This action might not be possible to undo. Are you sure you want to continue?
What is Management Science?
Management Science is concerned with developing and applying models and concepts that help to illuminate management issues and solve managerial problems. The models used can often be represented mathematically, but sometimes computer-based, visual or verbal representations are used. The range of problems and issues to which management science has contributed insights and solutions is vast. It includes scheduling airlines, both planes and crew, deciding the appropriate place to site new facilities such as a warehouse or factory, managing the flow of water from reservoirs, identifying possible future development paths for parts of the telecommunications industry, establishing the information needs and appropriate systems to supply them within the health service, and identifying and understanding the strategies adopted by companies for their information systems.
DECISION MAKING ENVIRONMENT
Elements of a Decision There are three components to any decision: (1) The choices available, or alternatives; (2) The states of nature, which are not under the control of the decision maker; (3) The payoffs. These concepts will be explained in the following paragraphs. The alternatives, or acts, are the choices available to the decision maker. Ford can decide to manufacture and assemble the door locks in Sandusky, or it can decide to purchase them. To simplify our presentation, we assume the decision maker can select from a rather small number of outcomes. With the help of computers, however, the decision alternatives can be expanded to a large number of possibilities. The states of nature are the uncontrollable future events. The state of nature that actually happens is outside the control of the decision maker. Ford does not know whether demand will remain high for the F-150. Banana Republic cannot determine whether warm-weather or coldweather teams will play in the NCAA basketball final. A payoff is needed to compare each combination of decision alternative and state of nature. Ford may estimate that if it assembles door locks at its Sandusky plant and the demand for F-150 trucks is low, the payoff will be $40,000. Conversely, if it purchases the door locks assembled
He identifies this state of nature. The various techniques used for solving these type of problems are: Linear Programming Integer Programming System of equations Dynamic programming Queuing models Inventory models Capital budgeting analysis Break – even analysis 2 . takes it for granted and presumes complete knowledge as to its occurrence. In this type of decision problems.000. the decision maker knows with certainty the results of selecting every course of action or decision choice. The main elements of the decision under conditions of uncertainty are identified schematically: Decision making under certainty: In this environment.MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 and the demand is high. the decision maker presumes that only one state of nature is relevant for his purposes. the payoff is estimated to be $22.
probabilities associated with the occurrence of different states of nature are not known i. the best decision is to select that course of action which has the largest expected payoff value.MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 Decision making under Risk: In this situation. But he is supposed to have believable evidential information. under uncertain situations. experience or judgment to enable him to assign probability values to the likelihood of occurrence of each state of nature. it is possible to determine EMV For each course of action if the decision could be repeated a large number of times. knowledge. the decision maker faces several states of nature. Suppose there is prior knowledge either on basis of past experience or on a subject basis that the state of nature Sj has a probability of occurrence P(Sj) So. Such situations occur when a new product is introduced in the market or a new plant is set up. Sometimes past experience or past records often enable the decision maker to assign probability values. What is EMV? Given a payoff table with conditional values(payoffs) and probability assessments for all states of nature. The problem of decision making under uncertainty is to choose an action (or decision) among many different available actions which gives (possibly) maximum expected profit or maximum expected revenue or minimum expected losses or minimum expected costs as the case may be. EMV for given course of action is just sum of possible pay offs of the alternative. Knowing the probability distribution of the states of nature. 3 .e there is no historical data available. The most widely used decision criterion for this type of problem is expected monetary value (EMV). EMV (A) = p11 P(Sj) + p12 P(Sj) + ………+ pim P(Sm) DECISION MAKING UNDER UNCERTAINITY Under this situation. each weighted by the probability of that payoff occurring.
