The face milling process for hardened steels used in die and mold manufacturing
involves precise parameter settings to achieve high-quality surfaces with high
productivity. However, due to the nonlinear relationship between the machining
parameters, it becomes challenging to optimize the machining process. This paper
focuses on optimizing the face milling parameters for the DIN 2714 hardened
steel by employing experimental approaches and AI-based modeling techniques.
Taguchi’s L9 orthogonal array-based design of experiments with machining
parameters of depth of cut, cutting speed, feed rate, and step-over is employed
for the optimization. The material removal rate (MRR) and surface roughness (Ra)
were considered as output responses.
The results showed that surface roughness is significantly affected by cutting
speed, while depth of cut primarily controls MRR. To improve the precision in
predicting the output responses, two artificial intelligence (AI) based
intelligent models were developed using hybrid modeling techniques—namely,
PSO-FIS and GA-FIS. The experimental results demonstrated that the GA-FIS model
was accurate in predicting the output response for surface roughness (Ra), with
a lower error value (0.0162 μm), while the PSO-FIS model was accurate in
predicting MRR with a high R2 value (0.9997), proving that fuzzy
modeling techniques can be effectively used to optimize machining operations
with high precision.