night stars image  

Decision Making Under Uncertainty


Criterion of Realism (Hurwicz) Decision Rule

The Criterion of Realism decision rule is an attempt to make a tradeoff between complete risk indifference (as in the Maximax rule), and total risk aversion (as in the Maximin rule). With this procedure, the decision maker will decisde how much emphasis to put on each extreme.

Tod do this, he must choose a Coefficient of Realism, called alpha (α), which is a decimal number between 0 and 1. This number provides the emphais on the optimistic view. The number (1-α), then, is the amount of emphasis that is placed on the most pessimistic outcome. For example, if a manager chose an alpha of 0.6, he would be placing 60% emphasis on a risky, high return (Maximax-type) outcome, and 40% emphasis (since 1-0.6 = 0.4 or 40%) on a low-risk, pessimistic, (Maximin-type) outcome.

Explained another way, the Criterion of Realism calculates a weighted average of the best and worst outcomes for each alternative, using α and (1-α) as the weights, respectively.

To determine the decision under the Criterion of Realism decision rule, a column is added on the right side of the payoff table. In this column the decision maker must calculate the factor by multiplying the best outcome in the row by α, multiplying the worst outcome in the row by (1-α),and adding the two result together.

Note that the sign on the payoffs is important, and frequently the worst payoff in a row will be a negative number. Make sure you add positive and negative numbers correctly!

From these calculated numbers in each row, identify the highest one, and that will be in the row of the decision alternative to be selected under this rule.

Maximax Payoff Table

In the example above, the manager decided to use an alpha of 0.8, meaning that he wanted to place 80% emphasis on a high-payoff alternative, and so only 20% emphasis on the low-risk alternative. Since the decision alternative Construct a Large Plant has the maximum payoff in the new Criterion of Realism column, it would be selected as the decision to implement.

Do you see how Mr. Thompson got the $124,000 in the Construct a Large Plant row?

He multiplied alpha (0.8) by the best row outcome of 200,000 and got 160,000. He then multiplied (1-α) (1 - 0.8 or 0.2) by the worst outcome in the row, which is -180,000 to get -36,000. 160,000 plus -36,000 gives 124,000.

Now you do it for the Construct a Small Plant row as a check on your knowledge.

Page 6 of 9