In this paper, a novel data envelopment analysis (DEA) model is proposed for ranking the association rules. In this regard, a mixed-integer linear programming (MILP) model is proposed to determine the most efficient association rules where, an N-person bargaining game is used to create an interactive competition between the existing N-weights to get a better ranking. In addition, the proposed model is fuzzified by setting the ambiguous threshold of the indicators' weight in each rule to improve the overall ranking of the rules. Finally, the risk, resilience and decongestion factors are also considered to increase the responsiveness of the models to different real-world conditions. The proposed model is validated by some random problems and an illustrative example of market basket analysis where, the proposed model shows better results than the competing models in the literature. In addition, the applicability of the proposed model is illustrated using a real case-study. [Received: 2 February 2020; Accepted: 5 July 2020]