Should we or should we not conduct the Experiment? Answer: If the expenses on the experiment are more than the increase in the expected pay-off then we do not conduct the experiment otherwise we conduct it. The management believes that there is a 0. NEW PRODUCT Example: A firm is considering a final “GO” decision on a new product. should the firm introduce the product? ii. minmax etc). Denote it by EMV. then firm gave “GO” advice eight times out of ten. To do so we may have to conduct some sort of Experiment (Sample Survey or Hire Some Expert). (i) When advice was given (either by client firm or by others in the market place) on products that later proved to be successful.000. We can always find the Expected Value of the MV and call it Expected Monetary Value Under Uncertainty or Expected Pay-off Under Baye’s Criterion. we have some information with which we can associate certain Monetary Value (to be denoted by MV and measured in terms of dollars) or Utility (to be measured in terms of Utiles).000 and if it is unsuccessful the loss is $300. The question then arises: Q. EXPERIMENT: To increase (or IMPROVE) the expected pay-off we obtain ADDITIONAL INFORMATION (Under Uncertain Conditions). ii. Its betting average in similar situations is as follows. Suppose. There is no profit or loss if the product is not introduced.20 probability (the odds are 2 to 8) that the product will be successful. We associate probabilities with this uncertainty. (ii) When advice was given (either by client firm or by others in the market place) on products that later proved to be unsuccessful. If the product is introduced and it is successful. Based on the above information and under a variety of other criteria (like maxmin.MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 i. A consulting firm specializing in new product introduction has offered its services to firm. i. the profit is $500. There may be some uncertainty involved with this information. 4 . then firm gave “STOP” advice fifteen times out of twenty.
It is an optimistic criterion. Optimal Profit = $500. Steps are: i.000 as consulting fee to give advice. profits.MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 The Firm charges $5. profit.2 Unsuccessful -300 0 . Maxi-max Criterion (Optimistic View): In the maximax criterion the decision maker selects the decision that will result in the maximum of maximum payoffs.8 DIFFERENT DECISION MAKING CRITERIA without using Probability (6 in all) 1. PAY-OFF TABLE-2 EVENTS OR STATES OF NATURE ACTIONS GO STOP Successful $500 $0 Unsuccessful -300 0 MV $500 $ 0 Answer: Max-max Action = Go. Should the firm be hired in order to maximize the expected profit? PAY-OFF TABLE-1 EVENTS OR STATES OF NATURE ACTIONS GO STOP Prior Probs.000 For Go Max payoff is $500 5 . ii. Successful $500 $0 . of these max. Choose the action which has the max. For each action choose max.
of these min . PAY-OFF TABLE-3 EVENTS OR STATES OF NATURE ACTIONS GO STOP Successful $500 $0 Unsuccessful -300 0 MV $500 $ 0 Answer: Max-max Action = Stop. Choose the action which has the max. Maxi-min Criterion (Optimistic View): In the maximin criterion the decision maker selects the decision that will reflect the maximum of the minimum payoffs. profits. Steps are: i. Optimal Profit = $0 For Go Max payoff is $500 For Stop max payoff is $0 Minimum of these above mentioned maximum values is $0 so choice is Stop decision 3. 6 . a pessimistic criterion. profit.profits. ii. For each action choose min. Choose the action which has the max.MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 For Stop max payoff is $0 Maximum of these above mentioned maximum values is $500 so choice is GO decision 2. profit. of these min. Min-max Criterion (Pessimists View): This criterion is the decision to take the course of action which maximizes the minimum possible payoff. ii. For each action choose min. Steps are: i.
Optimal Profit = $0 For Go Min payoff is . PAYOFF TABLE 5 Regret table EVENTS OR STATES OF NATURE ACTIONS GO STOP Successful $500.$0 = $500 Unsuccessful $0 .MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 PAY-OFF TABLE-4 EVENTS OR STATES OF NATURE ACTIONS GO STOP Successful $500 $0 Unsuccessful -300 0 MV $-300 $ 0 Answer: Max-min Action = Stop.The decision maker attempts to avoid regret by selecting the decision alternative that minimizes the maximum regret.$300 For Stop min payoff is $0 Maximum of these above mentioned maximum values is $0 so choice is Stop decision 4) Mini Max regret criterion: Regret is the difference between the payoff from the best decision and all other decision payoffs.$500 =0 $500 .-300 = $300 $0 -$ 0 = $0 Maximum Regret for Go decision is $300 Maximum regret for stop decision is $500 so We minimize the maximum regret and choose Go 7 .
