| Decision Theory and Decision Trees The one word that makes a good manager ~ decisiveness.” - lacocea, Lee PREVIEW In decision theory a set of techniques ate used for making decisions in the decision-environment of ‘uncertainty and risk Decision tree graphically displays the progression of decision and random events, in those cases ‘where a problem involves @ sequence of decisions (including a decision on whether to obtain additional information). Utiities theory helps to incorporate decision-makers! attitude towards risk into the analysis of those problems where there is possibilty of large losses. LEARNING OBJECTIVES ‘After studying this chapter, you should be able to ‘© understand the steps of decision-making process, ‘¢ make decision under various decision-making environments. ‘© determine the expected value of perfect information, expect opportunity loss and expected monetary value associated with any decision, ‘revise probability estimates using Bayesian analysis ‘© construct decision trees for making decision, ‘* understand the importance of utlty theory in decision-making CHAPTER OUTLINE 4144 Introduction + Self Practice Problems 8 41.2. Steps of Decision-Making Process + Hints and Answers 11.3 Types of Decision-Making Environments U7 Decision Troo Analysis 11.4. Decision-making Under Uncertainty 41.8 Decision-Making with Utilities + Conceptual Questions A + Self Practice Problems C + Self Practice Problems A + Hints and Answers + Hints and Answers Chapter Summary 115. Decision-Making Under Risk 1B Chapter Concepts Quiz 416 Posterior Probabilities and Bayesian Analysis | 0 Case Study ‘© Conceptual Questions 8‘Operations Research: Theory and Applications Fig. 11.4 Zones of Decision-making Decision aralysis is an analytical approach of comparing decision attematives in tors of expected ‘outcomes. States of nature fare outoomes due to random factors that effect the payot! of from Secision_altematives, Table 11.1 General Form of Payoff Matrix 44.1 INTRODUCTION “The suecess or failure that an individual or organization experiences, depends toa large extent, on the ability of making acceptable decisions on time. To arrive at such a decision, a decision-maker needs to enumerate feasible and viable courses of action (alternatives or strategies), the projection of consequences associated swith each course of action, and a measure of effectiveness (or an objective) to identify the best course of action, Decision theory is both descriptive and prescriptive business modeling approach to classify the degree of knowledge and compare expected outcomes due to several courses of action. The degree of knowledge is divided into four categories: complete knowledge (i.e. certainty), ignorance, visk and uncertainty as shown in Fig. H.1 ‘paar [Dewan [eon [oa ] Increasing rowledge as Irrespective of the type of decision model, following essential components are common to all: Decision alternatives ‘There is « finite number of decision alternatives available to the decision-maker ‘at each point in time when a decision is made. The number and type of such altematives may depend on the previous decisions made and their outcomes, Decision alternatives may be described numerically, such as stocking 100 units of a particular item, or non-numerically, such as conducting a market survey to know the likely demand of an item, States of nature A state of nature is an event or scenario that is not under the control of d For instance, it may be the state of economy (e.g. inflation), a weather condition, a politi jsion makers. development, etc. ‘The states of nature may be identified through Scenario Analysis where a section of people are interviewed ~ stakeholders, long-time managers, etc., to understand states of nature that may have serious impact on a decision. The states of nature are mutually exclusive and collectively exhaustive with respect to any decision problem. The states of nature may be described numerically such as, demand of 100 units of an item or non-numerically such as, employees strike, ete. Payoff it is « numerical value (outcome) obtained due to the application of each possible combination of decision alternatives and states of nature, The payofT values are always conditional values because of unknown states of nature. ‘The payoff values are measured within a specified period (e.g. within one year, month, etc.) called the decision horizon. The payofts in most decisions are monetary. Payofts resulting from cach possible Combination of decision altematives and states of natures are displayed in & matrix (also called payoff ‘air form as shown in Table 11-1 41.2 STEPS OF DECISION-MAKING PROCESS The decision-making, process involves the following steps 1 2 Identify and define the problem. List all possible future events (not under the control of decision-maker) that are likely to occurDecision Theory and Decision Trees 3, Identify all the courses af action available to the decision-maker 4. Express the payoffs (p,) resulting from each combination of course of action and state of nature Apply an appropriate decision theory model to select the best course of action from the given lst on the basis of a criterion (measure of effectiveness) to get optimal (desired) payoft Example 11.1 A firm manufactures three types of products. The fixed and variable costs are given below: Pied Cost (Rs) Variable Cost per Unit (Rs) Produet 4 25,000 12 Product B 35,000 9 Product 3,000 7 The likely demand (units) of the products is given below: Poor demand 3,000 Moderate demand 7,000 High demand 11,000 Ifthe sale price of each type of product is Rs 25, then prepare the payoff matrix. Solution LetD,,D, and D, be the poor, moderate and high demand, respectively. The payofTis determined as: Payoff = Sales revenue ~ Cost we calculations for payoff (in °000 Rs) for each pair of alternative demand (course of action) and the types of product (state of nature) are shown below: D\A=3%25-25-3%12 = 1 DA =7%25~25-7% 12 =66 D,B=3%25-35-3% 9 = 3 DB =7"25-35-7% 9 =77 D.C =3x25~ 53-3 7 1 D,C=7%25-53-7% 7=73 DjA = 11 *25—25~11 «12 = 118. . D,B = 1125-35-11 x D,C = 11 25-53-11* 7 The payoff values are shown in Table 11.2, 9= MI 45 4 4 66 118 B B 71 14) ic 1 B 4s 11.3. TYPES OF DECISION-MAKING ENVIRONMENTS To arrive at an optimal decision it is essential to have an exhaustive list of decision-alternatives, knowledge of decision environment, and use of appropriate quantitative approach for decision-making. In this section three types of decision-making environments: certainty, uncertainty, and risk, have been discussed. The knowledge of these environments helps in choosing the quantitative approach for decision-making Type 4 Decision-Making under Certainty is decision-making environment, decision-maker has complete knowledge (perfect information) of outcome due to each decision-altemnative (course of action). In such a case he would select a decision ‘alternative that yields the maximum return (payoff) under known state of nature. For example, the decision to invest in National Saving Certificate, Indira Vikas Patra, Public Provident Fund, ee., is where complete information about the future return due and the principal at maturity is know. Type 2 Decision-Making under Risk In this decision-environment, decision-maker does not have perfect knowledge about possible outcome of every decision alternative. It may be due to more than one states of nature. In a such a case he makes an assumption of the probability for oceurrence of particular state of nature Payoff is the quantitative measure of the outcome trom each pat of decision altematve ‘and a state of natue. Table 11.2 Decision making lunder cortainty 1s fan environment in whieh future ‘outcomes or slates of nature are known.Operations Research: Theory and Applications Decision making tunder risk is. an environment in ‘which the probably ‘of auteomes or states of ature can be quenttied Decision making under uncertainty is an environment in which the probebity of outcomes oF sales of nature can hot be quantified ‘Type 3 Decision-Making under Uncertainty In this decision environment, decision-maker is unable to specify the probability for occurrence of particular state of nature. However, this is not the case of decision-making under ignorance, because the possible states of nature are known. Thus, decisions under uncertainty are taken even with less information than decisions under tisk. For example, the probability that Mr X will be the prime minister of the country 15 years from now is not known. 