Computer experiments are widely used for the design and development of products. Therefore, a new method for constructing optimal experimental design(SOEDM) is developed in this paper. There are two major developments involved in this work. One is on developing a multi- objective optimal criterion by combining correlation and space-filling criteria. The other is on developing an efficient global optimal search algorithm, named as improved enhanced stochastic evolutionary (IESE) algorithm. Several examples are presented to show: the optimal designs are good in terms of both the correlation and distance criteria, the new method can be used in other experimental designs besides Latin hypercube design the number of levels of each factor is equal to the number of the experimental schemes, and the new algorithm is fast.