ABSTRACT
In many real-life decision situations, the decision-maker does not have access to the amount of information demanded to come up with precise numerical values and probabilities, nor have the ability to make precise and accurate estimates of values. In this chapter, we will see how we can circumvent two types of uncertainties. The first type derives from a lack of historical data and includes statistical variation, subjective judgments, linguistic imprecision, variability, inherent randomness, disagreement, and approximation. For example, in experiments, errors in the measurements of quantities give rise to statistical variation. The second type arises from the model chosen, for example, how to reasonably, correctly represent and express a utility function. Furthermore, uncertainty due to biases in communication and value differences is unavoidable.
