Title: A new DEA model for ranking association rules considering the risk, resilience and decongestion factors
Authors: Majid Khedmati; Ardavan Babaei
Addresses: Department of Industrial Engineering, Sharif University of Technology, P.O. Box 11155-9414 Azadi Ave., Tehran 1458889694, Iran ' Department of Industrial Engineering, Sharif University of Technology, P.O. Box 11155-9414 Azadi Ave., Tehran 1458889694, Iran
Abstract: 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]
Keywords: ranking association rules; data envelopment analysis; DEA; fuzzy logic; mixed-integer linear programming; MILP; game theory; risk; resilience.
European Journal of Industrial Engineering, 2021 Vol.15 No.4, pp.463 - 486
Received: 02 Feb 2020
Accepted: 05 Jul 2020
Published online: 12 Jul 2021 *