ABSTRACT
In this chapter solution methods for decision problems that have one evaluation criterion are explained. When there is no uncertainty regarding the outcome values, the solution approach is Single Criterion Optimization method. When there is uncertainty about the future outcomes of the decision to be made, several solution methods are available based on the nature of the problem. Laplace method assumes that all the future states (i.e., scenarios) of problem nature have equal probabilities of occurrence and the criterion is to be maximized. The Max-Min method assumes that the decision-maker (DM) is very pessimistic in outlook and just wants to choose the best alternative by considering only the worst possible outcomes overall the states of nature of each alternative. The Min-Max Regret method assumes that the DM wants to minimize the maximum opportunity loss where the opportunity loss is defined as the difference between the maximum value of a criterion and the achieved value of that criterion for an alternative. In the Expected Value method, it is presumed that the DM can determine the probability of occurrence for each state of nature or scenario. The preferred alternative is that alternative that has the maximum expected value. When the problem is complex enough to have multiple stages involving a sequence of decisions and outcomes, the use of the Decision Tree method is appropriate.