5) + 0$(.6) = $20 $0(.) then looking at the pay-off table.4 calculations are performed here.4) . A coefficient of optimism.6) = $0 EXPECTED VALUE OF PERFECT INFORMATION Suppose the FIRM knows (has perfect information) that (I) the product is going to be successful (I.5) = 0 6) The Hurwicz criterion is a compromise between the maximax and maximin criterion.5) . the event “successful” is going to occur). Decision Go Stop Values $500(. For alpha value of 0. for each decision. thus assuming that the states of nature are equally likely to occur.000 . then looking at the pay-off table. is a measure of the decision maker’s optimism.. Decision Go Stop Values $500(.300(.2 (ii) the product is going to be unsuccessful (i.4) .8 8 . p = . the action is evidently A1. . the event “unsuccessful” is going to occur.. The Hurwicz criterion multiplies the best payoff by and the worst payoff by 1.e.5) = 100 $0(.e. with profit $0 . the action is evidently A2.300(.MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 5) The equal likelihood ( or Laplace) criterion multiplies the decision payoff for each state of nature by an equal weight. and the best result is selected. p = .0(. with profit $ 500.
Value of Perfect Information (EVPI) = $100 -$ 0 = $100 Interpretation: The maximum amount the company should be willing to pay to obtain perfect information (additional information) is $100.20) + 0 (. EVPI equals the expected opportunity loss (EOL) for the best decision. E (Profit Under Perfect Information)….000..MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 PAY .OFF TABLE EVENTS OR STATES OF NATURE ACTIONS A1 = GO A2= STOP Prior Probs.8) = 100 The Exp.e. Therefore.2 Successful $500 $ 0 . .E (Value Under Perfect Information) i. EVPI equals the expected value given perfect information minus the expected value without perfect information. The expected value of perfect information (EVPI) is the maximum amount a decision maker would pay for additional information.8 Unsuccessful 9 . EVUPI = 500 (.
In the view of the fact that the firm would have to incur extra cost by payment to the agency conducting the marketing research.8 Exp.2 Unsuccessful -300 0 . through statistical sampling or through an expert opinion etc. may not be perfect. it would like to know the value to be placed on the information provided by the agency. Successful $500 $0 .000 CATEGORIZATION or Classification: We classify the Additional Imperfect Information obtained through Sampling/Experimentation (Expert/Consultant’s Advice) of “Consulting Firm” into the following Two Categories/Classifications: 10 . for the “Firm” it is possible to obtain Additional Imperfect Information through then services of a “Consulting Firm” at a cost of $ 5. Under such circumstances it may be worthwhile to obtain some additional information.OFF TABLE-6 is EVENTS OR STATES OF NATURE ACTIONS A1 = GO A2= STOP Prior Probs.000. say. MV $-140 $ 0 EXPECTED VALUE OF SAMPLE INFORMATION: Perfect information may be either impossible or very expensive.MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 Now the PAY . Such an information (Imperfect) may be obtained. EVSI = Expected pay – off with sample information – Expected payoff without sample information Suppose. Then Cost of Sampling or Cost of the Experiment = $5.
