The inventory order point is the basis for establishing a companies optimal inventory levels (the level between too much and too little) when the following variables are taken into account.

• Replenishment lead time: Normal time from order placement to receipt for both production orders and purchase orders
• Volatility of historical demand: Weekly standard deviation of demand
• Projected (forecasted) demand during the lead time: Statistical and/or user defined forecast over the items replenishment lead time.
• Desired level of customer service: This is the Z-Factor / service level factor used to calculate the statistical safety stock.

The order point is the trigger that tells the system it is time to reorder. Some call this the min.

As can be seen in the following graphic, there is a cost to carrying inventory. As such, DemandCaster follows a process that segments your data to help you apply the most optimal policies to help meet your companies Cost, Inventory, Service, and Quality objectives taking into consideration the business supply chain network and overall complexity.

The order point is managed in the Order Point Calc Analytic and is calculated by adding the forecast during the lead time plus the safety stock. The formula is:

Order Point = Forecast + Safety Stock (SS)

Safety Stock is a buffer to accommodate natural variations in demand and supply. DemandCaster has multiple methods to calculate safety stock however, for this exercise, we will apply the weekly standard deviation method using the following formula.

Square Root of Lead Time * Service Level * Standard Deviation of weekly demand

In short, the longer the lead time and higher the standard deviation the higher the safety stock required to maintain a desired level of service. To maintain high levels of service with longer lead times and variation, one must increase safety stock. However, increasing safety stock may improve customer service but hurts the cash flow!

If one desires to reduce inventory, the first thing that needs to be investigated is reducing lead time which is something that is often within an operations control. Thereafter, the items demand volatility may be addressed but this is more challenging since it typically lies outside of the control of operations. S&OP is often used to help manage volatility in addition to different types of vendor managed inventory processes or data sharing.

The second component of the order point is the items projected demand / forecast. DemandCaster uses numerous statistical methods to automatically calculate a best fit forecast. The forecast can also be user defined via a manual entry into DemandCaster, uploaded via a spreadsheet, or pulled in from a customer system.  If the forecast is created from within DemandCaster it is desired to have three years of history to help identify seasonality though less history may be used.

The images below illustrate how changes in demand volatility, service level, and longer lead times impact order points.

Though there are exceptions to this general approach, once the order point is established, DemandCaster suggests replenishment orders based on two distinct time horizons:

• During Lead Time
• Beyond Lead Time

Within lead time, DemandCaster sums the the quantity on hand and on orders and then subtracts open orders during lead time. Thereafter it compares the net available against the order point. If the net available is less than the order point, an order quantity is suggested. If it is greater, the calculation progresses the requirement week after week following the lead time until the projected inventory breaches the safety stock or 0. At that point a suggested order is created.