UNCERTAIN DESIGN BASED ON ROBUST DESIGN METHOD WITH AN APPLICATION TO T-PEEL TEST APPARATUS

Abstract

Measurement apparatus can accurately measure against errors to ensure high robustness, which is conventionally achieved through parameter design. This study proposes a robust design method, which is adaptive to any environment and can handle any uncertainty in the environment, by incorporating the set-based design concept into the conventional parameter design to ensure robustness. The proposed method represents design parameters as ranged values, obtains optimum conditions from both conventional and outside level values, and employs an evaluation index to reflect the designer’s intention. The algorithm is represented in five steps: (a) determination of design variables, (b) planning and conduct of an experiment, (c) calculation of signal-to-noise (SN) ratio for each experiment, (d) conduct of a supporting experiment, and (e) application of the set-based design method. The proposed method is applied to a T-peel test apparatus’ design. Owing to the proposed design method’s application, the SN ratio of the proposed method exhibited the feasibility of attaining higher values compared to the SN ratio of the conventional parameter design’s optimum condition. Furthermore, the designer could use the proposed evaluation index to select a range solution reflecting a subjective intention. The proposed method is more robust than the conventional parameter design and can design a measurement apparatus considering the designer’s intention.