Exclusion regions for optimization problems (Q2250096)

The authors consider the global optimization problem \[ \min f(x)\text{ s.t. }F(x) =0,\;x \in \mathbf{x} (\text{box})= [\underline{x}, \overline{x}]\in\mathbb R^n. \] Branch and bound methods using constraint propagation methods for solving this problem in a verified way suffer from the so-called cluster effect. This paper discusses the reasons for the occurrence of this so-called cluster effect, and how to reduce the cluster effect by defining exclusion regions around each local minimum found. In addition, a method is introduced for verifying the existence of a feasible point close to an approximate local minimum. Examples illustrate the given theory.