41.4 DECISION-MAKING UNDER UNCERTAINTY When probability of any outcome can not be quantified, the decision-maker must arrive at a decision only ‘on the actual conditional payof¥ values, keeping in view the criterion of effectiveness (policy). The following criteria of decision-making under uncertainty have been discussed in this section, (i). Optimism (Maximax or Minimin criterion (ii) Pessimism (Maximin or Minimax) eriterion (ii). Equal probabilities (Laplace) criterion iv) Coefficient of optimism (Hurwiez) criterion (W)_ Regret (salvage) criterion 14.4.1. Optimism (Maximax or Minimin) Criterion In this criterion the decision-maker ensures that he should not miss the opportunity to achieve the largest possible profit (maximax) or the lowest possible cost (minimin). Thus, he selects the decision alternative that represents the maximurn of the maxima (or minimum ofthe minima) payofTs (consequences or outcomes), The working method is summarized as follows: (a) Locate the maximum (or minimum) payoff values corresponding to each decision alternative. (b) Select a decision alternative with best payoff value (maximum for profit and minimum for cost). Since in this criterion the decision-maker selects an decision-alternative with largest (or lowest) possible payoff value, it is also called an optimistic decision criterion, 41.4.2. Pessimism (Maximin or Minimax) Crit In this criterion the decision-maker ensures that he would earn no less (or pay no more) than some specific ‘amount. Thus, he selects the decision alternative that represents the maximum of the minima (or minimum of the minima in case of loss) payoff in case of profits. The working method is summarized as follows: (a) Locate the minimum (or maximum in ease of profit) payoff value in case of loss (or cost) data corresponding to each decision alternative. (b) Select a decision alternative with the best payoff value (maximum for profit and mimimum for toss or cost). ‘Since in this criterion the decision-maker is conservative about the future and always anticipates the worst possible outcome (minimum for profit and maximum for cost or loss), it is called a pessimistic decision criterion, This eriterion is also known as Wald’ criterion. 11.4.3 Equal Probabilities (Laplace) Criterion Since the probabilities of states of nature are not known, it is assumed that all states of nature will occur with equal probability, ie. each state of nature is assigned an equal probability. As states of nature are mutually exclusive and collectively exhaustive, so the probability of cach of these must be: I/(number of slates of nature). The working method is summarized as fotlows: (a) Assign equal probability value to each state of nature by using the formula: 1 + (number of states of nature). (b) Compute the expected (or average) payot for each alternative (course of action) by adding all the payofts and dividing by the number of possible states of nature, or by applying the formula: (Probability of state of nature 7) * (Payoff value for the combination of alternative i and state of nature j.) (6) Select the best expected payoff value (maximum for profit and minimum for cost),Decision Theory and Decision Trees This criterion is also known as the criterion of insufficient reason, This is because except in a few cases, ome information of the likelihood of occurrence of states of nature is available, 11.4.4 Coeffi This criterion suggests that a decision-maker should be neither completely optimistic nor pessimistic and, herefore, must display a mixture of both. Hurwicz, who suggested this criterion, introduced the idea of a ficient of optimism (denoted by @) to measure the decision-maker’s degree of optimism. This coefficient lies between 0 and 1, where 0 represents a complete pessimistic attitude about the future and | a complete ptimistic attitude about the future. Thus, if is ahe.coefficient of optimism, then (1 ~ a) will represent the coefficient of pessimism. “The Hurwicz approach suggests that the decision-maker must select an alternative that maximizes nt of Optimism (Hurwicz) Criterion H (Criterion of realism) = 0 (Maximum in column) + (J ~ 0.) (Wininum in column) The working method is summarized as follows: (a) Decide the coefficient of optimism 0: (alpha) and then coefficient of pessimism (1 ~ 0. (b) For each decision alternative select the largest and lowest payoff value and multiply these with o- ‘and (1 ~ @) values, respectively. Then calculate the weighted average, H by using above formula. (©) Select an alternative with best weighted average payoff value. 11.4.5 Regret (Savage) Criterion ‘This criterion is also known as opportunity loss decision criterion or minimax regret decision criterion because decision-maker regrets for choosing wrong decision alternative resulting in an opportunity loss ‘of payoff. Thus, he always intends to minimize this regret. The working method is summarized as follows: (a) From the given payoff matrix, develop an opportunity-loss (or regret) matrix as follows: (Find the best payoff corresponding to each state of nature Gi) Subtract all other payoff values in that row from this value. (&) For each decision alternative identify the worst (or maximum regret) payoff value, Record this value in the new row. (©) Select a decision alternative resulting in a smallest anticipated opportunity-loss valve. Example 11.2 A food products’ company is contemplating the introduction of a revolutionary new product with new packaging or replacing the existing product at much higher price (S,). It may even make & moderate change in the composition of the existing product, with a new packaging at a small inerease in price (S,),or may mall a small change inthe composition of the existing product, backing it withthe word 'New* and a negligible increase in priee (S,). The thtee possible states of nature or events are: (i) high increase in sales (N,) (ji) no change in sales (V,) and (if) decrease in sales (N,). The marketing department of the company worked out the payoffs in terms of yearly net profits for each of the strategies of three events (expected sales). This is represented in the following table: Siates of Nature Sono 30nom 30808 s, | Which strategy should the concerned executive choose on the basis of (@) Maximin criterion (©) Minimax regret criterion (6) Maximax criterion (@ Laplace criterion? Regret criterion attempts 10 minimize the maximum opportunity. loss.Operations Research: Theory and Applications Solution (@) Maximin Criterion The payofT matrix is rewritten as follows: Nature S Sy 7,00,000 5,00,000 3,00,000 N 3,00,000 4,50,000 300,000, ”, 1,50,000 ° 3,00,000 Column (minimum) 1,50,000 ° 3,00,000<— Maximin Payoff The maximum of column minima is 3,00,000. Henee, the company should adopt strategy S;, (b) Maximax Criterion Strategies States of Nawre 5, s, Sy oF 7,00,000 500,000 3.00,000 My 3,00,000 4,50,000 3,00,000 ™, 1,530,000, o Column (maximum) 700,000 100,000 ‘The maximum of column maxi L Maximax Pavol (©) Minimax Regret Criterion Opportunity loss table is shown below: Staves of Strategies a is 7,00,000. Hence, the company should adopt strategy S}, Nature S, Ss Ny 7,00,000 - 7,00,000 _7,00,000 - 5,00,000_7,00,000 ~ 3,00,000 o 2,00,000 = 4,00,000 4,50,000 — 3,00,000—_-4,50,000 ~ 4,50,000 _4,50,000 ~ 3,00,000 50,000 0 = 1,50,000 300,000 ~ 150,000 3,00,000 ~ 0 3,00,000 ~ 3,00,000 = 1,50,000 = 3,00,000 =0 Column (maximam) 1,50,000 300,000, 4,00,000 Pe Minimax Regret Hence the company should adopt minimum opportunity loss strategy, 5}. (@) Laplace Criterion Assuming that each state of nature has a probability 1/3 of occurrence. Thus, Strategy Expected Renun (Rs) 5, (700,000 + 3,00,000 + 1,50,000)/3 = 3,83,333.33 = Largest Payoff 5S, 00,000 ~ 4,50,000 + 0)/3 316,666.66 sy (,00,000 = 3,00,000 + 3,00,000)/3 = 3,00,000 sir the largest