The nutrition diet decision problems on Multi-objective optimization are solved by using Compromise Difference Evolutionary (DE) algorithm. This method is equipped with a domination selection operator to enhance its performance by favoring non-dominated individuals in the populations. DE is a population based search algorithm, which is an improved version of Genetic Algorithm (GA). Simulations carried out involved solving nutrition decision using a method that relationships of dominant to determine the fitness, and finding Pareto optimum set for the nutrition decision problem. Compromise Difference Evolutionary found to be stable and more accurate in optimization compared to simple GA.