This paper presents the optimization of the surface roughness using the Taguchi technique to assess the machinability of the AISI 316Ti steel with PVD coated carbide inserts under different cooling conditions such as dry, conventional (wet) and cryogenic cooling with liquid nitrogen (LN2). Based on the Taguchi L9 (3 3 ) orthogonal-array design, the machinability tests were made utilizing a CNC lathe machine. Test parameters including the cutting speed, the cooling condition and the feed rate were taken and then the surface roughness (Ra) was measured to obtain the machinability indicator. An analysis of variance was performed to determine the importance of the input parameters for the surface roughness. The process parameters were optimized by taking the Taguchi technique into consideration. The Taguchi signal-to-noise ratio was employed with the smaller-the-better approach to obtain the best combination. On the basis of the first-order model, a mathematical model was created using the regression analysis to predict the Ra model. The results indicate that the feed rate is the parameter with the highest effect on the surface roughness and that the other parameters also have a statistical significance. In addition, cryogenic cooling is an alternative method for increasing the surface quality of machined parts. Keywords: AISI 316Ti, cryogenic cooling, machinability, optimization, surface roughness, Taguchi method lanek obravnava optimiranje hrapavosti povr{ine z uporabo Taguchi-jeve metode za oceno obdelovalnosti jekla AISI 316Ti s karbidnimi vlo`ki s PVD-nanosom v razli~nih razmerah ohlajanja, kot je suho, navadno (mokro) in kriogensko hlajenje s teko~im du{ikom (LN2). Preizkusi obdelovalnosti so bili izvr{eni s CNC-stru`nico na osnovi Taguchi-jevega ortogonalnega niza L9 (3 3 ). Izbrani so bili parametri preizkusov, hitrost rezanja, razmere pri ohlajanju in hitrost podajanja, nato pa je bila izmerjena hrapavost povr{ine (Ra) kot pokazatelj obdelovalnosti. Izvr{ene so bile analize variance, da bi ugotovili pomembnost vhodnih parametrov na hrapavost povr{ine. Procesni parametri so bili optimirani z upo{tevanjem Taguchi-jeve tehnike. Uporabljeno je bilo Taguchi-jevo razmerje signal -hrup s pribli`kom~im manj{e tem bolj{e za doseganje najbolj{e kombinacije. Na osnovi modela prvega reda je bil postavljen z uporabo regresijske analize matemati~ni model za napovedovanje Ra. Rezultati ka`ejo, da je hitrost podajanja parameter z najve~jim u~inkom na hrapavost povr{ine, vsi drugi parametri imajo statisti~no zna~ilnost. Dodatno je kriogensko ohlajanje alternativna metoda za pove~anje kvalitete povr{ine stru`enih delov.