About working with inventory control codes

Records in the inventory control code tables are identified by two keys. First they belong to one of the classes described above. Second, they must be identified by a key value, which can be:

• Item code for records of the item class
• Item group code for a record of the item group class
• Supplier number for records of the supplier class
• Volume value code for records of the volume value class
• Movability code for records of the movability code class
• A three position combined field of volume value code and movability code for records of the volume & movability class
• An arbitrary name for records of the nickname class
• A constant for the default class

As part of the key value key, you can also define a warehouse. By defining a warehouse, the entry you create applies only to that warehouse. By leaving the warehouse field blank, the entry you are defining applies to the entire company. This means that you can manage your stock at the warehouse level, or the company level, or a combination of both.

You define the alpha value by completing the Alpha value field in the Forecast control code table. The defined alpha value will be used for every period.

The value of the alpha depends on the length of the forecast period. The following table can be used as a guideline when you define an alpha value manually. Look at the values for the forecast period length that you have established for your company:

If Auto alpha update is set to YES in the Forecast control code table, then the system automatically updates the alpha value when the inventory calculation is run. A new alpha value will be calculated for each forecast period and stored it in the Forecast file per item/warehouse and forecast period.

You can control how low and how high the automatic alpha value will be by completing the Minimum alpha and Maximum alpha fields. The minimum alpha must be in the range of 0.01 to 1.00, and the maximum alpha must be greater than the minimum alpha. Each time the system calculates a new alpha value, it checks that it is greater than the minimum alpha and less than the maximum alpha.

The system uses the forecast error information as input for the calculation of the new alpha value. If the forecast error is large, the system increases the alpha value to get a faster adjustment to true demand. When the forecast error is small, the system maintains a low alpha to keep a stable forecast that adjusts slowly to changes in demand.

The size of the forecast error is retrieved from the absolute value of the tracking signal. The tracking signal is close to zero when there is a small forecast error and close to 1 when there is a large forecast error.

Note: See About working with forecasts for information about the forecast error calculation.

The Wilson method seeks to minimise the total cost of replenishing stock. There are two cost components involved. These are the ordering cost which is the estimated total cost for placing one order line, and the stock carrying cost which is the cost of having capital tied up in stock.

Ordering cost and the cost of having stock on hand are inversely related. That is, to minimise ordering costs you will need to purchase large quantities per order, reducing the total number of orders. But by reducing ordering costs in this manner, you increase the number of stock you have on hand, and increase the cost of tied up capital.

The Wilson method finds the optimum balance between these two costs. It adds the cost components together and selects the quantity (EOQ) that gives the lowest total cost.

This method is based on a fairly smooth demand pattern and functions best under such circumstances.

The Period dependent method is a modified version of the Wilson method that handles large fluctuations in demand. The Wilson formula is applied to calculate the best quantity to purchase. Wilson minimises the cost over one year, by analysing the year forecast from the Item file.

In the Period dependent method, the Wilson quantity is related to the total year forecast to get a time slice of one year that should be covered with every Wilson purchase. From this, you know how many days' consumption per average over the year to buy.

The final step is to look in the Forecast file to see what the expected demand is for the number of days that we want to cover at this time. This means that when the near future forecast is higher than average, the Period dependent method will give a higher quantity than the Wilson method. If the near future forecast is lower than average, the Period dependent method will yield a lower quantity than the Wilson method.

The Demand control method does not try to minimise cost as in the previous two methods. Instead, an average monthly forecast is calculated from the year forecast in the Item file. Using the average monthly forecast, the quantity to purchase is calculated as the forecasted demand during a specified number of months.

This is the theoretical method to calculate a statistical safety stock. It takes into account such factors as your service level, forecast uncertainty and lead time. The safety stock is based on the service level established in the Service control code table (at item/warehouse level), the length of the replenishment time, the size of the forecast error and the size of EOQ.

It will be high if you have long lead times, high uncertainty in your forecast, and high requested service level. It will be low if EOQ is high because a high EOQ means that you will maintain a high stock level. In this case, high EOQ acts as the buffer against unexpected demand that the safety stock would normally provide.

This method is used to set your safety stock. It is calculated as a specified percentage of the forecasted demand during lead time. It provides a simple way of calculating the safety stock.