Handle Almost Any Continuous Function
The LGO Global Solver allows the objective and
constraints to be general, convex or non-convex functions --
the only assumption made is that the functions are Lipschitz-continuous.
A function f is Lipschitz-continuous if there exists a
constant L > 0, such that for all pairs of points x, y, the
absolute value of f(x) - f(y) is no greater than L times the
distance between x and y. Among many others, every smooth function
(the type handled by the bundled GRG and Large-Scale GRG Solvers) on
a closed and bounded set is Lipschitz-continuous.
Use a Full Repertoire of Global and Local Search Methods
The LGO Global Solver includes a full repertoire
of global and local scope search methods. The global methods include
continuous branch and bound and adaptive random search;
the local methods include an exact penalty function approach,
a search based on sequential model linearization, and an exact
primal (Generalized Reduced Gradient type) nonlinear search
algorithm. These methods are invoked sequentially, in a fully
automatic mode. LGO can also be used as a stand-alone nonlinear
optimizer (seeking locally optimal solutions) for a wide range of
continuous problem functions.
Options for Mixed-Integer Nonlinear Problems
