Decision Making

In business, managers always make decisions in their operations from time to time.
For Instance:
 Choosing what particular marketing strategy to be used in introducing a new product;
 Determining how many sales representatives to be hired and;
 Knowing what should be the selling price of a particular product; are situations which need
decisions.
Decision – Making - the act of selecting a preferred course of action among alternatives.
Example: Making and implementing decisions are crucial parts of the management.
Remark: Decision makers should make careful evaluation of the different alternatives and select the best
alternative that would meet the requirement on solving the problem based on the criteria and in
accordance with the organization’s objective
Steps in Decision Making
1. Identify the problem.
2. Determine the criteria that can solve the problem.
3. Search for different possible alternatives.
4. Evaluate each alternative.
5. Select the best alternative.
6. Implement the chosen alternative.
7. Monitor the implementation to ensure it is in accordance with the organization’s objective.
Three decision environments that can be applied in decision making depending on the existing problem
of the decision maker:
1. Decision – Making Under the Condition of Certainty
The decision maker knows which state of nature will occur. At this situation, he will decide
definitely on the best possible solution favorable to him since he knows exactly what will happen.
2. Decision – Making Under the Condition of Uncertainty
When the decision maker has no information or estimates of the probability of the events, he
will decide using any of the different strategies under this condition. His choice of strategy in decision –
making reveals his attitude based on his experience and the influence or persuasion of the people in his
environment. The following strategies are applicable under this condition:
Decision Making
a. Maximax Strategy – use to select the maximum payoff for each alternative,
and then among the maximum payoffs choose the alternative with the
maximum payoffs. That is, the decision maker has to choose the alternative
with the maximum among the maximum payoffs. Hence, it is called
“maximax” which is also known as the “optimistic approach”.
Remark: In case the problem is cost minimization, to apply the preceding strategy, one has to choose the
minimum payoff for each alternative, then finally select the minimum. Thus, this is called “minimin” for
cost.

c. . simply do the same then finally. is the coefficient or index of optimism. Hurwicz Criterion of Realism – the strategy is a middle ground criterion between the maximax and maximin. Decision – Making Under the Condition of Risk . d. which will be having a special feature on function keys. choose the alternative with the minimum average payoff. we cannot do away with opportunity loss.the decision maker does not know exactly which one among the different states of nature will occur but can estimate that any one state will occur. Remark: In case the problem is minimization of cost. For Profit: Regret Value = Highest Column Entry – Every column Entry For Cost: Regret Value = Every Column Entry – Lowest Column Entry Remark: For payoff tables with states of nature written in columns and alternative written in rows. and finally select the minimum among the maximum regrets. What we would “regret” losing can be determined for alternative payoff values. Remark: For cost problem. choose the alternative that would minimize the maximum regret. that is. The decision maker can assign a value for coefficient or index of optimism denoted by the Greek letter (alpha) with a value between 0 and 1. Maximin Strategy – used to select the minimum payoff for each alternative then among the minimum payoffs choose the alternative with the highest payoff. 3. simply determine the average payoff for each alternative and then choose the alternative with the maximum average payoff (profit). the decision maker has to choose the alternative with the maximum among the minimum payoffs. To compute the value. e. this is called “minimax” for cost. Thus. Remark: Choose the maximum value of MR for profit (max MR) and minimum value of MR for cost (min MR).to apply the strategy. first determine the regret payoff for each alternative under each state of nature. . Laplace Strategy – used to select the alternative with the maximum average payoff. that means that he is optimistic about nature. then finally choose the minimum. one has to select the miximum payoff for each alternative. . to apply the preceding strategy.he is making the assumptions on the probability of the occurrence of each state of nature. Thus. Minimax Regret Strategy In business. One particular concern of the company has something to do with the keyboard that will be used in the system. Measure of Realism: Where. between optimism and pessimism. then determine the maximum regret or loss payoff for each alternative.b. Example: Consider the following problem: RGV Company is planning to manufacture its own new PC – based system.using this strategy. If a person assigns as 1. it is called “maximin” which is also known as the “pessimistic approach”. which intends to be marketed by next year under its own brand. . In other words.

