Nurse scheduling is an inherently complicated task due to the nature of the problem where a set of constraints must be met simultaneously. Manually altering the schedule of one nurse directly affects the schedule of all other nurses in both the total nurses in one day and the total shifts assigned to one nurse. Hence, finding an acceptable schedule can be time consuming and may result in inefficient and unfair schedule. Alternatively, mathematical programming, namely mixed integer linear programming (MILP), approaches have been extensively studied in the literature to optimize nurse schedules. However, the complex nature of this problem makes it classified as NP-Hard which may lead to not finding even a feasible solution in a reasonable time as the problem becomes larger. Thus, metaheuristic approaches have been developed to find approximate solutions, such as the genetic algorithm (GA), where a population of candidate solutions evolves over generations to find optimal or near optimal solution through selection, crossover, and mutation. In this paper, we introduce a novel mutation method to find a fair load distribution and resource assignment in nurse scheduling using the GA. The tested problem is real-world case of nurse scheduling at a pediatric intensive care unit at one of the largest hospitals in Saudi Arabia. The results of the GA model are then compared with the results of the exact approach using MILP.