摘要：Viscosity is the property of fluids to oppose movement when a cutting effort is applied on them to convey them from one point to another. Heavy oil has a high viscosity greater than 1000 cP, which makes it difficult to transport. The present work shows a mathematical model for the prediction of the viscosity of dead heavy oils produced in the fields of Monagas State, Venezuela. For the development of the work, 25 samples of oil were collected and the viscosity was measured at 5 temperatures, in addition to the API gravity and the percentage of Asphaltenes. The data were introduced in the Statgraphics Centurion XVI statistical package and through multiple regression analysis two mathematical models were obtained, 1) linear multiple and 2) multiple nonlinear; The best model being divided according to its coefficient of determination R2 and the average relative error (ARE). The selected model was compared with the Glaso, Bennison and Naseri models. The nonlinear multiple model with R2 of 0.9792 and ARE of 5.05% was obtained as the best model, surpassing the models of Glaso (35.5% ARR, Bennison (107.5% ARE) and Naseri (61.7% ARE).
