| Multidimensional Data Management
The Problem
In complex business environments, individuals who analyze demand data and use forecast results are normally interested in seeing the information presented in a way that helps them better manage certain aspects of their business, such as sales territories, major customers or product lines. As part of that process, they need an easy, efficient way to focus in and segment the data stored in their database systems.
Consequently, in today’s business world, there is usually no such thing as a simple, stand-alone item that is managed and forecasted in isolation. Each of your product items may be characterized by multiple dimensions, such as SKU number, sales region, market segment, customer, product line, etc. Demand data for each item will be measured and collected across all dimensions. In addition, other information, such as conversion factors, forecast parameter values and promotional scenarios, may need to be stored in your database and managed on an item-specific basis.
Furthermore, if your items are grouped into product lines, you may need to “filter” the data to select relevant subsets for special forecasting treatment and display. In handling product lines, you must be careful to forecast each item with an eye to the forecasts of all other items in the group. For instance, increased sales of one item in the product line may imply lower or higher sales of other items. Another possible complication is that you may need to keep switching focus between several dimensions, such as sales regions and market segments.
The Smart Solution
|
 |
| A multidimensional data table in SmartForecasts
Click here for full-size screenshot
|
With SmartForecasts’ multidimensional data management capability, you can label each item with as many dimensions as you require. Then you can use its special OLAP Data Filtering feature to “slice and dice” your database, selecting relevant subsets of the data for special treatment and display. For instance, you could focus in on an important product line and then use SmartForecasts’ Multilevel (Multiseries) forecasting feature to create either bottom-up or top-down forecasts of all the items in the product line and their total. Or you might want to key in on only those items with the highest average monthly demand, or with a large percentage of zero values (intermittent demand), or with short lead times and high service level requirements.
The Benefits
With SmartForecasts, you can drill down in multidimensional databases to focus on key subsets of data, such as particular markets or product lines. Items in product lines can be forecasted more accurately by taking advantage of any patterns of interaction among items or the greater statistical stability of aggregated data. The net result is a more focused, efficient and accurate forecasting process, with information presented in a form that enables you to better manage your business.
|
