Abstract: Neural networks and genetic algorithms are used for optimization design of laminated composite plate structures. Two steps are taken to meet the stability requirement of the composite laminate. At the first step, the thickness of plies with prescribed orientations is obtained by a neural network model. The basic laminate structure is formed when the total number of the laminate is given. At the second step, the ply stacking sequence of the laminate is optimized by a genetic algorithm. The optimal laminate design is acquired for a given weight.