6. ConclusionsThe ANFIS with multi inputs fuzzy model was developed topredict the flank wear during turning process. The inputs of thedeveloped model are cutting parameters of turning process (cutting speed, feed rate, depth of cut), and I-kaz coefficient for cuttingforce and feed force. The cutting force components were measuredusing in-house developed strain gauge sensor. The I-kaz methodwas developed based on the decomposed frequency signals, andintegrated kurtosis-based algorithm was used to extract the features of the cutting force and feed force signals. The changes ofcutting force signals due to flank wear were indicated by significant increasing the I-kaz coefficient values. Among the five inputparameters of ANFIS model, changes of the I-kaz coefficient in feedforce have the most effect on predicting flank wear value, followedby I-kaz coefficient in cutting force, feed rate, depth of cut, and cutting speed. It was found that the results generated by the ANFISmodel are close to the experimental results with the minimumand maximum average error of the flank wear of about 2.30% and5.08% respectively. The accuracy of the prediction may achieve upto 95.93–97.70%. The performance of the prediction shows that theestimated results are very accurate and encouraging to be appliedin real industry application.