Title: Artificial neural networks and case-based reasoning models for predicting tool life and tool-shim interface temperature

Authors: M. Anthony Xavior; S. Margret Anouncia

Addresses: School of Mechanical and Building Sciences, VIT University, Vellore – 632014, Tamil Nadu, India ' School of Computer Sciences and Engineering, VIT University, Vellore – 632014, Tamil Nadu, India

Abstract: The objective of this paper is to develop artificial neural networks (ANN) and case-based reasoning (CBR) models to predict the tool life and the tool-shim interface temperature during turning of different alloy steel materials. The tool life of multicoated carbide, cermet and alumina inserts, and the temperature (measured by placing a thermocouple between the tool and shim in the tool holder) under various turning conditions are experimentally determined. Further, the experimental values are used to develop the prediction models based on ANN and CBR. Twenty sets of validation experiments are conducted to evaluate the performance of the prediction models. The prediction models are compared based on the statistical measures such as mean absolute percentage error (MAPE), root mean squared error (RMSE) and the correlation coefficient (R), and it is confirmed that CBR model is superior to ANN model for the machining process considered.

Keywords: artificial neural networks; ANNs; case-based reasoning; CBR; tool life prediction; tool-shim interface temperature; alloy steel; turning; carbide inserts; cermet inserts; alumina inserts; prediction modelling; mean absolute percentage error; MAPE; root mean squared error; RMSE; correlation coefficient.

DOI: 10.1504/IJSOM.2016.076906

International Journal of Services and Operations Management, 2016 Vol.24 No.3, pp.379 - 396

Received: 11 Oct 2014
Accepted: 17 Dec 2014
Published online: 07 Jun 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article