Case retrieval(CR) is critically an important part of case-based design. However few studies attempt to research CR for customer-driven design and analyze the eect of other production factors besides similarity comput...Case retrieval(CR) is critically an important part of case-based design. However few studies attempt to research CR for customer-driven design and analyze the eect of other production factors besides similarity computation. This paper proposes a new CR method for customer-driven design,and requirement-weighting analysis. Fuzzy set theory are integrated into CR process to deal with the fuzzy and imprecision customer requirements. So the proposed method is called weighted fuzzy case retrieval(WFCR). Furthermore,similar case evaluation is added into CR process to demonstrate that the best case is selected not only on the basis of satisifaction for the given requriements,but also on the degree of preference over other cases according to multiple evaluation criteria. Meanwhile,WFCR system is developed and applied to power transformer design to validate its scientificity and practicality. Finally,the paper statistically validated the supriority of WFCR by comparing it with traditional fuzzy case retrieval methods(FCRs),and the comparison indicates that WFCR is more accurate than other FCRs.展开更多
基金the National Natural Science Foundation of China (Nos. 50775140, 50575142 and 60304015)the National High Technology Research and Development Program (863) of China (No. 2008AA04Z113)+1 种基金the National Basic Research Program (973) of China (No. 2006CB705400)the Shanghai Committee of Science and Technology (No. 08JC1412000)
文摘Case retrieval(CR) is critically an important part of case-based design. However few studies attempt to research CR for customer-driven design and analyze the eect of other production factors besides similarity computation. This paper proposes a new CR method for customer-driven design,and requirement-weighting analysis. Fuzzy set theory are integrated into CR process to deal with the fuzzy and imprecision customer requirements. So the proposed method is called weighted fuzzy case retrieval(WFCR). Furthermore,similar case evaluation is added into CR process to demonstrate that the best case is selected not only on the basis of satisifaction for the given requriements,but also on the degree of preference over other cases according to multiple evaluation criteria. Meanwhile,WFCR system is developed and applied to power transformer design to validate its scientificity and practicality. Finally,the paper statistically validated the supriority of WFCR by comparing it with traditional fuzzy case retrieval methods(FCRs),and the comparison indicates that WFCR is more accurate than other FCRs.
