Modeling and optimization of tribological performance of K usum oil‐based bio‐lubricant using response surface methodology and artificial neural network

Growing environmental concerns about petroleum‐based lubricants have encouraged the development of sustainable bio‐lubricants derived from renewable resources. Plant‐based oils have environmental benefits, but generally, they exhibit less effective tribological characteristics. The current study developed an epoxidized bio‐lubricant enhanced with copper oxide nanoparticles to overcome these limitations. The tribological behavior of the developed lubricant was evaluated by varying the concentration of CuO nanoparticles (0–0.6 wt%) to solve this restriction by mixing Kusum oil with loads (20–40 N), sliding speeds (2.0–3.0 m/s), and distances (500–800 m) while assessing their tribological performance under different conditions. The optimized condition (30 N, 2.5 m/s, 700 m, 0.4 wt% CuO) yielded a very low coefficient of friction (0.0252) and minimal wear (0.0309 mm 3 ). Response Surface Methodology (RSM) and Artificial Neural Network (ANN) modeling were employed to validate the findings. Response Surface Methodology was used to analyze the effects of operational parameters, while ANN modeling yielded greater predictive accuracy, with an R 2 value greater than 0.98. The results highlight the strong potential of CuO‐reinforced Kusum oil as a sustainable and high‐performance bio‐lubricant and validate the effectiveness of combining experimental investigation with data‐driven optimization.