An iterative optimization method based on linearization and linear programming is developed. The method can be used for the determination of the material distributions in a fast reactor which maximize or minimize integral reactor parameters that are linear functions of the neutron flux and the material volume fractions. The method has been applied to the problems of optimization of the fuel distribution in a reactor of fixed power output, constrained power density, and constrained material volume fractions so as to obtain (a) a maximum initial breeding gain, (b) a minimum critical mass, and (c) a minimum sodium void reactivity. Under this realistic set of constraints, numerical results show that the same fuel distribution yields maximum breeding gain, minimum critical mass, minimum sodium void reactivity, and uniform power density.