A stochastic multicriteria acceptability analysis–evidential reasoning method for uncertain multiattribute decision‐making problems. Issue 4 (14th June 2019)

• Record Type: Journal Article
• Authors: Zhang, Xiaoqi; Gong, Bengang; Yang, Feng; Ang, Sheng
• Other Names: Jiménez Alfredo guestEditor.; Herrero Álvaro guestEditor.
• Is Part Of: Expert systems. Volume 36:Issue 4(2019)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2019-06-14
• Subjects: evidential reasoning -- multiattribute decision making -- stochastic multicriteria acceptability analysis‐2 (SMAA‐2) -- uncertainty; Expert systems (Computer science); 006.33
• DOI: 10.1111/exsy.12426
• Languages: English
• ISSNs: 0266-4720

Abstract

Abstract: Evidential reasoning (ER) is an effective approach for assessing alternatives with uncertain attribute values in the context of decision making. For the ER approach to be able to handle variations in the weights of uncertain attributes in an appropriate manner, this paper proposes a method to solve problems of uncertain multiattribute decision making that involve both uncertain attribute values and uncertain attribute weights, which this method does by combining the ER approach and stochastic multicriteria acceptability analysis‐2 (SMAA‐2). First, the uncertainty in attribute values is described by using a belief decision matrix as in the ER approach. The analytical ER algorithm is then used to create the utility function in the SMAA‐2 model, and that function is used to calculate the probability of different sorting positions of the decision units under weight‐related restrictions. Finally, the results of ranking are obtained by combining the sorting weights. An example is provided to verify the effectiveness of the proposed method.