Prediction of surface roughness in low speed turning of AISI316 austenitic stainless steel

Gabriel Medrado Assis Acayaba, Patricia Munoz de Escalona

Research output: Contribution to journalArticlepeer-review

39 Citations (Scopus)


Surface roughness is an important quality in manufacturing, as it affects the product's tribological, frictional and assembly characteristics. Turning stainless steel at low cutting speeds may result in a rougher surface due to built up edge formation, where as speed increases the surface roughness improves, due to the low contact time between the chip and the tool to allow bonding to occur. However, this increase in cutting speed produces higher tool wear rates, which increases the machining costs. Previous studies have indicated that savings in cost and manufacturing time are obtained when predicting the surface roughness, prior to the machining process. In this paper, experimental data are used to develop prediction models using Multiple Linear Regression and Artificial Neural Network methodologies. Results show that the neural network outperforms the linear model by a fair margin (1400%). Moreover, the developed Artificial Neural Network model has been integrated with an optimisation algorithm, known as Simulated Annealing (SA), this is done in order to obtain a set of cutting parameters that result in low surface roughness. A low value of surface roughness and the set of parameters resulting on it, are successfully yielded by the SA algorithm.
Original languageEnglish
Pages (from-to)62-67
Number of pages6
JournalCIRP Journal of Manufacturing Science and Technology
Early online date9 Sep 2015
Publication statusPublished - Nov 2015


  • stainless steel
  • turning
  • surface roughness
  • multiple linear regression
  • artificial neural networks
  • simulated annealing


Dive into the research topics of 'Prediction of surface roughness in low speed turning of AISI316 austenitic stainless steel'. Together they form a unique fingerprint.

Cite this