.. _method-multifidelity_sampling-numerical_solve-model_reordering: """""""""""""""" model_reordering """""""""""""""" Approach for addressing model misordering during numerical solutions. .. toctree:: :hidden: :maxdepth: 1 method-multifidelity_sampling-numerical_solve-model_reordering-auto_reorder method-multifidelity_sampling-numerical_solve-model_reordering-fixed_order **Specification** - *Alias:* None - *Arguments:* None - *JSON argument key:* None **Child Keywords:** +-------------------------+------------------------+--------------------+--------------------------------+------------------------------------------------------------+ | Required/Optional | Description of Group | JSON Group Key | Dakota Keyword | Dakota Keyword Description | +=========================+========================+====================+================================+============================================================+ | Optional | | | `auto_reorder`__ | Address any model misordering in MFMC numerical solutions | | | | | | by reordering on the fly. | +-------------------------+------------------------+--------------------+--------------------------------+------------------------------------------------------------+ | Optional | | | `fixed_order`__ | Enforce a prescribed model ordering in MFMC numerical | | | | | | solutions through additional linear constraints. | +-------------------------+------------------------+--------------------+--------------------------------+------------------------------------------------------------+ .. __: method-multifidelity_sampling-numerical_solve-model_reordering-auto_reorder.html __ method-multifidelity_sampling-numerical_solve-model_reordering-fixed_order.html **Description** Multifidelity Monte Carlo is a hierarchical sampling method that requires an ordered set of model approximations, whether in terms of model response correlations (for analytic solutions) or model sample allocations (for estimator variance calculations in numerical solutions). In the latter case of numerical solutions, ordering of sample allocations can be accomplished using two different approaches, either reordering on the fly (:dakkw:`method-multifidelity_sampling-numerical_solve-model_reordering-auto_reorder`) as dictated by a given set of allocations or enforcing a prescribed model ordering (:dakkw:`method-multifidelity_sampling-numerical_solve-model_reordering-fixed_order`) through additional linear constraints on the numerical solution.