Sequential optimization and infeasibility

In cases of infeasible models, you can select the objectives that are related to the infeasibility and run the model through sequential optimization. Constraints that are listed as sequential objectives are treated as objectives and act as soft, rather than hard, constraints, allowing the model to solve. The constraint can be violated and the violation is added to the objective function as a penalty term.

The Constraints Summary table identifies the diagnostic excesses or surpluses related to the constraints. For example, if you have included production constraints and the model is infeasible, you can run the model with sequential optimization, using "Production" as the sequential objective. In the Constraints Summary output table, the Diagnostic Deficit column displays a value indicating how much production could not occur due to the production constraints.

As another example, assume that your model has an infeasibility which is caused by site capacities. However, you may not know this in advance. In this case, you can specify all the hard constraints in the table (flow, flow count, etc.) and give them all priority 1. The objectives are defined in the table in Objectives and hard constraints. After you run sequential optimization, review the Constraints Summary output table. Non-zero values in the Diagnosis Excess or Diagnosis Deficit columns indicate the source of infeasibility in the model.