Modeling uncertainty in multi-criteria decision analysis

his paper provides a review of multiple criteria decision analysis (MCDA) for cases where attribute evaluations are uncertain. The main aim is to identify different tools which can be used to represent uncertain evaluations, and to broadly survey the available decision models that can be used to support uncertain decision making. The review includes models using probabilities or probability-like quantities; explicit risk measures such as quantiles and variances; fuzzy numbers, and scenarios. The practical assessment of uncertain outcomes and preferences associated with these outcomes is also discussed.

► A review of multiple criteria decision analysis under uncertainty. ► Reviews probabilities, decision weights, risk criteria, fuzzy sets and scenarios. ► Reviews both preference models and practical assessment issues. ► Identifies recent trends and scope for future research.