عنوان مقاله [English]
One of the classical paradigms in decision making models is to develop a model with deterministic values for input data. However, in practical cases, inaccurate data may lead to either impractical decisions or weak solutions. The Analytical Hierarchy Process (AHP) is one of the most widely used techniques in Multi Criteria Decision Theory. Since, in AHP, interval judgments can be used to deal with uncertainty in a model, a robust model is required to extract appropriate weights from the corresponding interval comparison matrices. In this paper, two robust models are presented to reduce inconsistency and errors in the interval comparison matrix. The first model is based on robust optimization with a budgeted uncertainty set that generates point estimation weights. The second model is based on goal programming that extracts interval weights from interval comparison matrices. Hopefully, both models can be converted to linear programming problems. Experimental results on both models show that our proposed models can extract appropriate and robust solutions for comparison matrices of AHP under different circumstances.