You can model and solve Decision Optimization problems using the Modeling Assistant (which enables you to formulate models in natural language). This requires little to no knowledge of Operational Research (OR) and does not require you to write Python code. The Modeling Assistant is only available in English and is not globalized.

The basic workflow to create a model with the Modeling Assistant and examine it under different scenarios is as follows:

1. Create a project.
2. Add a Decision Optimization experiment (a scenario is created by default in the experiment UI).
3. Add and import your data into the scenario.
4. Create a natural language model in the scenario, by first selecting your decision domain and then using the Modeling Assistant to guide you.
5. Run the model to solve it and explore the solution.
6. Create visualizations of solution and data.
7. Copy the scenario and edit the model and/or the data.
8. Solve the new scenario to see the impact of these changes.

This is demonstrated with a simple planning and scheduling example .

See also Build, run, and deploy a Decision Optimization model for a quick video showing you how to run a sample Decision Optimization experiment to create, solve, and deploy a model by using the Decision Optimization experiment UI.

For more information about deployment see Deploying with watsonx.ai Runtime.