expected return is from strategy S}, the executive must select strategy 5. Example 11.3 A manufacturer manufactures a product, of which the principal ingredient is a chemical X. At the moment, the manufacturer spends Rs 1,000 per year on supply of X, but there is a possibility that the price may soon inerease to four times its present figure because of a worldwide shortage of theDecision Theory and Decision Trees chemical. There is another chemical Y, which the manufacturer could use in conjunction witha third chemical Z, in order to give the same effect as chemical X, Chemicals ¥ and Z would together cost the manufacturer Rs 3,000 per year, but their prices are unlikely to rise. What action should the manufacturer take? Apply the maximin and minimax criteria for decision-making and give two sets of solutions. Ifthe coefficient of ‘optimism is 0.4, then find the course of action that minimizes the cost Solution The data of the problem is summarized in the following table (negative figures in the table represents profit) States of Nature Courses of Action Sy(use ¥ and Z) Sy(use X) 1M, (Price of X increases) 3,000 ~ 4,000 N, (Price of X does not increase) 3,000 = 1,000 (i) Maximin Criterion Suates of Nature Courses of Action y, 3,000 ~ 4,000 Ny = 3,000 1,000 Columa (minimum) 3,000 ~ 4,000 1 Maximin Payoff Maximum of column minima 3,000. Hence, the manufacturer should adopt action S, (ii) Minimax (or opportunity loss) Criterion States of Nature Courses of Acton ~s, s, 3,000 ~ (3,000) = 0 3,000 ~ (4,000) h 1,000 ~ (3,000) = 2,000 1,000 = (— 1,000) Maximum opportunity 2,000 1,000 — Minimax Payoff Hence, manufacturer should adopt minimum opportunity loss course of action Sy. iil) Hurwice Criterion Given the coefficient of optimism equal to 0.4, the coefficient of pessimism will be 1 0.4 = 0.6. Then according to Hurwicz, select course of action that optimizes (maximum for profit and minimum for cost) the payoff value Hf = @& (Best payoff) + (1 ~ a) (Worst payoft) = a (Maximum in column) + (I~ c) (Minimum in column) volt Worst Payetf H 5, = 3,000 s; 1,000 Course of Action Since course of action S, has the least cost (maximum profit) = 0.4(1,000) + 0,6(4,000) = Rs 2,800, the nanufacturer should adopt this. Example 11.4 An investor is given the following investment altematives and percentage rates of return. ‘States of Nature (Market Conditions) Low Medium High ‘Regular shares 10% 15% Risky shares 1% 25% Property 18%Operations Research: Theory and Applications ‘Over the past 300 days, 150 days have been medium market conditions and 60 days have had high market inereases. On the basis of these data, state the optimum investment strategy for the investment. [Nagpur Univ, MBA, 1999) Solution According to the given information, the probabilities of low, medium and high market conditions ‘Would be 90/300 or 0.30, 150/300 or 0.50 and 60/300 or 0.20, respectively. The expected pay-offs for each ofthe alternatives are shown below: Sirateey Market Conditions Probability Regular Shares Risky Shaves Property Low 030 007%030 0.10030 0.15 «030 Medium 0.50 010050 0.12050 0.25*0.50 High 020 0.12%020 0.18020 0.30 0.20 Expected Return 0.136 0.126 0230 Since the expected return of 23 per cent s the highest for property the investor should invest in this alternative, CONCEPTUAL QUESTIONS A 4. What is @ solentiio decision-making process? Discuss the role of the statistical mathod in such a process. 55. Give an example of a good decision that you made, which ‘esuted in a bad outcome. Also give an example of a good ‘cision that you made and that had a good outcome. Why ‘was each decision good or bad? 1. Discuss the ltference between decision-making under certainty, under uncertanty and under risk 2. What techniquoe aro used to solve decision-making problems Under uncertainty? Which technique resuts in an optimistic | cision? Which technique results in a possimistic decision? ‘8. Explain the various quantitative methods that are useful for decision-making under uncertainty SELF PRACTICE PROBLEMS A 4, The folowing matrix gives the payott (in Rs) of ferent strategies Strategy State of Nature (altematives) §,, Sp and S, against conditions (events) N,N - ————— Nyand My My Ne i State of Nature 3S a 60 Strategy Ny % % Ne S 10 -20 s 40 150 5 4000-100 6,000 18,000, s 20,000 5,000 400 ° Select a strategy using each ofthe following docsion citeria: (a) s 20,000 15,000 © -2,000 1,000 Maximax, (0) Minimax regret, (c) Maximin, (J) Minimum tsk, ‘assuming equiprobable slates. indleate the decision taken under the following appraaches: (| 4, y4r Sethi has Ais 10,000 to invest in one of three options: A, B Possimiste, (i) Optmiste, (i) Equal probability, w) Regret. ”) |“ Oy G-the Tetum on his investment depends on whether the Hurwicz eriterion, the degree of optimism being 0,7, ftconomy experiences inflation, recossion, or no change at all Ina toy manufacturing company, suppose the product acceptance ‘The possible returns under each economic condition are given probablties are not known bul the folowing data is known: below: “Antpated First Year Profit (000 As) ‘State of Nature product Product Line | _Suategy | _totaton Recession No Change eceptance | Ful arial | a 2,000 200 7500 Aver fa a = 5 3,000 800 4000 Good 8 70 50 ¢ 2.500 1,000 4800 Fair a Pow | 28 ‘0 ° rat soul he dete, using the pessmiste creron,opbmste | efteion, equaly Brey citron and regret stein? Determine the optimal decision under each of the folowing | 5, a manulacuer’s representative has been offered a now product decision criteria’ and show how you arrived: at_it Tine. the accepts the new line ho can handle it in. one of the {@) Meximax, (B) Maximin, (¢) Equal liketnood and (4) Minimax two ways. Tho best way according to the manuiacturer would regrel? be to set a separate sales force 1D exclusively handle the now 3, The following is @ payol (In rupees) table for three strategies line. This would involve an ital Investment of FAs 1,00,000 i and two states of nature: ihe olfce, office equipment and the hiring and training of theDecision Theory and Decision Trees aleemen, On tho ater hand, tthe naw Hino ts handled by te | Teo and their experience with other products, the fxisting sales. fort earaent would only be Fs 80,000, principally fr raining is present salesmen. ae now product sells for As 250, The representative normally roceives 20 por cf of the sales pice on each unt Probabilty may rich 10 per cent is pakd as commission to hance he sete Soult. The manufacturer offers to pay 60 par cent of We (a) Set up a regret table. naw ge of each Unit sokd to the representative, Me oat stortaive sets UP a separate slos organization, Ot ewce Tee eeemal 20 per cont wil be paid. n eter case the salesmen we at0 por cont commision. Based on the am ef: she Sales of the now product rmaximin eiterion? HINTS AND ANSWERS ‘ales (in units) : 1,000 2,000 3,000 010 015 040 {b)_ Find the expected regret of each coy (P) mieh eourse of acon would have been best under the ‘ingesting ‘acs, the nial | presents Galmates the folowing probabilties for annual 4,000 §,000 030 005 yr8@ of action, LOS, SoS GD Sy GIS. OS Tecfoe, pay fnctioncoresponding 10 5, and Sa would yin) Minis, (2) Full o para (Patil | be Fe Tad.) 5p READ OSE REA @SHRESS |e 10,000 + 250 (G0 Joy} = « =—1,00,000 + SO 3 55 RI ign Choose B: Re 300, Choose CRS IT6S, | Sa = 250°» {20 ~ 10V100) «a = ~30,000 + 250 Choose C: Rs 50 to inet ono, we get~1,0,000 + S0a-= ~30,000 # 250 5, Let §;= instal new sls faites or a= 2800 ‘5, = continue with existing sales facilis UNDER RISK decision-making environment, decision-maker has sufficient infants assign probability to the repr each outcome (atte of nature). Knowing the probability distribution of outcomes (Gates of nature), the decision-maker needs to select & form ‘of action resulting a largest expected (average) payolT value. The expected payoft is the sum of nil possible weighted payofTs resulting from choosing @ decision alternative. "The widely used criterion for evaluating decision altematives (cousss of action) under risk is the Expected Monetary Value (EMV) ot Expected Urtlioy 41.5.1 Expected Monetary Value (EMV) “The expected monetary value (EMV) fora given course of actions obtained by adding payoff values multiplied by the probabilities associated with each state of nature ‘Mathematically, EMV is stated as follows: EM (Couse of actin. §) = $, uP where m= number of possible states of nature probability of occurrence of state of nature, N; Pi py = payotl associated with state of nature N, and course of action, The Procedure x Construct a payot? matrix Fisting all possible courses of action and #aiee of nature. Enter the const pay values associated with each possible combination of owns of action and state of saree tong with the probabilities of the oocurTence of cach ste of Ne aan aoe EMV for each course of action by multiplying the condons! jpayofts by the associated Probables and adding these weighted valucs for each course of action 3, Soleet the course of action that yields the optimal EMV. Example 11.5 Mr X fis quite often from town A t0 town B. He a the airport bus which costs Pease tie takes it, there is # 0.08 chance that he will miss the Tish The stay in a hotel costs Ree ne igG hance of being on time for the slight. For Rs 350 be can 1 taxi which will Take 99 per cent chance of being on time forthe fight If Mr catches the plane on time, he will rae oe et ceiness transaction that will produce a profit of Rs, 10,000, fotherwise he will lose it Fron ge of ranspart should Mr X use? Answer on the basis ofthe EMY erteion Expected monetary value i obtained by adding payotis for teach course of feefon, mutiptod by the probabilties associated with ach stale of ature,Operations Research: Theory and Applications Table 11.3 Table 11.4 Conditional Profit Value (Payofts) Solution Computation of EMV associated with various courses of action is shown in Table 11.3. Caches ]1o000-- 25092 9,177 10.000 270.96 9,340.80 10000 - 350. 099 9,553.50 the tim — | = 9.975 = 9780 9650 Mist te Mig] 25 008-20 -270 004-080-350 om 350 Expected moncary ats 9,330 9.550 vale EMV) Since EMV associated with course of action “Taxi is largest (= Rs 9,550), it isthe logical alternative. Example 11.6 The manager of a flower shop promises its customers delivery within four hours on all lower orders. All flowers are purchased on the previous day and delivered to Parker by 8.00 am the next morning. The daily demand for roses is as follows, Dozens of roses: 70 80 90 100 Probability ol 02 04 03 ‘The manager purchases roses for Rs 10 per dozen and sells them for Rs 30. All unsold roses are donated {0 @ local hospital. How many dozens of roses should Parker order each evening to maximize its profits? ‘What is the optimum expected profit? [Delhi Univ, MBA, Dec. 2004) Solution The quantity of roses to be purchased per day is considered as ‘course of action’ and the daily demand of the roses is considered as a ‘state of nature’ because demand is uncertain with known probability. From the data, it is clear that the flower shop must not purchase less than 7 or ‘more than 10 dozen roses, per day. Also each dozen roses sold within a day yields a profit of Rs (30 ~ 10) = Rs 20 and otherwise it is a loss of Rs 10, Thus Marginal profit (MP) = Selling price - Cost = 30 - 10 ~ Rs 20 Marginal loss (ML) = Loss on unsold roses = Rs 10 Using the information given in the problem, the various conditional profit (payoff) values for each combination of decision altematives and state of nature are given by Conditional profit = MP * Roses sold — ML * Roses not sold 20D, if Des 20D-10(S-D) = 30D-108, if D k) that exceeds the critical ratio, p= 0.428 is k= 3 units of cake, the optimal deeision is to prepare only 3 cakes. 11.6 POSTERIOR PROBABILITIES AND BAYESIAN ANALYSIS An initial probability statement to evaluate expected payoff is called a prior probabitiqy diswribution, but iF the probability statement has been revised due to additional information, then such 1 probability statement is called a posterior probability d sribution, In this section we will discuss the method of computing posterior probabilities, given prior probabilities using Baves’ theorem. The analysis of problems using posterior probabilities with new expected payols and additional information, is called prior-posterior analysis: Baye’s Theorem Statement pat Aiea dy be mutually exclusive and collectively exhaustive outcomes. Their probabilities P(4,), Pla)... P(4,) are known, There is an experiniental outcome B for which the conditional probabilities P(B\ Ay), PUB Aa), «»» PU |A,) are also known. Given the information that outcome B has occurred, the revised conditional probabilities of outcomes 4,, ie. PU, | B), t= 1.2 are determined by using the following relationship: )_ PA) PBA) PB) B)= where PB) = ¥ PUA) BLA) Since each joint probability can be expressed as the product of a known marginal (prior) and conditional probability, Le, PUA, OB) = PLA, )x PUBL, ) Example 11.19 A company is considering to introduce new product to its existing produet range. It has defined two levels of sales as ‘high’ and *low" on which it wants to hase its decision andl has estimated {he changes that each market level will occur, together with their costs and consequent profits or los This information is summarized below Sates of Mare Probabity Courses of Action Marker the Pro Do sot Marker the (R000) Produce (Rs “000 Wishes SS 150 0 Low sales 07 -40 f The company's marketing manager suggests that a market research survey may he unclertaken provide further information on which the company should base its decision, Based on the ewimpany's pastDecision Theory and Decision Trees. xperience with a certain market research organization, the marketing manager assesses its ability to give ‘90d information in the light of subsequent actual sales achievements, This information is given below Market Research Total Sales (Survey onteome) Market ‘high Market low os a Indecisive survey report 03 04 Low’ sales forecast 02 on ‘The market research survey costs Rs 20,000, state whether or not there is a case for employing the tarket research organization [Delhi Univ, MBA, 2000] olution The expected monetary value (EMV) for each course of action is shown in Table 11.29. Table 11.29 With no additional information, the company should choose course of action ‘market the product ‘owever, if the company had the perfect information about the ‘low sales’, then company would not go read with the decision because expected value would be (-) Rs 28,000. Thus, the value of perfect formation is the expected value of low sales. Let outcomes of the research survey be: high sales (S,), indecisive report (S,) and low sales (S,) and ates of nature be: high market (N,) and low market (N,). The calculations for prior probabilities of forecast ‘e shown in Table 11,30. High sales (S,) S|) = 05 AS ||) = 01 decisive report (5,) ASN) = 03 S,|N) = 04 Low sales (8) PAS, 1M) = 02 PS[ND = 05 Table 11.30 Jith additional information, the company can now revise the prior probabilities of outcomes to get posterior ‘obabilities, The calculations of the revised probabilities, given the sales forecast are shown in Table 11.31 ss op 03 s,m) = 05 ae P(N) = 03. 