In decision-making in some cases decision makers cannot or do not want to specify preferences with the exact values. To decide with these stochastic data Stochastic Multi-Criteria Acceptability Analysis (SMAA) is an effectively implemented decision support tool. Grey Relational Analysis (GRA) which is working only with deterministic data, is an alternative and a popular method for multi criteria decision making problem. In this study, two new methods SMAA-GRA and SMAA-DEMATEL-GRA are proposed: combination of SMAA-2 and GRA methods and also combination of SMAA-2 and DEMATEL-GRA. The aim of the article is to provide GRA cope with vague and imprecise data in other words, to establish stochastic GRA. And also with DEMATEL we can take into account relationship criteria with each other in decision making process. The proposed methods are applied to both drug benefit risk analysis problem in literature and a real life problem. The study shows that SMAA-GRA and SMAA-DEMATEL-GRA whose results are significant and consistent GRA and DEMATEL methods could be used with ambiguous and arbitrarily distributed data for weights and criteria measurements.