[vc_row gap=”10″][vc_column width=”2/3″][vc_column_text]

Details of Optimal Feed Formulation Savings Analysis

  1. Determine that the items and item-lots required by the formula(s) are in the Optimal database. The Optimal solution conforms to constraint limits, so the Lots table must contain data for all constraint calculations.
  2. The set of lots to be used in this study must be identified by a Lot Set name. If that set is not identified in the Lots table, then:
    1. Add a lot set name to the Lot Sets table, e.g., Savings Analysis
    2. In the Lots table, identify the lots to be used by assigning the new lot set name to them.  This is usually done by copying existing versions of the required lots and replacing the lot set name with the new one.
  3. Update Item Lot costs and attribute values.  The Lots table can have multiple uniquely named cost columns to support various uses of the system.  Decide which cost column is to be used for this study.
  4. Add an input set name to the Input Sets table and in the Input Set Product table add the product or products to be costed for the input set name just entered. The only data required is the batch size.
  5. Cost of existing approach (reference formula(s))
    1. From the Data menu, select Edit Data Tables.  In Formula Items table, for each formula in the study enter the items in the existing recipe including the amount specified in the recipe.  The amount can be specified either as weights or as percentages of a basis weight; in either case, the minimum amount must equal the maximum amount.
    2. In Formula Constraints table, for each formula in the study enter the name of all constraints that are to be used in the optimized formula(s) cost step.  In the Existing approach step, the constraints do not control the solution (set Minimum = Maximum = 0); they are input so that the value of the constrained formula characteristic is computed and displayed in the results.
    3. From the Optimize menu, open the Selections table.  Set Selection parameters to
  6. Run Comment = Reference formula (you just need to distinguish between runs; a run number is assigned to each run, so run comment can be anything)
  7. Input Set = input set name created in step 4
  • Model Type = Single Product or Multi-Product (depending on the number of products in the input set)
  1. Lot Set = see step 2
  2. Unlimited Quantity = check the box (the costing is not to be affected by item availability)
  3. Lot Attribute for cost = cost column selected in step 3

Click the Solve button to create and optimize the model. Since all items in the Formula Items table have fixed amounts or percentages, the optimizer does exactly what the recipe says to do and sends all results including costs to the results tables where they are identified by run number.

  1. Verify that the model terminated normally and produced anticipated results. If not, you should be able to find the problem in the data.
    1. The model completion message box informs of the completion status
    2. Look at Error Log and Warning Log for messages about problems
  • Look at SQL reports to see that all is as expected.
  1. Cost of Optimal approach (optimized formula(s))
    1. In Formula Items table, for each formula in the study enter the items that can be used in the formula (this might be different from the items offered in the reference case).  Optimal determines how much of each is item is used in the optimized solution; however, some items might have minimums and/or maximums in order to meet various requirements.  These minimums and maximums are entered either as weights or as percentages of various basis weights.
    2. In Formula Constraints table, for each formula in the study enter all constraints to be considered by the optimizer.  This is a critical step.  The optimizer does everything possible to reduce costs; the constraints provide the acceptable range for all of the product characteristics, to assure that the optimizer operates within the product specifications.
    3. From the Optimize menu, open the Selections table.  Set Selection parameters to
  1. Run Comment = Optimized formula
  2. All other parameters can be the same as in the reference case

Click the Solvebutton to create and optimize the model.

  1. Verify that the model terminated normally and produced anticipated results.
    1. The model completion message box informs of the completion status
    2. Look at Error Log and Warning Log for messages about problems
  • Look at SQL reports to see that all is as expected.

This version of the model might not be as easy to get correct as the reference formula(s) model because you are trying to influence how much of each item is used based on a few item limits and limits on the product characteristic.  You can anticipate that you might not get this exactly right the first time.  Also, the results reports must be studied carefully to verify that the product formulation is reasonable in all respects.

  1. Compare results.
    1. Verify that the formula characteristics from the reference formula(s) version are all within the constraint limits used in the Optimized formula(s) version. Otherwise, the comparison might not be valid.
    2. Having studied the Optimized Formula(s) to confirm that it is acceptable (meets all Item limits and Constraint limits as well as produced a reasonable Recipe), compare the cost and project a savings from the reference formula(s) to the Optimized formula(s) (Run the Optimal SQL report that does the comparison.)

[/vc_column_text][/vc_column][vc_column width=”1/3″][vc_single_image image=”1432″ img_size=”large”][vc_column_text]

[/vc_column_text][/vc_column][/vc_row]