1S, 1) = 02 MS, 1Ny = 0.1 PS, | N,) = 04. AS, | Ny) = 05 neat ads Tab nto dai Probabilities(Operations Research: Theory and Applications ‘The posterior probabilities of actual sales, given the sales forecast, are: Outcome Probability ‘States of Nanre Posterior Probability (5) PS) (y) PAN, | S)= PIN, © SPS) Si 022 x 0.151022 = 0.681 N 0.070022 = 0.318 5 oa7 Ny, 0.09/0.37 = 0.243, Ny 0.28037 = 0.756 5 oat yy 0.061041 = 0.146 N; 035/041 = 0.853 Given the additional information, the revised probabilities to calculate net expected value with respect (0 each outcome are shown in Table 11.32. High sales 130 3159 0146 1898 Low sales =4536 0.853 51.18 Fapected value of sales forecast, a7 3220 Probabiliyy of occurrence 037 O41 ‘Net expected valve Table 11.32 ‘(Expected value x Probability) = 31095 13.202 CONCEPTUAL QUESTIONS B 1. Given the complete set of outcomes in @ certain situation, how | 4. Brlelly explain ‘expected value of perfect information’ with Is the EMV determined for a specie course of action? Explain examples. in your own words 5. Describe a business situation where a decision-maker faces 2, Explain tho dliferonce between expected opportunity loss and expected valve of perfect information 3, Indicate the difference between decision-making under sk, and Uncertainty, in staisteal decision theory. decision under uncertainty and where 2 decision based on ‘maximizing the expected monetary value cannot be made. HOW do you think the decision-maker should make the required ecision? SELF PRACTICE PROBLEMS B 1. You are given the folowing payots of three acts Ay, A, and Ay ‘and the events E,, E>, Ey States of Nature A A & 25 = 128 5 400 400 5 650 750 ‘The probabiites of the states of nature are 0.1, 0:7 and 0.2, respectively. Calculate and tabulate the EMV and conclude which would prove to be the best course of action, 2. The managernent of a company’ is faced with the problem of choosing one of three products that it wants to manufacture, ‘The potential demand for each product may turn out to be go0d,, moderate or poor. The probabilties for oach of the slates of ature wero estimated a5 follows Product Nature of Demand Good «Moderate Poor x 070 0.20 0.10 y 050 0.30 020 Zz 0.40 080 0.10 The estimated proft or loss in rupees under the three st may be taken as: Product Moderate Poor x "20,000 10,000 y 20,000 20,000 Zz 40,000 15,000 Prepare the expected value table, and advise the management about the eboice of product4. The marketing staff of a certain industrial organization has submited the folowing payot! table (giving profs in mition rupees). conceming a certain proposal depending upon the rate | of technology advance Technological Decision ‘Advance econ Pact Much 2 3 Lite 5 2 None or 4 “The probabilities ara 02, 0.5 and 0.3 for Much, Ue and None technological advance respectively. What decision should be taken? 4. A physician purchases @ particular vaccine on Monday each week. The vaccine must be used within the following week, ‘otherwise it becomes worthless. The vaccine costs Fs 2 por ‘dose and the physician charges Rs 4 per dose. In the past 60 weeks, the physician has administered the vaccine in the following quantities: Doses per week 2 25 408 Number of weoks : 5 1525S Determine how many doses the physician should buy every 5. A grocery with a bakery department is faced with the probiem of deciding how many cakes it should buy in order to meet the day's demand, The grocer prefers not to sell day-olS goods in ‘competion wih fresh products; leftover cakes are, tnerofore, a Complota loss. On the other hand, i @ customer desires a cake 4nd all of them have been sold, the disappointed customer will buy from elsewhere and the sales wil be lost. The grocer has, therefore, collected information on the past sales on a selected 00-day potiod as shown in tabla below: ‘No of Days Probability | Sales per Day 2s 10 0.10 26 30 0.30 27 50 050 28 10 0.10 Construct the payoff table and the opportunity lose table. What 's the optimal number of cakes that should be bought each day? Also find and interpret EVPI (Expected Value of Pertect Information). One cake costs Re 0.60 and sells for Re 1 6. A producer of boats has estimated the following dietibution of demand for a particular kind of boat No. demandes: 0 1 2 3 4 5 6 Probabiliy| 0.14 027 027 018 0.09 0.04 001 Each boat costs him Rs 7,000 and he salls them for Rs 10,000 each, Boats loft unsold at the end of the soason must be disposed of for Fis 6,000 each. How many boats should be in stock so as to maximize his expected profit? 7. smal industy from the past data has found that the cost of ‘making an item is Ais 25, the lems soling price is Rs 9011 Ie sold within a week. t can algo be disposed of at Rs 20 per | item at the end of the week Wooky sas: <3 4 5 8 7 2B No, of weeks | 0-10-20. 4030 ° Find the optimum number of lems per week the industry should produce, 8. A fim makes pastries which it soll at Rs 8 per dozen in special boxes containing one dozen each. The direct cost of pasties | for the firm is Rs 4.50 per dozen. At the end of the week the | slale pasiis are sold off fora lower price of Rs 350 per dozen. | Decision Theory and Decision Trees ‘The overhead expenses attributable to pasty production are As 1.25 per dozen. Fresh pasties are sold in special boxes which cost 50 paise each and the stale pasties are sold wrapped in ‘ordinary paper. The probabiliy cistribution of demand per week is as under: Demand (in dozer): 0 1 2 8 4 5 Probably oot ats 02 05 01 Find the optimal prosuction level of pasiies per week The local football club wants your advice on the number of programmes that should be printed for each game. The cost of Printing and production of programmes for each game, as quoted by the local printer, is RS 1,000 plus 4 paise per copy ‘Advertsing revenue which has been agreed for the season 's Fs 800 for each game, Programmes are sold for 15 paise each. A review of sales during the previous saasons indicates that the folowing pattern is expected to be repeated during the coming season of 50 games: ‘Number of Programmes Sold ‘Number of Games 10,000 6 20,000 20 30,000 15 40,000 10 Programmes not sold at the game are sold as waste paper to f paper manufacturer at one palse per copy. ‘Assuming that the four options lisled are the only possibities (0) prepare a payott table (i) determine the number of programmes that would provide the largest profit, i @ constant number of programmes wore to be printed for each. game (i) caloulate the profit that would arise trom a perfect forecast ‘of the number of programmes which would be sala at each ‘game, ‘The probabllty Gistibution of monthly sales of an item is as follows: Monthly sales (units) Prababilties 0.01 0.08 029 0.30 022 a0 0.08 “The cost of canyng inventory (aneok! dung the month) is Re 80. por unt, per mont, and tho cost of un ehatage f Rs 7. Boeiernino the optimum stock to minimize experod Cost A moder home applaness dealer tins that he cost of noting 2 min ooking rang stock lors month Re 200 (stance minor detroralon, erst on borowed caplet). Customers tino cannot bia a cooking range immaiatoly tend Yo go fo oiner dealers and he estimates thal for evry customer who Cama get immediate delvery, ho nsas an averago of 50. Tio probit of damand of 0, 1, 2.3, 4, 5 mi cookng {anges ina month ato 008, 0.10, 020, 030, 020, and 015, feapectvely.elarmine the optimal sick level of cooking Tanges. Ao tnd the EVPL {Bam Unie, NBA, 2005 2 company manulacturing large olectical equipment ancipating fe poosbiy of afta ora pata copper stein tren hau, t's alorping To stl wate! 1 stone 2 lege amount of copper at an addon cost oF 5. Siekpie's smal amount costng an adaionl Re 20,000; oro Stele no" aditnal copper at ah Tha stocking code,consisting of excess storage, holding and handling costs and so forth, are over and above the actual material costs HW there isa paral strike, the company estimates that an addtional cost of Fe $0,000 for delayed orders wit be incurred if thore is no stookpie at all a foal strike occurs, the cost of | delayod orders [6 estimated at Rs 1,00,000 if there i only @ ‘small stockpile, and Rs 2,00,000 it there is no stockpile. The company estimates the probabilly of a total siike as 0.1 and that of a partal stike as 0.3, (a) Develop a conditional cost table showing the cost of all futcomes and course of action combinations. (b} Determine the proterred course of action and iis cost, What is EPI? (c) Develop a conctonal opportunity loss table. (@) Without calculating the EOL, find the EVPI. AA wellknown departmental store advertises female fashion garments from time to timo in the Sunday press. Only one garment is adverisod on oach occasion. The management's experience in garmonts in the price range Rs 90-40 leads the ‘management to assess the statistical probablty of demand at various levels, after each advertisement, as fellows ‘Aller Aaverising Demand Alter Advertsing in One in Two (No. of garments) Newspaper ‘Newspapers 20 0.10 0.00 40 0.25 015, 50 0.40 035, 50 0.25 0.40 | 7 0.00 0.10 ‘The next garment to be advertised at As 35 wil cost Re 15 to make and can be disposed of tothe trade, it unsold, for Rs 10. The store's advertising agents wil charge Rs 97.60 for an artwork and blockmaking for the advertisement and each newspaper will charge As 50 for the insertion of the advertisement. You are required to: (a) Calculate how many garments to the neatest 10 should be purchased by the store in order to maximize the expected {91038 prof, after adverising in: () one newspaper, {i {Wo newspapers. (0) Caleviate the amount, if any, of the expected net proft aftor adverising: () in one newspaper, (i) two newspapers. 