<p>Mat foundation design can pose important soil-structure interaction problems that are often oversimplified. According to Winkler’s hypothesis, the deformation of foundations under applied loads is confined to the loaded regions, but this is seldom accurate. In structural analysis, Winkler springs, defined by the coefficient of subgrade reaction (k_s), are mostly used to represent the ground. To simplify the modeling process even more, designers tend to consider a single value of k_s. In 1995, the American Concrete Institute published a special review on the design and performance of mat foundations (SP-152) suggesting the iterative discrete area method (DAM), which accounts for variable k_s values across the mat; however, this method is seldom used in practice. This study investigates the impact of the DAM on the steel reinforcement design of large mat foundations and compares it with the reinforcement that results from using a constant k_s. Validation was performed using the finite element method. With the variable k_s model, an optimized steel reinforcement distribution is generated. Most analysis show a reduction in top steel reinforcement and an increase in bottom steel reinforcement.</p>