Magnetic abrasive finishing (MAF) exhibits considerable potential as a promising process for enhancing surface finish quality for softer as well as harder materials. This work proposes a novel approach to harness the potential of MAF for obtaining superior surface quality of precipitation-hardening martensitic steel (17-4PH steel) while performing periodic re-distribution of magnetic abrasive particles. The experiments were conducted using the Box-Behnken design of experiments and response surface methodology (RSM). RSM-based desirability function approach and genetic algorithms were used for multiobjective optimization. The set of pareto-optimal solutions obtained through multiobjective optimization were ranked using the Technique for Order of Preference by Similarity to an Ideal Solution (TOPSIS) by assigning equal importance to the selected performance parameters (percentage change in surface roughness (PCSR) and material removal rate (MRR)). The optimized values of PCSR and MRR obtained through the desirability function of RSM and genetic algorithm were validated experimentally and compared with the predicted values obtained through the regression model. Results showed that the desirability function of RSM provided better results than the genetic algorithm for the selected conditions and confirmed the effectiveness of the proposed approach.