Multi-performance optimization of multi-roller burnishing process in sustainable lubrication condition

Sustainable machining processes are efficiently achieved using the selection of optimal parameters. In this study, the minimum quantity lubrication-assisted multi-roller burnishing (MQLAMRB) operation is proposed and optimized to reduce the total energy consumption (TE), mean roughness depth (MR), and roundness deviation (RN). Burnishing parameters are the burnishing speed (BS), depth of penetration (DOP), the quantity consumed of the lubricant (QO), and the pressure value of the compressed air (PA). The embodied energy of the lubricant (Eel) and burnishing tool (Eeb) are developed and integrated into the TE model. The artificial neural network (ANN) model of the energy consumption in the burnishing time (Ebo), MR, and RN is proposed regarding the MQLAMRB parameters. The best-selected solution is determined using an efficient glowworm swarm optimization (GSO) algorithm and the TOPSIS. The results indicated that the 4–25-21-25-3 ANN structure effectively used to construct the MQLAMRB performances. The optimal outcomes of the BS, DOP, QO, and PA are 94 m/min, 0.12 mm, 130 ml/h, and 0.7 MPa, respectively. Moreover, the TE, MR, and RN are decreased by 12.2%, 14.2%, and 42.5%, respectively. The reductions in the MR and RN of the burnished surface are 90.23% and 88.18%, respectively, as compared to the pre-machined conditions. © 2021 Taylor & Francis.