The strength of concrete is a function of the proportions of the constituent materials, namely, cement, water, fine and coarse aggregates. The conventional methods used to determine the mix proportions that will yield the desired strength, are laborious, time consuming and expensive. In this paper, a mathematical method based on modified regression theory is formulated for the prediction of concrete strength. The model can prescribe all the mixes that will produce a desired strength of concrete. It can also predict the strength of concrete if the mix proportions are specified. The adequacy of the mathematical model is tested using statistical tools.