Decision Theory and Decision Trees

DecisionMaking Tools


and the script must be choreographed and staged by the 2  . Achieving good results for local outcomes is an important objective for individual operational units and individual operations managers. the orchestra. the music.Introduction • At the Operational level hundreds of decisions are made in order to achieve local outcomes that contribute to the achievement of the company’s overall strategic goal. the costumes. all these decisions are interrelated and must be coordinated for the purpose of attaining the overall company goals. However. • Decision making is analogous to a great stage pay in which all the actors.

Decision Theory Decision Theory represents a general approach to decision making which uses quantitative methods suitable for a wide range of operations management decisions. including: • Capacity planning • Location planning • Product and service design • Equipment selection  3 .

one of which may be to do nothing 4 . • Develop a list of possible alternatives. Future states of nature may be high demand or low demand for a product or good economic conditions or bad economic conditions etc.Decision Theory Process • Identify possible future conditions called states of nature.

revenues. given the various states of nature.  5 . Payoff’s are typically expressed in terms of profit.Decision Theory Process • A payoff table is a means of organising and illustrating the payoffs from the different alternative decisions.

determine the likelihood of each possible future condition • Evaluate alternatives according to some decision criterion and select the best alternative 6 .Decision Theory Process (Cont’d) • If possible.

probability are assigned to the states of 7 .e.Decision Environments • Uncertainty .Environment in which it is impossible to assess the likelihood or the probabilities of various future events (Or probability are not assigned to states of nature) • Risk .Environment in which certain future events have probable outcomes i.

Decision Making under Uncertainty • Maximin – (Pessimistic nature) Choose the alternative with the best of the worst possible payoffs • Maximax – (Optimistic nature) Choose the alternative with the best possible payoff. The decision maker chooses the most favorable states of nature for each alternative decision. • Laplace .Choose the alternative with the best average payoff of any of the alternatives 8 .

50.00.000 9 .000 5.00.000 3.00. This is represented in the following table: Which strategy should the concerned executive choose on the basis of of Nature States i) Maximin Criterion ii) N2 Maximax criterion iii) Minimax regret Strategies N1 N3 criterion iv) Laplace criterion?    S1 S2 S3 7. The three possible states of nature or events are: i) high increase in sales.Example • A food products company is contemplating the introduction of a revolutionary new product with new packaging or replace the existing product at much higher price or a moderate change in the composition of the existing product with a new packaging at a small increase in price or a small change in the composition of the existing product except the word ‘New’ with a negligible increase in price.000 4.00. ii) no change in sales and iii) decrease in sales. 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).000 0 3.50.000 3.000 1.

Expected value is computed by multiplying each outcome by the probability of its occurrence and then summing these products according to the formulae: EMV(x) = ∑ p(xi)*xi Where xi = ith outcome p(xi) = probability of ith outcome The decision maker selects the decision that has  10 .Decision Making under Risk Expected Monetary Value (EMV).

However. The information has some maximum value that is the limit of what the decision maker would be willing to spend. enabling the decision maker to make a better decision.Decision Making under Risk Expected value of perfect information: Occasionally additional information is available. a company could hire an economic forecaster to determine more accurately the economic conditions that will occur in the future. it would be foolish to pay more for the information than it stands to gain in extra profit from having the information. This value of information can be computed as an expected 11  . regarding future events. For example. or can be purchased.

By consensus.Example A glass factory specializing in crystal is experiencing a substantial backlog. management estimates the respective  12 . or high. which may be low. C) Do nothing (no change) The correct choice depends largely upon demand. and the firm's management is considering three courses of action: A) Arrange for subcontracting. medium. B) Construct new facilities.

Example The management also estimates the profits when choosing from the three alternatives (A. in thousands of dollars are presented in the table below:  13 . These costs. B. and C) under the differing probable levels of demand.

4 = 62 -120*.1) + (50*0.5) + (200*0.4 = 46 Decision: B EVPI = (Best outcome for 1st state of nature) * p(x1) + (Best outcome of 2nd state of nature) * p(x2) + (Best outcome for 3rd state of nature)* p(x3) – EMV under risk 14 = (20*0.4) -80. Further it is find the best decision out of A. medium and high) hence we use decision criterion under Risk (EV & EVPI).5 + 60*.5 + 90*.1 + 40*.5  .5 +200*. B or C Expected Values 10*0.4 =80.5 20*.Solution Probability are assigned to the states of Nature (Low.1+ 50*.1 + 25*.5 =26.