Scheduling rules

You can review the parameters related to the sourcing of products only. There are other parameters defined as part of scheduling rules that are used for date determination purposes.

Scheduling rules are set up at an Enterprise level. Sterling Order Management System Software uses the rule defined by the Enterprise of the order transaction. When using an on-the-fly inquiry, Sterling Order Management System Software uses the rule defined by the primary Enterprise of the organization code making the request.

Use geography?

This parameter can be used to turn node prioritization on or off based on the distance between the node and the ship-to location. If "Use geography" is set to "Yes" and you are optimizing node selection based on "Priority", Sterling Order Management System Software calculates the node priority as:


Weight factor for distance * distance in miles as calculated based on longitude 
and latitude + Weight factor for priority * priority setup in the distribution 
node group.

This combined priority is used to select the node that has the lowest priority number. The weight factors are also set up as part of the scheduling rules.

If you want to give first preference to the node priority setup, you would want to set up the weight factor of "Priority" as 100,000 irrespective of the distance that is given the first preference. When the priorities of two nodes are the same, the distance is the tiebreaker. If you want to give distance more importance, you set up the weight factor for priority as 0 or another low number. You can work out the weight factors and priority numbers you want to use based on the fact that the distance (in miles) calculation is done internally.

Optimization type

When multiple nodes and dates are available for sourcing, this parameter can be used to make sourcing selections based on:

  • Priority of nodes
  • Distance of nodes from the ship-to location
  • Date when the delivery can be made
  • Number of resultant shipments
  • Cost-based scheduling

The optimization type that is required can be set as part of the scheduling rules. The following optimization types can be set:

  • Priority - If this is set, node selection is based on the priority setup in the distribution group and the distance from the node to the Ship-To location. Priority is calculated using Weight of node priority * Node priority + Weight of distance * Distance from the node to the ship-to location. Distance calculation is done based on the longitude and latitude of the two locations, and is only a "straight-line" distance. Distance calculation has an approximately 20% error margin from the actual road distance in most scenarios.
  • Distance of nodes from the Ship-To location - When optimizing a scheduling solution, the schedule can take into consideration the node priority cost. The priority of the node is converted into schedule cost using the priority cost factor. The product of the node priority and the priority cost factor results in additional cost for using that particular node. For example:

    Distribution Group Passed on Line DG1: N1 – Priority 1, N2 - Priority 10.

    Final leg shipping cost from N1 is $10, from N2 it is $2 (assuming no other costs are being used).

    The optimum solution is N2.

    After the node priority cost factor is enabled,

    Priority Cost Factor - $1

    Option1 - N1 - 1*$1+$10=$11

    Option2 - N2 - 10*$1+$2=$12

    The optimum solution is N1 as it has the lesser cost and a low priority of node.

  • Date - Sterling Order Management System Software selects the node that can make the earliest delivery of the product. Delivery date is calculated based on the transit time calculated between the shipping and ship-to locations.
  • Number of shipments - Shipment date and node selection is done in a way that it reduces the total number of shipments finally made. This is also accomplished by consolidating the shipments against future inventory. As a result, the shipment might get delayed beyond what is informed to the customer. To address this issue, you can configure maximum number of days (delay window) by which a shipment can be delayed.
    Figure 1. Date range optimization logic
    Date_Range_Optimization_Logic

    The previous figure illustrates how the windows refer to the optimization date range, which is configured as 5 days. The items are available on various dates. Item1 is available on 2/3/06 and continues to be available until 2/15/06. Similarly, the Item2 is available on 2/6/06 and Item3 on 2/8/06 until 2/15/06. Item 4 is available only on 2/15/06. If the customer places an order for all the items, and has agreed to wait for the shipment, the items are consolidated in one shipment and is shipped on 2/15/06 which reduces the cost of multiple shipments. This applies the rule of Date Range Optimization Logic. But if the customer wants the items to be shipped when they are available, the shipment optimization is considered and the customer is shipped items1 and 2 at the earliest date 2/6/06 which falls within a single window.

    A Delay window can be specified at the item level to account for high ticket items. This window can be set to 0 days which means that the product is never delayed. You can also choose not to delay shipments to be consolidated with future inventory since information (dates) about future inventory is often not accurate. This ensures that onhand and procured inventory is shipped immediately.

