Prioritization of development activities in software companies iscommonly performed arbitrarily and subjectively, based solely onthe experience of the company’s employees. Aiming to contributeto the automation of this process, this paper presents a methodof automatic prioritization of software development activities, usingan algorithm that combines the use of a numerical regressionbasedArtificial Neural Network (RNA) with an ordering method.To validate the accuracy of the proposed method, a case study wasconducted, with real data from a software development company.The experiments allowed to evaluate the accuracy of the proposedmethod qualitatively and quantitatively, generating satisfactoryresults, which indicate the feasibility of implementing the proposedmethod in the target company of the case study.