Conceptual Design Evaluation Using Interval Intuitionistic Fuzzy-Z-Number for Multiple Uncertain Information from Decision-Maker
Jin Qi,Jie Hu,Yinghong Peng
DOI: https://doi.org/10.1007/s40815-023-01559-1
IF: 4.085
2023-07-10
International Journal of Fuzzy Systems
Abstract:In new industrial product development, human-involvement conceptual design evaluation (CDE) is an information-intensive and multi-criteria group decision-making (MCGDM) process. To capture and formalize the uncertain information from decision-makers (DMs), several derivations of fuzzy number such as triangular intuitionistic fuzzy number (TIFN), Z-number, and interval number have been employed in CDE, which only cover limited information. The main contributions of this paper are as follows: firstly, a new kind of fuzzy number by taking advantage of TIFN, Z-number and interval number, i.e., interval intuitionistic fuzzy-Z-number (IIFZN), is presented to utilize DM's preference value (PV) for alternative, as well as his important rating (IR) and familiarity degree (FD) judgements for evaluation criterion. IIFZN is composed of not only the membership and non-membership of PV or IR, but also the affiliated confidence coefficient transformed from FD. Then, in terms of the IIFZN-based decision-making matrix, this study proposes new ideal solution definition rules and distance measurement metric in technique for order preference by similarity to ideal solution (TOPSIS) to form a new TOPSIS-IIFZN method. The proposed method will enable DM easily find the alternative which is most preferable for important and familiar criteria and least preferred for less important and unfamiliar criteria as the best one. It is more rational than traditional PV-only CDE method, which will help company to save manufacturing resources. Besides theoretical research, one practical example and three more complex comparative experiments have been carried out to validate the performance of TOPSIS-IIFZN. By comparing with other MCGDM methods using different types of fuzzy numbers, it proves that TOPSIS-IIFZN can obtain more reasonable evaluation results. Meanwhile, besides PV, the RI and FD factors also play great influences in CDE.
computer science, information systems,automation & control systems, artificial intelligence