An advanced fuzzy approach for modeling the yield improvement of making aircraft parts using 3D printing

Abstract

Forecasting the yield is critical for making aircraft parts using three-dimensional (3D) printing. However, the existing methods for the yield forecasting exhibit a common problem: they employ the logarithmic or log-sigmoid value, rather than the original value, of the yield to simplify the computation. To address this problem, an advanced fuzzy approach was proposed in this study. The focus of this study is to investigate the effectiveness of the advanced fuzzy approach for forecasting the yield of a 3D-printed aircraft part. The advanced fuzzy approach derived the direct-solution (DS) versions of the existing fuzzy yield learning models. These DS versions use the original yield value directly, thereby optimizing the forecasting performance. The proposed methodology was applied to the process of making an aircraft part using 3D printing to evaluate its effectiveness. Experimental results revealed significant improvements in the forecasting accuracy of the proposed methodology compared with the aforementioned methods. Furthermore, when the proposed methodology was applied to various fuzzy yield learning models, different improvements were obtained.