If the condition of economy is HIGH. In the absence of information about the likelihood or probability of occurrence on each alternative. the best decision is to buy from local manufacturer. The company can buy the keyboards from a local manufacturer. The profit contribution is in thousand pesos. Alternatives Future Sales Level Low Moderate High Manufacture -30 20 110 Buy from Local 20 60 50 Buy from Japan 10 45 80 Note: The Decision maker has to consider the different costs and benefits of each of the given alternatives. However. which of the three alternatives would you recommend? Note: The answer to the above question depends on the attitude of the decision maker on how he perceives each alternative under the economic demand situation. The following are the decisions of the management applying the above mentioned strategies: 1. Alternatives Maximum Manufacture 110 Buy from Local 60 Buy from Japan 80 . Buying from local supplier would be better if demand will not be as high as expected. then finally select the maximum. b. The payoff table is given below. buying from Japan would be of high quality but the problem might be in delivery schedule. 2. the decision of the management will be as follows: a. The company can manufacture its own unique keyboard. choose the maximum for every alternative. Manufacturing would require major investment in the new production paraphernalia. Decision – Making Under the Condition of Uncertainty Considering the payoffs under each alternative and the different economic situations. If the condition of economy is MODERATE. Applying the Maximax Strategy. the best decision is to buy the units from the local manufacturer. The company can buy the keyboards from Japan. c. the best decision is to manufacture. He has to assess the advantages and disadvantages of each.The following are the different decision alternatives identified by the management: a. Decision – Making Under the Condition of Certainty If the management is certain that the condition of economy is LOW.

c.33 Buy from Local 130 43. select the maximum.Decision: Since the maximum among the three maximum values refers to the alternative of manufacture. choose the minimum of each alternative then finally.33 Buy from Japan 135 45. Alternatives Minimum Manufacture -30 Buy from Local 20 Buy from Japan 10 Decision: Since the maximum among the minimum payoffs refers to the decision on buying from local suppliers. get the average payoff for each alternative.40)(-30) = 54 Buy from Local = (0.00 Decision (based on the table): The maximum average refers to the third alternative. Alternatives Row Total Row Average Manufacture 100 33.40)(10) = 52 Decision (based on the result of computation): The best alternative is to manufacture. which is to buy the units from Japan d.40)(20) = 44 Buy from Japan = (0. .60)(110) + (0. Alternatives Future Sales Level Low Moderate High Manufacture -30 20 110 Buy from Local 20 60 50 Buy from Japan 10 45 80 The succeeding shows the row total (getting the sum of each alternative from different sales level and row average. therefore RGV Company will manufacture its own unique keyboard. b. and finally choose the maximum. therefore RGV Company will buy the keyboards from local suppliers.60)(60) + (0. Applying the Laplace Strategy. Applying Hurwicz Strategy with α= 60% as coefficient of realism: Manufacture = (0.60)(80) + (0. Applying the Maximin Strategy. row total divided by the total number of columns).

That is. which is not fitted to the actual economic situation. Thus. the mathematical expected value is computed as EV = P(X). he has to decide the alternative that can give the best payoff to him.makes use of the concept on regret or known as opportunity loss. Decision Making Under Condition of Risk Decision – making under the condition of risk needs probability estimates of the occurrence of each state of nature. To employ this strategy.e. Applying the Minimax Regret Strategy . the decision of the RGV company should be based on the alternative with the minimum regret. compute the opportunity loss of each alternative under each economic condition by simply getting the difference of the highest entry for each column and entry on each column Alternative Future Sales Level Low Moderate High Manufacture -30 20 110 Buy from Local 20 60 50 Buy from Japan 10 45 80 REGRET TABLE Alternatives Future Sales Level Maximum Regret Low Moderate High Manufacture 50 40 0 50 Buy from Local 0 0 60 60 Buy from Japan 10 15 30 30 The table reveals that if we only know that the future sales will be LOW. the best decision alternative is also to buy from local suppliers. thus. Finally. evaluating the values under the Maximum Regret column. In case an event has several possible outcomes with probabilities. let P represent the probability of an event and X represent the amount of money to be received for that particular event. -to apply the strategy. Hence. In symbols. or expected value (EV) has to be computed by getting the product of the probability of an event and the amount to be received upon the occurrence of that event. it would be best for the company to manufacture its own units. applying the Minimax Strategy. denoted by . However. If the future sales will be MODERATE. the mathematical expectation (ME). The decision maker has to compute the chance of each event based on the existing information and has to compute the expected value of each alternative or what he can expect to get for every alternative considering the different states of nature. 3. the minimum is 30 indicating that the best decision under this strategy is to buy the units from Japan. we are to analyze the payoff in each alternative if in case we have committed a decision. once the sales demand will be high. the best decision is to buy from local suppliers.

