Knowledge Service Technology for Supporting Intelligent Product Design

Table 1

Related research literature.

Related research literature	Method advantages and disadvantages
Zhang and Yu [7] proposed two distance measures of intuitionistic and interval-valued fuzzy sets. Sharaf-El-Deen et al. [8] proposed a new hybrid case-based reasoning approach for medical diagnosis systems. Qian et al. [9] proposed an improved requirements-driven self-adaptation approach that combines goal reasoning and case-based reasoning. Gu et al. [10] proposed a CBR system based on the weighted heterogeneous value distance metric. Tan et al. [11] described the hesitant fuzzy Hamacher aggregation operators for multicriteria decision-making. Song et al. [12] proposed a general type of similarity measure for an intuitionistic fuzzy set with two parameters. Mahmood et al. [13] proposed the notion of complex hesitant fuzzy hybrid vector similarity measures. Feng and Khan et al. [14, 15] proposed a case library optimization method to improve detection efficiency.	The existing research on similarity calculation investigations has improved the accuracy of case retrieval and has made the retrieval process more reasonable. The proposed methods have the following advantages: (1) they enhance the ability of a CBR system to process nonlinear data; (2) they overcome the problem of difficulty in obtaining decision-making knowledge in cases of discontinuous information; (3) they can handle continuous and discrete attributes simultaneously.
	However, some of the methods have certain limitations, which are as follows: (1) the classification of attribute types is not detailed enough; (2) the research on the similarity calculation of fuzzy attribute sets is not deep enough.
Feng et al. [16] combined the genetic algorithm and the particle swarm optimization algorithm in the calculation of attribute feature weights. Xiong et al. [17] have concluded that the normalization factor directly determines the influence of the corresponding attributes on the filtering result. Ahmadvand and Pishvaee [18] proposed a credibility-based fuzzy common weights data envelopment analysis approach. Li et al. [19] developed an integrated cumulative prospect theory based on the hybrid-information multicriteria decision-making approach. Lin et al. [20] reported that the traditional TODIM method could be extended to handle the HFLTSs based on the novel comparison function and distance measure. Chen et al. [21] proposed a two-stage logic scoring of the preference-ELECTRE III-based approach. Sun et al. [22] proposed an iterative weight update method based on the optimistic coefficient.	The existing research on multiattribute decision-making not only can improve the matching accuracy but can also avoid excessive reliance on decision-makers to set the weight value subjectively. The proposed methods have the following advantages: (1) they consider the psychological behavior of decision-makers; (2) the reasoning logic pattern in the human cognition process and the compensation among attributes are taken into account; (3) multiple methods are included to deal with multiattribute decision-making problems.
	However, some of these methods have certain limitations: (1) it takes a long time to find an optimal solution; (2) the coefficients of the attribute weight set are not continuously and dynamically updated.