.. _method-multilevel_blue-global_local:

""""""""""""
global_local
""""""""""""


Use a hybrid global-local scheme for solving an optimization sub-problem



.. toctree::
   :hidden:
   :maxdepth: 1



**Specification**

- *Alias:* None

- *Arguments:* None


**Description**


This sub-problem solver option employs multiple methods, both
sequenced and competed.  It is currently the most comprehensive (and
expensive) option.  First, the sequence is from a global subproblem
solve, followed by competed local solves that are initiated from the
best global solution.  The global solver is DIRECT and the competed
local solves are sequentual quadratic programming (``sqp``) and
nonlinear interior point (``nip``), from NSPOL and OPT++ respectively.