A thinning method based on precoded subarray structures is proposed for planar arrays. It introduces the matrix descriptions, including the subarray structure matrix <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</i> and the coefficient vector <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> , and transforms the thinning of a planar array into the arrangement of predetermined subarrays with specified thinning rates. This process effectively leads to order reduction and achieves fast optimization. We apply the evolutionary algorithm to accomplish the ideal radiation performance iteratively. The dependence of the array performance on <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> is also investigated. For demonstration, the 10×20- and 23×23-element arrays are optimized. By comparing with the traditional thinning methods, it is verified that the proposed method not only provides us with lower sidelobes and narrower beamwidth but also avoids the random element arrangement. In addition, the introduction of precoded subarrays may further reduce production and maintenance costs, especially from engineering implementation and mass production perspectives.