When advice was given on products that later proved to be successful (either by the client firm or by others in the marketplace). then firm gave “STOP” advice fifteen times out of twenty. X2 = When advice was given on products that later proved to be successful. the firm gave “GO” advice eight times out of ten. Based upon the above imperfect information the above data can be interpreted as conditional probabilities (imperfect additional information) as follows.8 . the “Consulting Firm” gave STOP. To do so we obtain. (“Consulting Firm” gave GO. With the help of these probabilities we shall REVISE our PRIOR INFORMATION. ii. the “Consulting Firm” gave GO.8 sum of the row is not 1 Now in above we have both Prior Probabilities and Conditional Probabilities.2 1 .75 1 .2 Unsuccessful .e. P(X=Xi/E=Ei) CONDITIONAL PROBABILITIES P(X/E) ACTS X1= GO X2= STOP Sum of the Columns Prior Probs. Xi/ a state of nature E is given). i. When advice was given on products that later proved to be unsuccessful (either by the client firm or by others in the marketplace).25 . X1 = When advice was given on products that later proved to be successful. 11 .MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 i. EVENTS OR STATES OF NATURE Successful . etc.
6=.2 = . 12 .8 = . X1= GO + .2 *.64 Prior Probs.36 X2= STOP +. Join Probabilities. sum 1.04 .25= .MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 i.16 EVENTS OR STATES OF NATURE Successful Unsuccessful EFFICIENCY OF EVSI Efficiency of sampling = EVSI/EVPI *100 Efficiency of EVSI enables the decision maker to decide whether or not to seek the information and compare various sources through which the needed information may be obtained.75=.8*.2*.2 = .6 .8* .8 . Marginal and Revised (Posterior) Probabilities Joint Probabilities Table P(X and E) = P(E) P (X/E) JOINT PROBABILITIES P(X/E) ACTS Marginal Prob. and from there we obtain.2 .2 .04 . ii.16 .00 .
EMV* with prior probabilities ENGS (Expected net gain under sampling = EVSI – Cost of sampling information Efficiency of sampling = EVSI/EVPI *100 13 .MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 SUMMARY EMV = expected monetary value EOL is expected opportunity loss EOL* and EMV* will always result in the same decision EOL* =EVPI EVPI is Expected value of Perfect information = Expected value under perfect information – EMV* with prior probabilities EVSI is Expected value of Sampling information = EMV* (with posterior and marginal probabilities) .
branches. A decision tree enables the decision maker to see the various elements of the problem in a systematic way. probability estimates and pay offs. PORTRAY OF DECISION TREE Q1 CHANCE NODE Q3 Q4 Q5 A1(Decision branch) Q6 Q7 Decision node Q10 Q9 Q11 Q12 14 . There are two types of decision trees: Deterministic decision tree Probabilistic decision tree Deterministic decision tree involves making only one decision under conditions of certainty and in probabilistic decision tree. Each situation is shown by a distinct path through the tree.MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 DECISION TREE A decision tree consists of network of nodes. a problem involving only one decision is to be made under risky conditions. A decision tree is a graphical representation of the sequences of action-event combinations available to the decision maker.
MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 EXAMPLE: RULES AND CONVENTIONS TO PREPARE A DECISION TREE: Identify all decisions to be made and the sequence in which they are to be made. Obtain outcomes of the possible interactions among decision alternatives and events. Calculate the expected value of all possible decision alternatives. Estimate probabilities that possible events or states of nature will occur as a result of the decision alternatives. 15 . Identify the chance events or nature of their state. The tree is constructed from left to right. Select the decision alternative offering the most attractive expected value. Develop a tree diagram showing the sequence of decisions and chance events.
Decision tree is useful in cases where an initial decision and its outcome affect the subsequent decisions.MANAGEMENT SCIENCE IN DECISION MAKING 20-MBA 2012-14 ADVANTAGES OF DECISION TREE: Decision tree diagram is useful for portraying the inter-related. It can be applied in various fields such as introduction of a new product. the decision maker will be in a position to visualize the entire complex of the decision problem in all its dimensions as also the actual processes and stages for arriving at the final choice. By drawing a decision tree. 16 . Enables the decision maker to see the various elements of his problem in content and in a systematic way. sequential and multi-dimensional aspects of any major decision-problem within the system’s framework. Focuses attention on the critical elements in a decision problem over the duration of its solution. apart from bringing to light the relationship between the presently available courses of action and the network of future events.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.