Optimization Model

What does the optimization model do?

S&OP Navigator uses an optimization model to get the best solution, instead of using a set of if-then rules.

In an optimization model, you have a set of decision variables (possibilities the model can compare in multiple combinations) and set of constraints (rules the model must follow). The optimizer finds the value for every variable that provides the best objective (the answer you are looking for) that satisfies the set of constraints. This form of optimization is useful in solving highly complex problems.

The outcome of the model is controlled by the constraints and objective you define.

In any model, there are constraints.

• hard constraints that must be held, such as a limitation of inventory based on physical space available.

• soft constraints can be relaxed in some cases, such as an arbitrary resource limit based on expected demand. An example of a soft constraint is if there is too much demand, the resource limit might be violated. This means that the model will report a utilization of more than 98%. When there are multiple soft constraints, you must indicate the order of priority.

The Options widget is the space to toggle whether to enforce certain hard constraint, or relax it.

• Hard constraints

  • Allow over utilization

    When this options is off, all resources must operate within capacity. This is a hard constraint. However, this hard constraint can be relaxed resource by resource. After this switch is on. Right click on this switch:

    Once in the dialog, you may use the filters to select the targeted resources, and apply Select or Unselect for them.

    

    Selected means allowing over utilization for this resource. Unselected means over utilization is not allowed for this resource.

  • Allow unfulfilled demand

    When this option is off, all demand must be met. This is a hard constraint.

  • Strict capacity limit

    Options with strict in the name usually means this is a hard constraint. This constraints refer to the capacity calculated by MAX Maximum Utilization Limit.

  • Min inventory coverage

    The minimum inventory required by inventory coverage functionality cannot be violated. This is a hard constraint.

  • Strict min inventory limit

  • Strict max inventory limit

  • Minimum production

  • Batch Production

  • Batch Supply

  • Supplier proportion

  • Supplier minimum proportion

  • Supplier maximum proportion

  • Frozen Periods

    In the actual production planning practice, production plans may already have been determined 2 or 3 cycles prior, meaning that in the next 2 or 3 cycles production volume should be the predetermined numbers. To achieve that, you can use attribute Production Volume Input in table "Bill Of Resource Data" to provide predetermined input data. And with a right click on the switch, a dialog will show up to toggle which resources should follow input data in frozen periods.

• Soft Constraints

When you have enough capacity to meet all these constraints, you have the best possible solution. When you don’t have enough capacity, some of these constraints must be violated.

The model has the following priority of the constraints. The constraint with the lowest priority (Priority 5) will be violated first, while constraints with the highest priority (Priority 1) will be violated last.

Priority 5) Above Inventory Target Constraint. Behavior: If needed, the model will build up inventory above the target to meet future demand.

Priority 4) Below Inventory Target Constraint. Behavior: If needed, the model will draw down below the target level to meet current demand, and will try to bring inventory back to target level in future periods.

Priority 3) Inventory Maximum. Behavior: If needed, the model will build up inventory above the maximum level to meet future demand.

Priority 2) Inventory Minimum. Behavior: If needed, the model will draw down inventory below the minimum level to meet current demand and will try to bring it back up in future periods.

Priority 1) Resource Capacity Constraint (Setting 1, Setting 2 and Setting 3). Behavior: If needed, the model will use more capacity than it has available.

In general, violations of Priority 4 and 5 constraints don’t require much attention, as you are still within your inventory and capacity limits. Violations of Priority 1, 2, and 3 constraints require attention, as you are operating outside the specified limits and run risk of stock out. Cells are marked clearly in the application in this case.

When a solve is not feasible

A dialog indicates this solve can't get an optimal solution. Click OK to go back to the page and check the following:

• Uncheck hard constraints, or relax the ranges of hard limits.

• Check if there is initial inventory.

• There may not be enough initial inventory, and the second month's capacity is not enough.