Investigation of the effect of process parameters on surface roughness in drilling of particleboard composite panels using adaptive neuro fuzzy inference system

Particleboard wood composites are immensely used for many general and manufacturing applications. In this study, an analysis of various machining conditions has been performed to obtain good surface quality in the hole making of particleboard by varying the input parameters. The surface roughness (R a ) values obtained are ranging from 6.03 to 28.32 µm, and the minimum value is achieved at a higher speed, lower feed, and smaller point angle combinations. From ANOVA analysis, it has been observed that the model developed is adequate, and the influence on surface roughness is strong for feed (56.68%) followed by a point angle (28.42%) and then speed (9.37%). Mathematical models have been developed using two different criteria such as response surface methodology (RSM), adaptive neuro-fuzzy inference system (ANFIS) and compared for their effectiveness. The coefficient of determination (R 2 (R-Sq)) values of 98.5% (RSM) and 99.9% (ANFIS) indicates that the models are useful to predict R a of particleboard. The average checking error percentage (0.20098) has been obtained for the ANFIS model trained using 'gaussmf' membership function with 100 epochs.