# I-Kaz methodology for predicting tool life of AlCrN—Coated WC—Co inserts in the machining of AISI 304 steel

## TL;DR

This paper introduces a new method using a sensor and statistical feature to monitor tool wear in machining stainless steel, improving accuracy and efficiency.

## Contribution

The first use of a Microflown PU sensor with the I-Kaz™ feature for real-time tool wear prediction in milling.

## Key findings

- The I-Kaz™ method achieved R2 > 0.96 in predicting flank wear with high accuracy.
- AlCrN-coated inserts significantly reduced flank wear compared to uncoated tools.
- Optimal cutting parameters minimized surface roughness and tool wear.

## Abstract

This study proposes a real-time tool wear monitoring approach for dry milling of AISI 304 stainless steel using a Microflown PU sensor and the I-Kaz™ statistical feature. A Taguchi L18 orthogonal array was adopted to optimize cutting speed, feed rate, depth of cut and tool type (uncoated and AlCrN-coated WC–Co inserts) based on surface roughness and flank wear responses. ANOVA revealed that tool type was the most significant factor, contributing approximately 85 % to flank wear variation, followed by depth of cut and cutting speed. The optimal combination: AlCrN-coated insert, cutting speed of 1250 rpm, depth of cut of 0.50 mm and feed rate of 0.04 mm/rev minimized both responses. The proposed I-Kaz-based monitoring approach established a strong inverse power-law relationship between the I-Kaz coefficient and flank wear Z2α =a(VB)-n, achieving R2>0.96, indicating high accuracy and stability across repetitions.•This study introduces the first application of a Microflown PU sensor for near-field acoustic monitoring in milling operations.•The I-Kaz™ feature is demonstrated as a computationally efficient and accurate method for real-time tool wear prediction.•The research integrates cutting parameter optimization and predictive monitoring within a single experimental framework.

This study introduces the first application of a Microflown PU sensor for near-field acoustic monitoring in milling operations.

The I-Kaz™ feature is demonstrated as a computationally efficient and accurate method for real-time tool wear prediction.

The research integrates cutting parameter optimization and predictive monitoring within a single experimental framework.

Image, graphical abstract