14. A car manufacturer uses @ special control device in each car that he produces, Two allornative methods ean be used to tect and avoid a faulty device. Under the fist method, each device is tested before iis installed. The cost of this method '8 Rs 2 per test, Atleratvely, the contol device can be installed without being tested, and a faully device ean be detectad and Fendared after the car has been assembled, at a cost of Re 20 per faulty device Rogartless of which mothod is used, faulty devices cannot be repaired and must be decarded. ‘A manulacturer purchases the control devices in batches of 10,000. Based on past experience, he estimates the proportion of defective components and the associated probability to. be: Proportion of Probability Faulty Devices 08 029 ose 070 0.16 0.0 (a) Which inspection method should the manufacturer adopt? (0) What is the expected value of perfect information (EVP)? HINTS AND ANSWERS EMV(A,) = 412.5, EMV(4) = 455, EMV(4,) = 4175 EMV(X) = 26, | EMV(Y)~= 43," EMV(Z) = 19.5; Company should manufacture product Y. EMV (accept) = 2.6, EMV(reject) = 2.8; reject. Conditional profit value = MP * units sold — ML > units unsold [(4-2)D=2D : DBs ={te-2)0-25-oyn4p-25 Des where D is the number of units demanded and S is the number of units stocked EMV* = EMV (Purchase 40 dozen) = Rs 54. Conditional profit value = MP x eake sold ~ ML cake not sold = (1 = 0.80) * cake sold ~ 0.80 * cake not sold [o.200 : DBs \020D-0.80(s-D) ; Des Where D is the number of units demanded and S is the number of units stocked EMV" = Rs 5; EOL = 0.22 (stock 26 units of cake) Conditional profit value = MP x boats sold — ML boats unsold 10,000 ~ 7,000) boats sold ~ (7,000 ~ 6,000) boats unsold (200 ™ [3,000 D— 1,000(S- 1) = 4,000 — 1,000 5 Des 3 Des EMV* = 4,080, stock 3 boats. (30-25) D ; DBS {0-25 9-cas-an(s-pyat0D-58 | Des id S is the number of sold — ML » item unsold 20) item unsold where, D is the number of item sold; $ is the number of item produced, EMV = Rs 26, produce 6 items. 8, Conditional prof value MP * boxes sold ~ ML » boxes unsold = (8 450 ~ 125 ~ 0.50) boxes sold ~ (4.50 + 1.125 ~ 5.50) boxes unsold 78D : Ds *[ursp-02818-p)=20-0258 : des EMV* = Rs 4.28, produce 4 dozen pasties. 9. (a) Concltional profit » Sales revenue {0.15D + 800 + (P — D) * 0.101) 0.14 D-0.03 P-20 *{o.11 0-200 Cost (1,000 + 0.04P) :D

PDecision Theory and Decision Trees Rs (35 — 15) whet Phe numberof programs prin and Dis the | 13. Magna of (MP demand for programmes. (b) largest expected profit ~ Rs 2,260 accrues for P = 30,000 copies per gam | (©) In case of perfect forecast, the expected profit is 10% 0.1 +2,000 0.44 3,100 0.3 +4,200%0.2=Rs2,660. | vensens VPI = Re 2560 — Bs 2.260 = Rs 400 | he ae 0. Cost function ={(P—S) 5 DSS | Max. EMV is correspond |200(s-D) ; Des = aan where D = monthly demand (or sales); § = umber of units purchased (a) EMV(S,)=Rs 600; EMV(S) EMV (S,) = Rs 1,037.5; EMV (S,) = Rs 10125. the expected cast of Rs 46 is minimum for course of action 4, the optimum slock to minimize the cost is 4 units per sooth, purchase 60 garments 500(D-S) ; D2 Ae 14, (@) Fist altemative, Rs 20,000 200(S-D) } D Rs mae On the positive side of the curve, it eventually bend away from the straight line, This indicates that increasing, units of money are resulting in smaller additional gains in utility.Decision Theory and Decision Trees 11.8.3 Construction of Utility Curves Suppose, there is a project that would payoff with Rs 1000 or (~) Rs 1000. Then what probability of success would make the decision-maker just about indifferent as to whether to undertake the project or not? Suppose he says that he would require an 80 per cent probability of success to make him indifferent about undertaking the project. Suppose utility of Rs 1000 is 10 for an individual. Then U Red) = (0.80) U(Rs 1000) + (0.20) U(- Rs 1000) 0 = 0.80 (10) + (0.20) U( Rs 10) or U(- Rs 10) = —40 The utility curve will have three points: UG 1000)=10; U@)=0 and U¢-1000)=—40 Other points on the curve can be obtained by interacting with decision-maker to know his/her utility values to a payoff Example 11.27 A manager must choose between two investments A and B that are calculated to yield net profits of Rs 1,200 and Rs 1,600, respectively with probabilities subjectively estimated at 0.75 and 0.60. Assume, the manager's utility function reveals that utilities for Rs 1,200 and Rs 1,600 are 45 and 50 utils, respectively. What is the best choice on the basis of the expected utility value (EUV)? Solution The expected utility value is expressed as: Buv= © Usp, where U, = utility value of state of nature { , = probability value of state of nature i ‘Then BUV (A) = pyll, = 0.75 * 45 = 33.75 utils and EUV (B) = pyU, = 0.60 * 5 30 utils Since EUV(A) > BUV(B), therefore, the best choice is investment A. Example 11.28 A businessman is thinking whether to invest in bonds with an assured return or in a new venture that is likely ¢o fetch him Rs. 20,000 or nothing with equal probabilities, The businessman says he would prefer an assured return in itis Rs. 10,000 or more, would be indifferent to the two alternatives if the assured return is Rs. 8,000 and would opt for the risky alternative if this amount is less than Rs. 6,000. The businessman had purchased a minicomputer last year which is lying with him unused at present. The ‘minicomputer can fetch him a profit of Rs, 8,000 by way of consultancy. He says that he would not like to sell the computer but would be indifferent to an offer of Rs. 3,000. Recently a Govt agency invited bid offers for a contract worth a profit of Rs. 8,000. Bidding expenses are reimbursable ifthe contract materializes and not otherwise. There is an equal chance of winning or losing the bid. AfRer thinking over the possibilities, the businessman says that he would be indifferent to submitting the bid ata bidding expense of Rs. 2,000. Construct the businessman's utility curve by using the CME technique, Solution Let arbitrary utility values of 100 and 0 be assigned to the most favourable and least favourable outcomes, i. Usp oq0 = 100 utiles, Uy = 0 utile 0.5 x 100+0.5 x 0 ‘The businessman is indifferent between his utilities for Rs. 3,000 and of a chance of Rs. 8,000 or nothing. Thus, Us.900 = 0-5 Us.oo9 + 0.5 Uy = 0.5 * 50 +05 x 0 = 25 utiles. The decision to enter the bid at an expense of Rs. 2,000 may get him either gain of Rs 8,000 or lose (bidding expenses of) Rs. 2,000 with equal probability. Since he is indifferent to submitting the bid, Uy = 0.5 Ugcen + 0-5 U_.090 0=0.5*50+0.5U, Given that Us.000 = 05 Uro.ogo + 05 U, 5 utiles. OF U2999%~50 utiles.Operations Research: Theory and Applications Following in the summary of points for the Utiles businessman's utility curve Amount (Rs) uty utes) 20,000 100 8,000 50 3,000 25 0 0 2,000 50 7 Fig. 11.10 =e — 5,000 5,000. 10,000 15,000 20,000 Ublty Curve The utility curve is plotted in Fig. 11.10. This curve ay reflect a reasonable estimate of the businessman's 50 behaviour when faced with risky ventures. Example 11.29 Mr XX has an after-tax annual of Rs. 90,000 and is considering to buy accident insurance for his car. The probability of accident during the year is 0.1 (assume that at most one accident will occur), in which case the damage to the car will be Rs. 11,600. With autility function of U(x) = v/x , what is the insurance premium he will be willing to pay? Solution Let A represent the venture when Mr XX does not by the accident insurance for his car. Then, in ‘case of accident he would spend Rs. 1,600 on damages and will be left with Rs, 78,400. In case of no accident he retains Rs. 90,000. Then we have Ug = Ure.00 Since UX) = U,= VF Ursaan U =280*0.1 +300 *0.9=798 ils, So the amount Rs. x which will give the same utility of the venture: A= (298)? U)= vx or X=(U@F] x 0.1 + Usg 78,400, go 0.9. 