The modeling of blending tank operations in petroleum refineries for the most profitable production of liquid fuels in a context of time-varying supply and demand is addressed. A new mixed-integer nonlinear programming formulation is proposed that using individual flows and split fractions as key model variables leads to a different set of nonconvex bilinear terms compared with the original work of Kolodziej et al. These are better handled by decomposition algorithms that divide the problem into integer and nonlinear components as well as by commercial solvers. In fact, BARON and GloMIQO can solve to global optimality all problems resulting from the new formulation and test problems from the literature. A tailored global optimization algorithm working with a tight mixed-integer linear relaxation from multiparametric disaggregation achieves a similar performance. (c) 2015 American Institute of Chemical Engineers