22. Solve the following LPP graphically. Minimize Z = 5x, + 6x2 subject to the constraints xy + X22 50, x14 2x> <4 3x, + 4x2 < 100 and x1 20; x2 20 ‘Converting the given constraints into equation and then plo} lure and then use the inequality conditions to mark shown in the following figure. 10 20 30 “40> ‘there exists no value of x, and x, simultaneously satisfy all ‘and the non-negative restrictions, Hence, the given problem does ile solution.‘AL SOLUTIONS eal TION AND GRAPHIC a pp: FORMULA qotal 3(3x, + 5x) - 100%) + 2) S xy + 5x2 ce Also, 20x; + 50%, < 800 ve and x1 +220 me The mathematical model of an LPP is given by “ Max. Z = -%1 + 5X2 ott subject to the constraints a 20x, + 50x, < 800 => 2x, + Sx, < 80 08 xy +X) $20 xy x22 0 Following the usual procedure, we draw the following graph Fig. 31 2 a 2 32 36 40 ies 8 12 (16 2 Solving simultaneously, the equations of the corresponding intersecting lines of the feasible region, we get the coordinates of the vertices of the feasible region is 0(0, 0), A(O, 16), B(20, 0) and D (2.4) corner points as given below: The value of the objective function at these ‘ive function Extreme point | Coordinates | Value of the object (xy x2) z= "xy + 5¥ (0, 0) -04+5x0=0 (0, 16) “1x0 +5 x 16 = 80 (ma) (20, 0) ax 20+0%5= 20 (33) Ee the maximum value of Z is 80 at the point x, = 0x2 = 16and B. It produces (eal pottling machines A 1 ; EXAMPLES. drink Pier pe as16 ‘ounce bottles. The following data is il e ————— | 100/minute 40/minute 60/minute 75/minute The machines can be run 8 hours per day, 5 days per week. Weekly a production of the ‘drinks cannot exceed 300000 ounces and the vince botties andazano sixteen market can absorb 25000 eight 01 ounce bottles per week. Profit on these bottles is 15 paise and 25 paise per bottle respectively. The planner wishes to maximize his profit subject to all the production and marketing restrictions. Formulate the problem as an LPP and then solve graphically. {et the planner produce x; and x2 num d 16 ounces respectively per week. Then his profit in @ is Z = 0.15x, + 0.25x2 otal production of soft drink per week to fill up these bottles = 8x, + 16x2 ‘Since, the production cannot exceed 300000 ounces. ‘Therefore, 8x, + 16x < 300000 otal time taken to produce these bottles on machine A x x iraeo" 40x60 °° bers of bottles of 8 an ‘On machine B = —~! X2 4 0x60 "75x60 "OS tmachines can run for 8 x 5 = 40 hours per week x ae <40 and 1 2 60x60 "75x60" *° “xy + 5x2 < 480000 and 5x, + 4x» < 720000 x1 < 25000 and x, < 7000 x, X2 20 tical model of the given LPP as follows: >——Ee_—_ % cal Nees LPP: FORMULATION AND GRAPHICAL SOLUTIONS onding intersecting lines, Solving simultaneously, the equations of the corresponding inksrso eM algy: the coordinates of the vertices of the convex polygon are (25000, 6250), R(23500, 7000), S(0, 7000) Now, we draw dotted lines through the origin corresponding to Z = 0. Drawing {ines parallel to this line away from the origin we see that the farthest line for maximum Z in the permissible region passes through the vertex Q(25000, 6250) of the convex polygon which is the point of intersection of the lines 8x, + 16x) = 300000 and x; = 25000. Hence, the optimal solution is given by x1 = 25000, x» = 6250 and Max. Z = 0.15 x 25000 + 0.25 x 6250 = 7 5312.50 We can also solve this LPP by corner point method as follows: |Coordinates Value of the objective function Gp x2) Z = 0.15x + 0.25x2 0) [015x0+025x0=0 (25000, 0) | 0.15 x 25000 + 0.25 x 0 = 3750 (25000, 6250)| 0.15 x 25000 + 0.25 x 6250=5312.50(max) (23500, 7000)| 0.15 x 23500 + 0.25 x 7000 = 5275 (0, 7000) _| 0 + 0.25 x 7000 = 1750