Sourcing rule determination

You might create conflicting sets of sourcing rules by leaving some of the parameters void. Sterling Order Management System resolves such conflicts by using a priority order for the key parameters.

The priority order Sterling Order Management System uses is:

  1. Seller Organization
  2. Order Sourcing Classification
  3. Item ID
  4. Primary Item Classification
  5. Secondary Item Classification
  6. Tertiary Item Classification
  7. Geographical Region of the ship-to location

Fulfillment type is matched exactly with the order line parameter or inquiry parameter.

If you specify just "Item ID" for one rule and "Primary Item Classification" for another rule and both rules match the values for the product being sourced, Sterling Order Management System chooses the rule specified for the Item ID over the one specified for Primary Item Classification. It also gives preference to the rule where a value is specified for the parameter over the rule where the value is left blank.

Example 1: sourcing rules

This example describes a basic sourcing scenario. Your sourcing rule parameters are defined as described in the following table.

Table 1. Sourcing rule parameters for preference example
Rule# Item ID Primary item classification Geographical region
1 ITEM1 - -
2 - TV -
3 ITEM1   California
4 - - California
5 ITEM1   San Francisco, CA

If you are sourcing an order line with its Item ID, Item Classification and Shipping Location values of ITEM1, TV, and New York, Sterling Order Management System uses rule #1. Even though rule #2 also meets the criteria, rule #1 is more specific over the other rule. The Item ID parameter is given higher preference over the Primary Item Classification parameter.

If you are sourcing an order line with its Item ID, Primary Item Classification, respectively, and Shipping Location values of ITEM1, TV, and Los Angeles-CA, Sterling Order Management System uses rule #3. Even though rules #1, 2, and 4 also meet the order line criteria, rule #3 is given higher priority over the other rules because each of these rules have some parameters as blank (void) and rule #3 does not. The Item ID parameter is given higher preference over the Geographical Region parameter.

If you are sourcing an order line with its Item ID, Primary Item Classification, and Shipping Location values of ITEM1, TV, and San Francisco-California, respectively, Sterling Order Management System uses rule #5. Even though rules #1, 2, 3, and 4 also meet the order line criteria, rule #5 is given higher priority over the other rules because each of these rules have some parameters as blank (void) compared to rule #5. Rule #5 is given higher preference than rule #3 because the region specified in rule #5 is more specific compared to rule #3's region.

Example 2: sourcing rule region hierarchy

When selecting a sourcing rule, the rule matching the lowest level region is given higher preference over any rule that has a higher-level region specified.

A hierarchy of region is defined as:

Source_Rule_Region_Hierarchy

Assume that the sourcing rules are defined as follows:

  • Use "Node1" for shipping to any address in New York
  • Use "Node2" for shipping to any address in Downtown Manhattan.

The Sterling Order Management System node selection for the shipping address chooses Node2 when shipping to Downtown Manhattan and Node1 for any address in Manhattan other than Downtown Manhattan.

Using sequence of sourcing templates

For each sourcing rule, you can specify a sequence of sourcing templates to be used for sourcing the order line.

Sterling Order Management System tries to source the product from the highest sequence (lowest number) sourcing template. When multiple choices are available to Sterling Order Management System, node selection is optimized based on settings in the scheduling rule associated with the order.

If there is no available product in the sourcing template specified in this sequence, Sterling Order Management System tries to source from the next sequenced sourcing template.

When would you want to use sequencing?

In certain situations you may want to have more control of the optimization logic used by Sterling Order Management System for sourcing.

For example, you need to ship to a location in El Paso, Texas. You have multiple warehouses in North America. You want to first look into United States warehouses and only if there is no inventory available in the United States warehouses would you want to look in the Mexican warehouses. If you specify all North American locations into the same sequence, Sterling Order Management System optimizes on the distance between the ship-from and ship-to locations (this can also be controlled to optimize on other parameters) and that may result in shipment from a Mexican warehouse. Being able to control the sequence of distribution groups to consider enables you to handle such geopolitical situations.

Another reason to provide specific sequences is for system performance reasons.

Suppose a shipment needs to be sent to a New York address. You have 30 to 40 distribution centers spread across the U.S. and you want to make the shipment from the closest distribution center that has the product. You could group your shipping nodes into 4 to 5 distribution node groups based on geographical regions such as east coast nodes, west coast nodes, central U.S., and southern U.S.

For this particular situation you would set up the sequence so that Sterling Order Management System first tries to source from the East Coast nodes and then the subsequent regions. This is illustrated in the following figure. Notice that the same node can be included as part of multiple groups.

Figure 1. Sequencing sourcing templates - example 2
Sequencing_Nodes2

One obvious question in the above setup is why shouldn't you just specify a single sequence and associate a distribution node group that includes all United States locations? Since Sterling Order Management System already optimizes to select the closest location, this requires less setup with the same results.

The primary reason why you would want to set this up as multiple sequences is to prevent Sterling Order Management System from performing redundant inventory searches and thus obtain better response times from the promising functions. Since Sterling Order Management System optimizes the selection between all the specified nodes, it needs to read inventory in all the specified nodes before it makes a final selection. If all the United States locations were to be specified, Sterling Order Management System reads the inventory of every location before making the final selection. If you can create a smaller subset of nodes to look into, it in all likelihood save you system resources and provide better response times. The choice between the lesser setup and better performance is mainly based on how many locations you are searching. It is reasonable to put 5 to 10 locations in a single sequence. Any more locations in a single sequence could result in some performance degradation.

Priority of nodes when using multiple sequences

Sterling Order Management System looks at each of the sequences set up in the sourcing rule to source the product until its finds product availability. The priority of an individual node at any given time is calculated as the highest priority of the node amongst all the groups in the sequence. If a node is explicitly mentioned as one of the sequences, priority is treated as 0 (highest).

For each sourcing rule, you can also specify a sequence which enhances scheduling that considers future inventory based on sourcing rule configuration. If preferred substitute is configured for an item, then that needs to be shipped before the onhand inventory. For example, consider:

Sequence number Sequence
Sequence 1 Source from Regional DC
  Consider future inventory that is coming in the next 5 days
  Substitution allowed
Sequence 2 Source from Regional DC
  Consider future inventory that is coming in the next 30 days
  Substitution allowed

The Onhand inventory is 5 and the Preferred substitute is 5 which would be coming in 2 days and 10 in 20 days.

If Customer1 orders 5 items, then the preferred substitute is shipped to the customer as it is coming within 5 days.

When Customer2 orders 5 more items, then the onhand inventory is shipped as the preferred substitute now has 0.

Now, when Customer3 orders 5 items, he gets them within 30 days as the preferred substitute and onhand inventory now has 0.

Therefore, these configurations prioritize selection of additional nodes and looking at future inventory.