15)(110) = 21. they know which decision will be the best for the company. X 1 2 n then the expected value for this alternative would be the sum of the products of the expected value on each state of nature computed based on the formula given: EV  P  X   P ( X )      P  X  1 1 2 2 n n Example: Considering the same problem. Buy from Local 3. definitely. Manufacture 2. Buy from Japan Expected Value EV = (0.P1 . However. using this strategy.25)(20) + (0. Remark: Most of the time managers make their decision under this strategy that is why their decisions are always associated with risk. Pn and if X denotes a discrete variable which can be represented by X . determine what alternative should be chosen by the decision maker under this strategy. the best decision is to buy from local supplier since it has the highest expected value.25)(10) + (0.0 EV = (0.    .60)(60) + (0. Consider the table below showing the assessed probability on each state of nature. the information that they can get from different sources can help lessen this risk.15)(80) = 41.60)(45) + (0. X . The payoff increases as we determine the certainty of a particular expected value of perfect information. P2 . Alternatives Future Sales Level Low Moderate High Manufacture -30 20 110 Buy from Local 20 60 50 Buy from Japan 10 45 80 Note: The payoff amount is in thousand of pesos.15)(50) = 48.5 EV = (0. State of Nature Low Moderate High Probability 25% 60% 15% The expected value for each alternative can be computed as follows: Alternatives 1.25)(-30) + (0. The Expected Value of Perfect Information (EVPI) Note: If the managers are able to determine which among the states of nature will occur.    .60)(20) + (0.5 Thus. The certainty of a particular expected value of perfect information can be determined using the following methods: .

5 – 48. The difference is the value of perfect information. The following are the steps to compute the value of perfect information using Method 1: 1.25) + (60 x 0. EVPI = Expected Value Under Certainty – Maximum EV Under Risk Example: Consider again the preceding example: Best Payoff Under Each State of Nature State of Nature Low Moderate High Probability 25% 60% 15% Manufacture Buy from Local 110 20 60 Buy from Japan (20 x 0.60) + (110 x 0.5 and thus. 4. EVPI is the smallest expected value. 2.15) = 57.0 (EVPI) Method 2: Find the expected regret value for each alternative then. 57.Method 1: Take the difference between the expected payoff under certainty and the expected payoff under condition of risk. Example: Consider again the preceding problem under the Minimax Regret Strategy Alternatives Future Sales Demand Maximum Regret Low Moderate High Manufacture 50 40 0 50 Buy from Local 0 0 60 60 Buy From Japan 10 15 30 30 State of Nature Low Moderate High Probability 25% 60% 15% . 3. Multiply the best payoff under each state of nature to its probability. Subtract the highest payoff of the result in decision – making under the condition of risk from the sum of the combined weights.5 EV under certainty The expected payoff under condition of risk is 48. Get the sum of these combined weights.5 = 9.

5 0(0.25 Moderate P = 0. Manufacture: 2. Maximin c.25) + 0(0. which refers to our EVPI.0 10(0. 3.0 Conclusion: Based on the computation.15) = 36.15) = 16.500.500. the lowest expected regret value is 9. Alternatives Approve Disapprove Renew the contract Php 750. and National. Minimax Regret Strategy b. Sharp. Wash Me Laundry Center must decide to purchase one of the three washing machines displayed in XYZ Appliance Center namely. Hurwicz (assume α=75%) 2. The performance of each machine differs because of the technology applied by its . Steve. Next year. Alternatives Future Sales Level Expected Value Low P = 0. A business consultant has estimated the present value of Parreños Grill Restaurant as shown below for the two alternatives. Laplace d. the lease contract of Parreños Grill Restaurant is to expire. LG. must decide whether to renew the contract for another five (5) years or relocate near the site of a proposed restaurant.15) = 9.0.60) + 0(0.60) + 30(0.60 High P = 0. Exercise Set: 1.60) + 60(0.The following are the steps to compute the value of perfect information using Method 2: 1.000 What course of action would you recommend using each of the following strategies? a. Maximax e. Multiply each entry by the probability value of each state of nature of every alternative. Each of these three machines can wash clothes at different speed. Get the sum of these combined weights. the owner. The City engineer is currently confused on the merits of granting approval to the restaurant.25) + 15(0.15 Manufacture 50 40 0 Buy from Local 0 0 60 Buy from Japan 10 15 30 Solution: 1.25) +40(0. Buy from Local: 3.000 Relocate the restaurant Php 6.000 Php 300.000 Php 4. 2. Buy from Japan: 50(0. The lowest expected regret value is the EVPI.

Square . . Remark: The decision maker has the option on what to choose if it is a square. Compute the expected value of each brand. 2. b. the occurrence of event is beyond his control. or low. Machines High Medium Low 45% 30% 25% Sharp 350 200 -100 LG 300 250 -60 National 450 150 -70 a.manufacturer. Survey also reveals that the likelihood profit for each alternative is summarized in the given table below. Values are in thousands of pesos. customer’s demand for the center’s service will be high.a schematic model of decision – making showing the available alternatives for the decision – maker including the possible consequences. Based on the marketing survey result. medium.represents a decision point. .represents a chance event. Circle . however. which of the three brands will be the best for Wash Me Laundry Center to purchase? Decision Tree Analysis . Types of Nodes that can be used in Making the Tree Diagram 1.a systematic way of analyzing problems is through the use of a diagram. when it comes to circle. Based on the expected value approach.