Building a design quest

A design quest flow contains multiple tasks to prepare data, define the setup model, apply preconfigured solvers, and compare the results.

  1. Select a quest.
  2. Click Create new.
  3. Specify a name. These characters are valid: A-Za-z0-9@#$&:[]-`.+, "s_.
  4. Optionally, specify a description.
  5. Select the group from the Data Collection list to load single or multiple datasets at once. Once the group is selected, all the datasets that are tied to that group are reflected in the dataset collection space.
  6. Complete these steps to build the flow of the design quest. Additional datasets can be added to the already loaded quest group if needed.
    1. Transform data into a useful structure with preprocessing activities, such as normalization, transformation, or reduction.
    2. Define the constants, decision variables, constraints, and objective functions.
    3. Apply a preconfigured solver, then run individually or the entire quest.
    4. Compare the results generated from multiple models under the results tile.
    5. Move the model to the production phase and deploy it as an end point.