Imprecise data envelopment analysis model for robust design with multiple fuzzy quality responses.

In this research, Imprecise Data Envelopment Analysis (IDEA) model wasutilized to improve fuzzy multiple responses in robust design. The combinationof process factor levels at each experiment was considered as a DecisionMaking Unit (DMU) with responses treated as inputs and outputs for allDMUs. The Fuzzy C-Means Clustering (FCMC) technique is used to fit theresponse fuzziness by clustering the average values, relative to each response,into a suitable number of clusters with triangular / trapezoidal membershipfunctions. IDEA models were used to estimate the fuzzy triangular / trapezoidalefficiency values for each DMU. Finally, the preference degree-based rankingapproach was used to discriminate between the fuzzy efficiency valuesand identifying the best combination of factors levels that would improvefuzzy multiple responses. Two case studies are utilized to illustrate the proposedapproach, including optimizing wire electrical discharge machining andsputtering process parameters. The results showed that the proposed approachprovides better anticipated improvements than the fuzzy multipleregression based approach. This approach would provide great assistant toprocess engineers in improving process performance with fuzzy multipleresponses over a wide range of business applications. [ABSTRACT FROM AUTHOR]