This paper presents an improvised version of Manta-Ray Foraging Optimization (MRFO) by using components in Genetic Algorithm (GA). MRFO is a recent proposed algorithm which based on the behaviour of manta rays. The algorithm imitates three foraging strategies of this cartilaginous fish, which are chain foraging, cyclone foraging and somersault foraging to find foods. However, this optimization algorithm can be improved in its strategy which increases its accuracy. Thus, in this proposed improvement, mutation and crossover strategy from GA were adopted into MRFO. Crossover operation is a convergence action which is purposely to pull the agents towards an optimum point. At the meanwhile, mutation operation is a divergence action which purposely to spread out the agents throughout wider feasible region. Later, the algorithms were performed on several benchmark functions and statically tested by using Wilcoxon signed-rank test to know their performances. To test the algorithm with a real application, the algorithms were applied to an interval type 2 fuzzy logic controller (IT2FLC) of an inverted pendulum system. Result of the test on benchmark functions shows that GMRFO outperformed MRFO and GA and it shows that it provides a better parameter of the control system for a better response.