280 utiles, and Usa ogg "90,000 = 300 utiles. Thus, Rs. 88,804. [Because Thus Mr XX will be indifferent to an amount of Rs. 88,804 with certainty and the venture A. The amount he will be willing to pay as car premium would be 90,000—88,804 = Rs. 1,196. SELF PRACTICE PROBLEMS C Raman Industries Lid. has a new product that they expect has great potential. At the moment they have two courses of action open to them: To test market (S,) and to drop product (S;). TE they tos tit wil cost them iS 50,000. The response could be either positive or negative, with probabities 0.70 and 0.30, respectively. I it i positive, they could oithor market it with ful effort or drop the product. If they market with full scale, then the result ight be low, medium, or high demand and the respective het payots would be — Rs 1,00,000, Rs 1,00,000 or As 5,00,000. These outcomes have probabilities of 0.25, 0.55 and (0.20, respectively. i the rooutt of the tost markotng fs negative they would secide to drop the product. I, at ary point, they drop the product there is a net gain of Rs 25,000 from the sale of scrap. All financial values have been discounted to the present. Draw a decision tree for the problem and indicate the most preferred. decision, ‘A manufacturing company has just developed a new produc. ‘On the basis of past experience, a product such as this woud either be successful, with an expected gross relun of AS 1,00,000, oF unsuccesslul, with an expected gross retum of Ris 20,000. Similar products manutacturad by the company have a record of being suocesstul about 50 per cent of the time. The production and marketing costs of the new product are expected to be Ris 50,000. The company is considering whether to market this new product or to drop ft. Before making its decision, a test marketing blot can be conducted at a cost of Rs 10,000. Based on past ‘experience, test marketing results have been favourable about 70 per cent of the me, Furthermore, products favourably tested have been suocesstul 60 per cont of the time. However, when the test marketing result has been unfavourable, the product has only been successiul 30 per cent ofthe time. What course of action should the company pursue? 3. XYZ Company manulactures guaranteed tennis balls. At present, ‘approximately 10 por cent of the tennis balls are defecive. A ‘fective ball leaving the factory cosis the company Fee 0.50 to honour its quarantee. Assume that all detective bali are returned. ‘At a cost of Re 0.10 per ball, the company can conduct a test that always correctly Identlies both good and bad tennis ball. (a) Draw a decision tree and determine the optimal course of sclion and its expected cost (&) At what fest cost should the company be inferent o testing? 4, The Ore Mining Company is attempting to decide whether or not 1 cartain pigce of land should be purchased. The land costs RS 300,000. I there are commercial ore deposits on the land, the estimated value of property is Fs §,00,000. If no ore deposits exist, however, the property value is estimated at Rs 2,00,000. Before purchasing the land, the property can be cored at a cost ‘of Re 20,000. The coring wil indicate if conditions are favourable ‘oF unfavourable for oe mining. I the coring repot is favourabie the probabilty of recoverable ore deposits on tho land is 08. However, i the coring tenor le Unlavourable then probability Is ‘only 0.2. Prior to obining any coring information, the management ‘estates that the odds of ore being present on the land is 50-50. Management has also received coring reports on pieces of 'acd similar to the ore in question and found that 60 per cent of the coring reports were favourable. | Construct a decision tree and determine whether the company should purchase the tand, decine to purchase i, or take coring test before making iis decision. Specily tho. optimal course of action and EMV. XYZ company, dealing with a newly invented telephone device, 's faced with the problem of selecting out of ‘he following avaliable courses of action (i) manufacture the device tse or (i) be paid on a royalty basis by another manulacturer; or (i) sel the rights for ts invention for e fump sum. The profit (in ‘0008 Fs) that can be expected in each case and the probabilties associated with the level of sales are shown a the following table: Provabilty Manufacture Royaties Soll Ar Outcome ‘tsar Fights High sales 01 78 as 15 Medium sales og 25 2% 45 Low sales 06 -10 108 Represent the company’s problem in the form of @ decision tree. Further redraw the decision tree by inducing the following Aadaitonal information (@) It manufactures tse, and the sales are medium or high, {hen the company has the opportunity of developing a new version of is telephone; From past experience the company estimates that there is 2 50 per cent chance of successt development The cost of development is As 18 and relums alter ‘eduction of development cost are As 80 and Re 10 for high ancl medium sales, respectively ‘An automobile owner faces the decision as to which deductible amount of comprehensive insurance coverage to select Comprehensive coverage includes losses due to fre, vehicle thet, vandaliom and natural calamites. The possible choices are 2210 deductible coverage for Rs 60 per year or Rs 50 deductible overage for Rs 45 per yoar. (The owner pays tho fist As 0 of any loss of at least Ais 500.) Considering accidents covered by the comprehensive portion of the policy some of te owners | tity values are given below. ‘Amount (Fs) > 95 60 Unity 02 04 (@) Sketch the owner's utlity curve. avoider, an EMV'er of a risk taker? (©) On the basis of the utity curve drawn in part (a), should the owner take the zero deductibie or the Rs 50 dedvetile ‘comprehensive coverage? | Suppose a company has several independent investment pportunities each of which has an equal chance of galning Fs 100,000 or losing Fs 60,000. What is the probabiliy that the company will lose money’ on twa such investments? On three Such Investments? On four such investments? a company has a number of independent investment ‘©pportuntias, in each of which the tnancial risk ig relatvely sl ‘compared to its overall asset postion, why should the company try to maximize EMV, rather than expected uty? An oil ring company is considering the purchase of mineral | Fights on @ property of Re 100 lakh. The price inchides tess 10 indicate vhether the property has type A geological formation or ‘ype B geological formation. The company wil be unable 10 fel {he type of geological formation unt the purchase i made. i's known, however, that 40 per cent of the land in tis area as "ype A formation and 60 per cent type B formation. # the. | o) () -50 -45 0 045 047 08 Is the owner @ risk it Decision Theory and Decision Trees ‘company decides to chill on the land it will cost Rls 200 lakh. tt the company does dill may hit an of wel, gas wel ot diy hole Dring experience indeates thal the probably of sling an ol well is 0.4 on type A and 0.1 on type B. formato: Probabilly of hiting gas is 0.2 on type A’and 0. on type & formation. The estimated discounted cash valve from on Chee is Rs 1,000 lakh and from a gas well is Fs 500 lakh Tris includes everything except cost of mineral rights and cost ay ‘ling. Use the decision-tiee approach and recommend whether the company shoukt purchase the mineral rights? The Sensual Cosmetic Co. has developed a new pertume which management feels has a tremendous polenial. I not only interacts with tho wearer's body chemistry 10 create a unique tragrance but is algo espacialy long lasing, A total of is 10 tek has already been spent on iis dovelopment. Two. marketing Plans have’ been «devised: (8) The fist plan folows the company’s usual poly of giving smal samples of the new products when othar items in the company’s product lines are. purchased and placing ‘advertisements in women's. magazines, ‘The plant would Cost Rs 5 lakh and t's boleved that i might result ina high ‘moderate or low market response with probabiliy of 02, 2.5 and 0.3, respectively. The ‘net profit excluding development and promotion costs in those cages would bo FRo 20 lekh, Fis 10 lakh