Intended learning outcomes: Differentiate between usage statistics and turnover statistics. Identify an outlier and abnormal demand. Differentiate between sales transaction/statistics and bid transaction/statistics.
Statistics on particular events can provide an important basis for various calculations in requirements planning and inventory management.
Usage statistics analyze the quantity of all inventory transactions.
For each transaction, the following attributes should be recorded:
- Date of transaction
- Identification of the item or
the item family - Moved quantity
- Employees responsible for the
recording of the transaction - The two customer, production,
or purchase orders or inventory stock positions (target and actual, “before”
and “after” position of the transaction)
As the number of recorded transactions is usually very large, in practice it is often impossible to make older transactions available for online queries. Moreover, too much time could be required to process many queries, particularly those pertaining to entire groups of items.
Turnover statistics condense the most important data on inventory transactions to gain rapid information about an item’s movements.
Turnover statistics are updated, for example, daily, to include all transactions for that day. Managers maintain sales records for every item over the last statistical period, for example, the last 24 months and also over the three previous years. For all these periods, the following data are recorded:
- Total inventory issues, that is, items released from an inventory for use or for sale
- Partial inventory issues
- Inventory issues that were sold
- Total inventory receipts, that is, items released from an inventory for use or for sale
- Partial inventory receipts
- Inventory receipts that were
purchased or produced
For each of these attributes, depending on need and the data storage capacity of the system, the following can be recorded:
- Number of transactions
- Turnover expressed in quantity
- Turnover expressed in value
The attribute partial issues allows to record outliers or abnormal demand. (The analogous considerations apply to the attribute partial receipts.)
An outlier is a data point that differs significantly from other data for a similar phenomenon.
Abnormal demand — in any period — is demand that is outside the limits established by management policy. See [ASCM22].
For example, if the average sales for a product were 10 units per month, and one month the product had sales of 500 units, this data point might be considered an outlier. Abnormal demand may come from a new customer or from customers whose demand is increasing or decreasing. In general, outliers and abnormal demand should not be taken as a basis for demand forecasting. Care must be taken to evaluate the nature of the abnormal demand: Is it a volume change, is it related to the timing of some orders, or a change in product mix?[note 1102]
Usage and turnover statistics do not always suffice. This is the case when a relatively large time span lies between the demand to estimate and measured usage. A good example is capital goods having a considerable lead time of several months. In this case, we need statistics that are constructed in principle in the same way as the usage and turnover statistics above, but relate to more current events. A favorite measurement time point is the moment of sale or — even more up-to-date — the moment of bidding.
A sales transaction of an item records the sending of the order confirmation and thus the moment the customer order is accepted. Sales statistics analyze all sales transactions.
Sales statistics are more up-to-date than usage statistics — by the amount of the lead time for the order. However, the corresponding sales data files tend to be less precise, for customers may cancel or alter placed orders.
A bid transaction of an item records the sending of a bid to the customer. Bid statistics analyze all bid transactions.
Bid statistics are even more up-to-date than sales statistics due to the time that on average lapses between the formulation of the bid and the sale. But again, the corresponding data are less precise. In fact, the order success probability (see Section 5.2.1) shows the approximate percentage of bids that translated into sales. However, the latter cannot be ascertained reliably for every individual product, or even for every individual product family.
Course section 11.2: Subsections and their intended learning outcomes
11.2 Usage Statistics, ABC Classification, XYZ Classification, and Other Anlayses
Intended learning outcomes: Present statistics on inventory transactions, sales, and bid activities. Explain the ABC Classification and the Pareto Chart. Describe the XYZ Classification and Other Analyses and Statistics.
11.2.1 Statistics on Inventory Transactions, Sales, and Bid Activities
Intended learning outcomes: Differentiate between usage statistics and turnover statistics. Identify an outlier and abnormal demand. Differentiate between sales transaction/statistics and bid transaction/statistics.
11.2.2 The ABC Classification and the Pareto Chart
Intended learning outcomes: Explain the principle of the ABC classification, shown as a Pareto chart. Describe the ABC classification for each ABC category.
11.2.3 The XYZ Classification and Other Analyses and Statistics
Intended learning outcomes: Explain the XYZ classification. Produce an overview on exception lists.
