Paper Title
A Robust Optimization Model for Multiple Responses Based On Fuzzy Logic Reasoning Method
Abstract
As the traditional parameters optimization process less consider the influence of noise factors, and the multiple
response dimension reduction has limitations in weight setting. This paper proposes an improved robust optimization model
based on fuzzy logic reasoning and neural network. Firstly, using the desirability function to calculate the desirability value
of each responses, and converts the desirability value into a single fuzzy reasoning grade(FRG), then selected the optimal
parameter combination by the main effect chart. Secondly, constructing the BP Neural network prediction model of multiple
quality characteristic including controlled factors and noised. Finally, applying this method to optimization the surface
roughness in end milling process, the two responses are all get optimization and the S/N ratio is improved, so the
effectiveness of this method is to be verified.
Keywords: Fuzzy logic reasoning, Desirability Function, BP neural network, Multiple responses, Robust optimization.