The estimation of parameters in semi-empirical nonlinear models through the error-in-variables method has been widely studied from a computational standpoint. This method involves the minimization of a quadratic objective function subject to the model equations being satisfied. Due to the nonlinear nature of these models, the resulting formulation is nonconvex in nature. The approaches to solve this problem presented so far in the literature, although computationally efficient, only offer convergence to a local solution df a model which may contain multiple minima. In this paper a global optimization approach based on a branch-and-bound framework will be presented to solve the error-in-variables formulation. Various estimation problems were solved and will be presented to illustrate the theoretical and computational aspects of the proposed method.