Contracted capacity setting is a discrete and nonlinear optimization problem in consideration of expenses on the electricity from the utility and costs of the self-owned generating units (SOGUs) at the same time. This paper analyzes and compares the proposed improved Taguchi method and the cultured differential computation algorithm (CDCA) in solving this problem. The improved Taguchi method provides fast converging characteristics in searching the optimal solution through quality analysis in orthogonal matrices. The cultural algorithm (CA) used in the CDCA extracts and saves the domain knowledge or problem properties during the evolution process. The differential computation of the CDCA provides fast converging characteristics in searching the optimal solution through operations of mutation, crossover, and selection operations, which are efficient and different from the existing generic algorithm (GA). To compare the proposed methods, the paper employs the real data obtained from an optoelectronics factory in Taiwan. In comparison with the existing optimization methods in different optimization problems, the proposed approaches have superior results as revealed in the numerical results with respect to the computation time and the solution quality.