A study about multi-objective optimization of turning process has been done in this paper. Twenty-five experiments of a matrix designed according to Taguchi method have been performed. In each experiment, the values of five parameters were changed, including tool nose radius, tool holder length, spindle speed, feed rate, and cutting depth. The three output parameters determined for each experiment include surface roughness, roundness deviation and material removal capacity. Four different methods were used to calculate the weights of the output parameters. The FUCA method was used to solve the multi-objective optimization problem. This work was repeated four times with four corresponding weight sets of the criteria. The purpose of solving the multi-objective optimization problem is determining the values of the input parameters to ensure both the surface roughness parameter and the roundness deviation parameter are the smaller the better, and the material removal capacity is the larger the better. A surprising thing happened, the optimal values of the set of input parameters were exactly the same when using four different weighting methods. Accordingly, the optimal values of tool nose radius, tool holder length, spindle speed, feed rate and cutting depth correspondingly are 0.8 (mm), 40 (mm), 587 (rev/min), 0.316 (mm/rev) and 0.6 (mm).

This paper studies the efficiency improvement of AISI 4140 steel external cylindrical grinding process. Experiments were carried out with two types of grinding wheels, a conventional and the rubber-pasted grinding wheels. With each type of the wheel, nine experiments were performed. Cooling fluid, spindle speed, feed rate and depth of cut are variables in the experiments. Two outcomes used to evaluate grinding efficiency are surface roughness and material removal rate (MRR). Experimental results demonstrate that the surface roughness achieved in the grinding operation using the rubber-pasted grinding wheels is smaller than using the conventional wheel. The Data Envelopment Analysis-based Ranking (DEAR) method was applied to determine the optimal values of the input parameters for the “minimum” surface roughness and “maximum” MRR in the cases of using each grinding wheel. It is found that the optimization of the input parameters in this circumstance are equal. Several grinding experiments for examining the values of the variables were also performed for both the wheels. The results also confirm that the surface roughness in the grinding process with the rubber wheels is about 19.42% smaller than the conventional wheels.