Statistical analysis, modeling and multi-criteria optimization using MOOSRA, COPRAS, MABAC, and DF approaches for eco-efficient and productive machining of Inconel 718 under dry and MQL conditions

Abstract Inconel 718 is a critical nickel-based superalloy; however, its machinability is challenging owing to low thermal conductivity and a pronounced tendency to work-harden. While lubrication generally improves machinability, alternatives to conventional flood cooling are actively sought to mitigate its environmental and practical limitations. This study presents the findings of an experimental investigation on the turning of Inconel 718. A Taguchi L9 orthogonal array was employed, with the lubrication strategy (Dry, Minimum Quantity Lubrication (MQL) with soluble oil, MQL with pure oil), cutting speed (CS), feed rate (fr), and depth of cut (DoC) as control variables. A statistical analysis was performed, followed by the modeling of key performance indicators: surface roughness parameters (SR Ra and SR Rz ), main cutting force (Ft), energy consumption (Ec), and material removal rate (Q). Analysis of variance (ANOVA) and regression analysis identified the feed rate as the predominant factor influencing surface roughness, whereas the depth of cut exerted the strongest effect on cutting force and energy consumption. Furthermore, MQL with pure oil significantly reduced both surface roughness and cutting forces compared to dry machining. Highly accurate linear regression models (with R 2 values exceeding 0.96) were developed to predict roughness, forces, and energy consumption, thereby validating the relationships between the input parameters and machining outcomes. A comparative multi-objective optimization was conducted using the Desirability Function (DF), COPRAS, MOOSRA, and MABAC methods. The results reveal distinct trade-offs: COPRAS and MABAC prioritized the minimization of constraints (quality and energy consumption) at the expense of productivity (Q). In contrast, MOOSRA favored maximum productivity despite higher values for the other parameters. The Desirability Function approach provided a well-balanced compromise. This work is pertinent for researchers engaged in MQL lubrication techniques and those interested in the comparative performance of multi-criteria decision-making (MCDM) methods in machining optimization.