An experimental investigation was done on a variable compression ratio diesel engine fuelled with blends of Karanja oil methyl ester (B100) and conventional diesel oil (B00) in the proportion of 20:80 (B20), 30:70 (B30), and 40:60 (B40). The performance and emission tests were carried out on diesel engines for three input parameters and six out parameters. The load, compression ratio (CR), and biodiesel blends are used as input operating parameters. The output parameters are brake specific fuel consumption (BSFC), brake thermal efficiency (BTHE), and emissions like Hydrocarbons (HC), Carbon monoxide (CO), Nitrogen oxides (NOX) and Carbon dioxide (CO2). To find the optimal response, a set of experiments was conducted using the design of experiments suggested by Response surface methodology. Design Expert-10 was used to get the set of experiments for optimization. The predicted maximum BTHE (23.21%) and experimental maximum BTHE (23.27 %) were obtained at the optimal operating parameters of load 8 kg, CR 14, blend B20. The predicted minimum BSFC (0.37kg/kWh) and experimental minimum BSFC (0.37 kg/kWh) were obtained at the optimal operating parameters of load 8 kg, CR 14, blend B20. The predicted minimum CO2 (2.64%) and experimental minimum CO2 (2.54%) were obtained at the optimal operating parameters of load 2.63 kg, CR 15, blend B30. The predicted minimum HC (16.68ppm) and experimental minimum HC (16ppm) were obtained at the optimal operating parameters of load 4 kg, CR 16, blend B20. The predicted minimum CO (0.065%) and experimental minimum HC (0.059%) were obtained at the optimal operating parameters of load 8 kg, CR 16, blend B40. The predicted minimum NOx (168ppm) and experimental minimum NOx (131.03ppm) were obtained at the optimal operating parameters of load 2.63 kg, CR 15, blend B30. It is found that the predicted and the experimental results were almost the same.