ABC classification, also known as Pareto analysis, is a useful method for classifying forecast items based on their relative importance to the organization. Many companies adopt different procedures for creating, reviewing and monitoring forecasts based upon an item’s classification. This enables the forecasting team to focus its efforts on those items which have the greatest impact to the organization.
Items are typically categorized as follows:
- A: important high volume items
- B: medium volume items
- C: slow moving items
Forecast Pro TRAC can automatically assign items into ABC categories based upon their volumes. Alternatively, you can create a custom filter to define your own classifications. In this edition of Tips and Tricks, we explore both approaches and illustrate how sorting and filtering reports based on the Pareto output can be extremely useful.
Automatic ABC Classification
Forecast Pro TRAC can automatically assign ABC codes to your items based on their relative volumes. You can concentrate on your most important items by sorting reports, filtering reports, and structuring your forecast process based on these codes.
The Pareto Analysis tab on the Settings menu allows you to customize the thresholds used to determine which items will fall into which category—A, B or C.
The dialog box provides you with a number of options. First, you can choose the Weighting Basis. This determines whether the history, forecasts, or both history and forecast values are used to determine the ABC classifications. The spinners next to the two categories allow you to specify a range of periods within the history or forecasts for Forecast Pro to consider. The classification occurs at a single level of the hierarchy—the Hierarchy level drop down menu allows you to choose the level at which you wish to classify. In this example, Forecast Pro will classify the SKU level based on the last 12 periods of history and the first 6 periods of forecasts.
The second section, Ranking Basis, allows you to specify the actual thresholds used to determine classifications. There are three different ways you can classify your items:
- Volume (%). The Volume (%) option allows you to set the codes based upon the cumulative volume of each group. For example, if A is set to 80%, B is set to 15% and C is set to 5%, Forecast Pro TRAC will first rank all of the items from highest volume to lowest volume and then sequentially place the top ranked items into the A group until their cumulative volume equals or exceeds 80% of the total volume. It will then sequentially place items into the B group until the cumulative volume of the A and B items equals or exceeds 95% (80% + 15%). The remaining items (5% or less of the cumulative volume) are assigned type C.
- Series Count (#). The Series Count (#) option allows you to assign specific numbers of items to each group. For example, if you have 1,000 items and set the A group to 100, the B group to 200 and the C group to 700—the 100 highest volume items will be assigned to group A, the next highest 200 items will be assigned to group B and the remaining 700 will be assigned to group C.
- Series Count (%). The Series Count (%) assigns the specified percentage of the total number of items you have to each group. For example, if A is set to 20%, B is set to 50% and C is set to 30%, Forecast Pro TRAC will first rank all of the items from highest volume to lowest volume and then sequentially place the top ranked items into the A group until the number of A group items equals 20% of the total number of items. It will then sequentially place items into the B group until the number of B group items equals 50% of the total number of items. The remaining items (30% of the total number of series) are assigned type C.
Finally, the Group Zero Volume check box, when selected, will put any items with zero volume in a fourth group labeled “D.”
In our example, we have instructed Forecast Pro to rank the SKUs based on volume, with group A comprising 75% of total volume, group B the next 15% and group C the remaining 10%. We’ve also elected to include a group D for zero volume items.
Forecast Pro TRAC can display the automatically generated ABC classifications and associated statistics in any of the five report views. In the Item Report Design dialog box shown below, you can see the controls used to specify which Pareto codes and statistics to display in the Item Report.
You can use the Pareto rankings and associated statistics to sort and filter your reports. The screenshot below shows the Item Report, filtered so that we are only viewing SKUs, and sorted according to ABC rankings.
The Pareto field displays the assigned code. The % of Total field displays the percentage of total volume that the item represents. The Ranking field is self explanatory. The Cumulative % field displays the percentage of total volume for the current item plus all higher-ranked items.
Custom Pareto Classification
Forecast Pro TRAC Version 4 allows you to define custom filter fields in a separate input file which can be used to filter reports. Among other uses, a custom filter field can be used to define Pareto codes. Since you define the codes, you are not limited to using only A, B and C categories, or basing them on volume—you can have as many different codes as you wish and base them on any criteria that you choose.
In Forecast Pro TRAC, you can display this custom field in any report. For example, in the Item Report Design dialog box, you can turn on custom fields using the Select filter attributes to display drop down menu (shown below). All report design dialog boxes have this control.
As with the automatically generated ABC codes, you can sort and filter based on these customized codes.
Pareto analysis is a useful tool for evaluating your forecast items and prioritizing your forecast review process. You can avoid wasting valuable time and resources by focusing efforts on your “big hitters.” Forecast Pro TRAC can generate ABC codes automatically based on volume, or you can specify your own codes using a custom filter.
Custom Pareto filters are just one of the useful features in Forecast Pro TRAC Version 4. Click here to schedule a customized live WebEx-based demonstration of Forecast Pro TRAC to learn more about how the solution can help to improve your forecasting.