and As 1 lakh, respectively. it lator appears that the market response is going to bo low it would be possile to launch a TV advt. campaign, This would cost another As 7.5 lakh. t would change the market esponso to high or moderate as previously described but with probability of 05 each ‘The second marketing plan is much more aggressive than the first. The emphasis would be heavily upon TV. ‘advertising, The total cost ofthis plan would be Re 16 lak bat the market response woul be elther excellent or good, with probabilties of 0.4 and 0.6, respectively. The ‘ott {excluding development and promotion costs would be Rs 30 lakh and Rs 25 lakh for the two outoomes Advise on the sequence of strategy to bo followed by the ‘company The investment staff of TNC Bank is considering four investmont Proposals for @ client shares, bonds, real estate and saving Certificates. Those investments will be held for one year, The pest deta regarding the four proposals are given below: ‘Shares: Thore Is @ 25 per cent chance that shares will decine ‘8 10 per cent, @ 30 per cont chance that they will remain stable and a 45 per cent chance that they will increase in value by 15. Ber cent. Also the shares under consideration do not pay any dividends. Bonds: These bonds stand a 40 per cent chance of increase In value by § per cent and 60 per cent chance of 1emaining stable and yield 12 per cont eal Estate: This proposal has @20 percent chance of increasing 30 per cent in value, a 25 per cent chance of increasing 20 pet Cent in value, a 40 per cent chance of increasing 10 per cent in value, a 10 per cent chance of remaining stable and a 6 per cent chance of losing § per cont ots value Savings Certlcates: Those coriicates yield 8.6 per cent with certainty Use a decision tree to structure the altematives to the investment stall, and using the expected value orilerion, choose the ltemative with the highest expected value ‘A company has developed a new product in its R&D laboratory, The company has the option of setting up production faci to ‘market this product straighiaway. tthe product is success, ‘then over tho thrae years expected product life, the returns wil bbe Rs 120 lakh with @ probabilly of 0.70. tthe market does not respond favourably, then the reluins wil be only Fs 15 lakh, wih 2 probabilty of 0.30 ©)Operations Research: Theory and Applications “The company is considering whether il should tast market this product by building a smal pilot plant. The chance that the test ‘market wil yield favourable response Is 0.60. i the test market es a favourable response, then the chance of successtul total market improves to 0.5. If the test market give @ poor response, then the chance of success in the total market i ony 0.90 [As beloce, the retums from a successful market would be As 120 lakh and from an unsuecessiul market only Rs 15 lakh. The. instalation cost to producion is Rs 40 lakh and tho cost of test. rmarkating tho plot plant Is AS 5 lakh. Using decision-tree analysis, draw a decision tree diagram, camry out necessary analysis to determine the optimal decisions {Doni Univ, MBA, 2002) HINTS AND ANSWERS 1, The company should test matket rather than drop the product. test matket + product otherwise drop the product sult is positive, the company should market the 2. ‘Test market should be followed; EMV = Rs 13,800 3. (a) Do not test, Re 0.05: (b) Re 0.05 4, Test: If favourable ~ putchase; If not ~ do not purchas 64,000 Decision Point Outcome (Drill il well Dry hole Gas (Do not drill D, ~—@) Drill Oil welt Dry hole Gas (ii) Do not drill D, Gi) Type A formation (i) Type B formation The company should purchase the mineral right 9. Aggressive Plan, Probably ou 06 03 04 04 02 04 06 Royalties; IMV = Rs 15.5 If p is the probability of high sale, then ()) manufacture itself when p > 0.24 (i) paid on royalty basis when 0.07

1aC+ (a) EOL (e) emv (6) p= IGKIC + 1) (d) p= uc + 1p) | (6) Hurwice (@), Maximex Rooke wT 2 8F 47 BF 6r 7T BF O9F WF 11. a course of action or strategy 12. risk 13. highest 14. optimism 15. EOL exterion 16. risk 17. minimum 18. weighted average 19, states of nature 20. posterior probablites 21. (8) 22. (a) 23.) 24. (bY 25. (6) 28. (a) 27. ()—28. (6). (a) 90. (0) | Sta) 92) B®) 3H (e) 35. (0) CASE STUDY Case 11.1: Oil India Corporation The corporation has recently got feaschold drilling rights on a large area in the westem part ofthe country. No seismic coverage is available and to condvet a detailed survey Rs 3 milion is required, If ol is struck, a large reserve may result in a net profit of Rs 30 million, whereas a smaller marginal reserve may result in a net profit of Rs 18 milion The cost of driling a wildcast well is Rs 7 million Scismie is thought to be quite reliable in this area. Uncertainty pertains to whether or nota structure exists, The company assesses a probability that the test producing good, fair or bad result is 0.40, 0.30 and 0.30, respectively, (On the basis of past drilling records and experiences indicating the probabilities of striking oil in large reserve, smaller ‘marginal reserve or dry hole, even in the presence of good, fair and bad reading of seismic study are as under Good 030 025 Pair 030 040 od 0.10 070 Exploratory group has suggested two possible exploration strategies: (@)_ Drill at once on the basis of present geologic interpolation and extrapolation (b) Conduct a seismic study and defer drilling till seismic data is reviewed ‘AS a-member of strategic group of the company evaluate the two strategies suggested by the exploratory group. Case 11. Moola Farms Mr. Moola, the owner of Moola Farms is attempting to decide which of three erops he should plant on his 100 acre farm, The profit from each erop is strongly dependent onthe rainfall doring the growing season. He has categorized the amount of the rainfall as substantial, moverate oF light. He estimates his profit for each crop as shown in the table below: ‘Substantial 7,000 2,500 4,000 Moderate 3.500 3,500 4,000 1,000 4,000, 3,000 Based on the weather in previous seasons and the current projection for the coining season, he estimates the probability of substantial rainfall as 0:2, that of moderate rainfall as 0.3, and that of light rainfall as 0Decision Theory and Decision Trees Services of forecasters could be employed to provide a Uetled survey of eurent rain all prospects as shown in the tabe Moderate 030 0.60, 0.10, 010 020 070, You in the capacity of Block Advisor to farmers are expected to suggest (a) Which crop should Mr. Moola plant based on existing data (6) Whether it would be economical for the farmer to hie the services ofa forecaster. If yes, then how will the result affects the decision process Case 11. Matrix India Matrix India, a company in FMCG sector is planning to launch a new product that can be initially introduced in ‘Westem India or in the whole country. If the product is introduced only in Western India, the investment outlay will be Rs 12 million. After two years, company can evaluate the project to determine whcther ic should cover the whole country: For such an expansion it will have to incur an addtional investment of Rs 10 million. To introduce the product in the whole country right inthe beginning would involve an outlay of Rs 20 milion. The product, in any ease, will hhave a life of 5 years, after which the plant will have ze10 net val. If the product is introduced only in Westem India, demand would be high or low with the probabilities of 0.8 ‘and 0.2, respectively and annual cash inflow of Rs 4 million and Rs 2.5 million, respectively. If the product is introduced in the whole country, right atthe beginning the demand would be high ot low with the probabilities of 0.6 and 0.4, respectively and annual cash inflows of Rs 8 million and Rs 5 milion, respectively, Based on the observed demand in Wester Indi, ifthe product is introduced in the entire country the following Probabilities would exist for high and low demand on an all Indie bass, High demand 0.90 ‘Low demand 040, ‘The hurdle rate applicable to this project is 12 percent. Advise Matrix India on the investment policy it should follow. Support your advice with appropri e reasoning.