Much of the fuel ethanol industry’s current interest centers on maximizing ethanol yield and overall profits. This can be achieved by knowing the potential yield of input corn and working to identify what parameters are inhibiting reaching 100% fermentation efficiency. On average, ethanol plants produce 2.82 gallons of ethanol per bushel of corn, as compared to 2.51 gallons per bushel in 1994 (Renewable Fuels Association 2015). With the focus on improved starch production and access, corn quality is one of the best indicators of ethanol yield, as the amount of starch determines the theoretical amount of ethanol. Near-infrared spectroscopy (NIRS) is one such method that can be used to evaluate corn composition and, with an appropriate model, corn composition can be used to predict ethanol yield. Many current models are held back by real world applicability, in that they are restricted to lab-scale validation, direct NIRS calibrations, or proprietary models/equipment. At the commercial level, corporately-produced propriety models have been developed by DuPont Pioneer and Monsanto. Neither Monsanto nor DuPont Pioneer’s products are available outside of company databases, and both are only applicable to Foss Infratec units, which left a need for a more universal method. Burgers et al. developed a multiple-linear regression equation for predicting corn ethanol yield based on near-infrared spectroscopy (NIRS) measurements of protein, oil, and density on a 15% moisture basis (Burgers, Hurburgh, and Jane 2009). Unlike corporately-developed models, this equation was intended to function independently of corn hybrid, corn supplier, growing location, and NIRS instrument make/model used, as long as the calibration database was consistent. Iterations of the model were evaluated, and the most current version was chosen to use in the rest of the research. A comparison of the model predicted yield, based on inbound grain composition, and corresponding reported ethanol yield from the same grain was performed to validate the model. The slopes for the plants’ predicted and reported ethanol yields did not differ significantly from one another. Overall, the combined model for the linear regression produced a low R2 value (0.23) which shows that a significant amount of variability in the data is not explained by the model. On average, the data validated the prediction model. Day to day or batch by batch variability in processing was not accounted for in the equation, but the variability of the corn composition was. From the linear regression analyses performed on each plant, the slopes are the same, but there is a plant-specific bias. This equation identified key corn quality parameters. Because the equation validated for all plants, the equation is validated to function independently of corn hybrid, corn supplier, growing location, and NIRS instrument make/model used. The model validated with a root mean square error of 0.13 gal/bu, and no difference (0.0008 gal/bu) between overall reported and predicted yield means.