.. _auto_dataset_compute_pareto_front_optimal:

*COMPUTE PARETO FRONT OPTIMAL FROM PROVIDED DATASET*
====================================================

Computes the Pareto-optimal point from a multi-objective dataset by filtering dominated solutions across the specified objective columns. Supports min/max criteria per objective, optional design constraints, target values, and optional meta-model fitting (polynomial or RBF) to refine the optimal point beyond the existing discrete samples.

When to use
-----------

Classification: **process**.

Tagged: ``design_exploration``, ``dominance``, ``meta_model``, ``multi_objective``, ``optimization``, ``pareto``, ``pareto_front``, ``pareto_optimal``.

Inputs
------

.. list-table::
   :header-rows: 1
   :widths: 20 20 20 20 20 20

   * - Label
     - ID
     - Type
     - Default
     - Required
     - Description
   * - Pareto Dataset
     - pareto_dataset
     - dataset
     - —
     - 
     - Input dataset containing the design space; each row is a design point and columns include the objective and input variables to evaluate for Pareto optimality.
   * - Columns
     - columns
     - scalar
     - —
     - 
     - Comma-separated list of column names from the dataset that represent the objectives to be optimized; must match column headers exactly (dynamic list populated from pareto_dataset).
   * - Criteria
     - criteria
     - scalar
     - min
     - 
     - Comma-separated optimization direction per objective column (e.g. 'min,max'); number of entries must equal the number of columns specified — defaults to 'min' for all objectives.
   * - Constraints
     - constraints
     - dataset
     - —
     - 
     - Optional dataset defining inequality/equality constraints on design variables used to filter infeasible points before Pareto front computation; leave empty if no constraints apply.
   * - Target Values
     - targets
     - textarea
     - —
     - 
     - Optional target values for each objective specified as colon-separated key-value pairs (e.g. 'mass:20,stress:30'); used to bias the optimal-point selection toward desired values.
   * - Fit Meta Model
     - fit_meta_model
     - select
     - —
     - 
     - Meta-model type to fit over the Pareto front for refining the optimal point beyond discrete samples; choose 'no' to return the best existing point, or 'polynomial_1', 'polynomial_2', RBF, etc. to interpolate a potentially improved optimum.
   * - Inputs if a Meta-model based Optimum is Chosen
     - inputs
     - scalar
     - —
     - 
     - Optional dataset or column list of input/design variables (distinct from objectives) used when fitting the meta-model to map design space to objective space.

Outputs
-------

.. list-table::
   :header-rows: 1
   :widths: 20 20 20 20

   * - Label
     - ID
     - Type
     - Description
   * - dataset_compute_pareto_front_optimal_output_1
     - dataset_compute_pareto_front_optimal_output_1
     - dataset
     - Single-row dataset representing the identified Pareto-optimal design point, including all original columns and the computed optimal values after applying the selected criteria, constraints, targets, and optional meta-model.

Disciplines
-----------

- ai_ml.surrogate
- data.dataset.transform
- design_exploration.optimization

Runnable example
----------------

A runnable example is registered for this worker. Open the example workflow on the d3VIEW canvas: `/api/workflow/example?id=dataset_compute_pareto_front_optimal <https://www.d3view.com/api/workflow/example?id=dataset_compute_pareto_front_optimal>`_

.. raw:: html

   <hr style="margin-top:2em">
   <p style="font-size:11px;color:#888">
   Auto-generated from <code>transformation</code> schema. Worker id: <code>dataset_compute_pareto_front_optimal</code>. Schema hash: <code>7a5fbfd32fa7</code>. Hand-curated docs in <code>workerexamples/</code> override this page when present.
   </p>