PT. Tiga Dimensi Mandiri Bekasi often has problems in carrying out maintenance because the company partners are very many and there are so many production machines that sometimes the profit results targeted by the company do not match. PT. Tiga Dimensi Mandiri Bekasi has a standardization in handling maintenance by conducting a survey between the distance, expenses and work income from many partners. Based on the above problems to overcome these problems, a system is needed to optimize the machine maintenance process based on the engine category and condition. The goal is to make it easier to identify and simplify maintenance and provide recommendations after the optimization process to find out which machines need to be handled first and optimize the number of employees. The purpose of this system is to produce output in the form of optimization of company profits. Algorithm is a meta-heuristic method that can solve production machine maintenance problems. The proper formulation of the fitness formula can provide a value from the chromosomes in the genetic algorithm, so that the genetic algorithm can work optimally to solve maintenance problems. The right genetic algorithm parameters can provide maximum results. The test shows the best results for the number generation of 100, the number of chromosomes by 20, the crossover rate of 0.8, and the mutation rate of 0.1. This study suggests that genetic algorithms can solve the complexity problem of optimizing profits in machine maintenance work.