Credit risk models are vitally important for organizations whose corporate purpose is to operate profitably in the loan or credit business. Technological developments have enabled the application of different statistical techniques to create functions that assist in measuring, and consequently in managing, exposure to credit risk; however, these models must be periodically reassessed and optimized to ensure that they fulfill their objectives. This study addresses problems that have been observed in the model for reading the credit history of customers of a company in the real sector, contributing to the design of a risk-scoring model using the discriminant analysis technique.