Define Processing Settings

MSO uses the Run screen to set up scenarios for processing (where they become 'workers' for subsequent queuing and processing).

The Processing Settings screen is used to define default processing parameters and other general behaviours.

Warning: These settings can potentially affect all optimizations significantly, so should be used with caution. See Solver Emphasis & Optimality.

See Solver Emphasis & Optimality.

Activity steps

  1. Create an MSO scenario and define a block model.

  2. Configure other settings as required.

  3. Display the Run screen.

    The Worker Configurations table shows an entry for every scenario that exists for your project.

  4. Click Settings.

    The Processing Settings screen displays.

  5. Choose what to do if a result file already exists using Existing Data Action.

    • Prompt – Choose what to do in each case.

    • Overwrite – Forcibly overwrite (replace) the existing data without prompts. Be careful with this one!

      Skip – Ignore the new data and leave the old data in place.

  6. Choose the Default Worker Configuration:

    • Cases to Process – This can either be Base Case, meaning only the primary scenario case is processed, or Base Case & Sensitivities (all possible case variants are processed to understand the impact of parameter changes).

  7. Define Solver Settings:

    Note: You should only change these settings if you understand their potential impact on results.

    • Concurrent Threads – The default number of threads to assign to each new case. This is the default value that appears when you Set up Worker Configurations.

    • MIP Solver – The default solving engine used for processing, either LP Solver or CPLEX (requires a separate license).

    • MIP Emphasis – See Solver Emphasis & Optimality.

    • Time Limit (Hours – Set a maximum time limit for solution calculation. After the limit has been reached, the best solution will be presented (or no solution, if all outcomes are sub-optimal).

    • Gap Tolerance – The precision of result where smaller values equate to more precise calculations (and a potentially longer runtime). See Solver Emphasis & Optimality.

    • Integrality Tolerance – The allowed deviation from integer solutions. See Solver Emphasis & Optimality.

    • Pre-Solve – Pre-solve is a preprocess of the solver model. If checked, MSO looks for ways to simplify the model. For example, it can delete unused variables and restrictions, substitute fixed variable values by a constant and so on. The result is a new model that is less complex than the original model and (hopefully) gives rise to more rapid solution calculations.

    Related topics and activities