  • Cost-based scheduling - Cost-based optimization enables the scheduling process to consider landed cost when determining a fulfillment option. Landed cost consists of the following costs:
    • Item Inventory Cost uses the average cost maintained in Sterling Order Management System Software to compute a unit cost of an item at a given node.
    • Node Capacity Cost Factor - This factor pertains to the consumed capacity percentage of inventory for a node. During scheduling optimization, when the system compares two possible nodes to determine the best one to use during costing, the landed cost will be based on the lowest consumption factor.
    • Handling Cost, both input and outbound handling costs, can be configured per unit, per line, or per shipment. Inbound cost is considered when procurement is selected during scheduling, and is applied at the procure-to node. Outbound handling cost is considered for outbound shipments and is applied at the ship node.
    • Transfer Cost is considered when a procurement option is selected when scheduling. Transfer cost can be configured per unit, per unit weight, per unit distance. Alternatively, a fixed transfer cost can be specified for an individual transfer schedule.
    • Final Leg Cost is used to calculate the cost of a final shipment to a customer. It is usually computed for a specified carrier or carrier service. A user exit is provided to compute the final leg cost.
    • *Hours of Supply Cost Factor - This factor pertains to the hours of supply cost for a node. During scheduling optimization, when the system compares two possible nodes to determine the best one to use during costing, the hours of supply cost factor determines the weight that is assigned to the hours of supply cost in the cost calculation.
    • *Shipment Delay Cost Factor - This factor pertains to the shipment delay cost for a node. During scheduling optimization, when the system compares two possible nodes to determine the best one to use during costing, the shipment delay cost factor determines the weight that is assigned to the shipment delay cost in the cost calculation.
    The *Node attribute Costs Calculation Per Unit option in the Landed Cost window enables per-unit calculation of node attribute costs, which include hours of supply cost, shipment delay cost, node priority cost, and node consumption cost.

    When scheduling, the landed cost options are evaluated to determine the least landed cost, which is used as the cost-based option. Landed cost is prorated to reflect landed cost per unit. The priority is configured in the scheduling rules. In addition to the various costs listed previously, the priority can be converted into cost using the cost factor. For example, if a node within a distribution group has a lower priority (higher priority number), the node will contribute more to the cost, compared to a higher priority node.

    To determine the best cost-based optimization type, a higher number of solutions should be evaluated. To evaluate a higher number of solutions, you can increase the MaximumRecords in the promising and scheduling APIs. If date optimization is used, cost-based scheduling will not be used because the fastest option will be considered first.

Cost-based scheduling: An example

In the following example, landed cost is computed for a given option. Let us assume that a customer orders a DVD player and three DVDs. The DVD player is available at DC1, and the DVDs are available at DC2. In this case, the schedule has two options:

  • Option 1 is to ship individual items to the customer as two separate shipments from DC1 and DC2.
  • Option 2 is to transfer both items to DC3, and then ship a single shipment to the customer from DC3.

Because option 2 is less expensive, it is selected by the schedule as the cost-based schedule option. The options are described in the following sections.

Option 1

Assuming that all the costs are configured, the landed cost for option 1 is computed as follows:

  • In shipment 1, the DVD player is shipped directly to the customer. The following equation is used to calculate landed cost for shipment 1:

    (Total Item Cost + Outbound Handling Cost + Final Leg Cost) / Total Units

    If this equation is used, the landed cost for shipment 1 is $120.50, as shown in the following calculation:

    (1 unit * $100 (average cost) + $3 (per shipment) + $17.50 (ground shipping to customer)) / 1 unit = $120.50

  • In shipment 2, three DVDs are shipped from DC3 directly to the customer. The following equation is used to calculate the landed cost for shipment 2:

    (Item Cost + Outbound Handling Cost + Final Leg Cost) / Total Units

    If this equation is used, the landed cost for shipment 2 is $14.33, as shown in the following calculation:

    (3 units * $10 (average cost) + $3 (per shipment) + $10 (ground shipping to customer)) / 3 units = $14.33

The calculations for shipment 1 and shipment 2 are added together in order to determine the landed cost to fulfill option 1, which is $124.83.

Option 2

Assuming that all the costs are configured, the landed cost for option 2 is computed as follows:

  • In shipment 1, the DVD player is shipped from DC1 to DC3. The following equation is used to calculate the landed cost for shipment 1:

    (Total Item Cost + Outbound Handling Cost) / Total Units

    If this equation is used, the landed cost for shipment 2 is $106, as shown in the following calculation:

    (1 Unit * $100 (average cost) + $3 (per shipment) + $2 (fixed transfer cost between DC1 and DC3) + $1 (per shipment at DC3)) / 1 Unit = $106

  • In shipment 2, the DVDs are shipped from DC2 to DC3. The following equation is used to calculate the landed cost for shipment 2:

    (Total Item Cost + Outbound Handling Cost + Transfer Cost + Inbound Handling Cost) / Total Units

    If this equation is used, the landed cost for shipment 2 is $11.67, as shown in the following calculation:

    (3 Units * $10 (average cost) + $3 (per shipment) + $1 (fixed transfer cost between DC2 and DC3) + $1 (per shipment at DC3)) / 3 Units = $11.67

  • In shipment 3, the DVD player and DVDs are shipped from DC3 to the customer. The following equation is used to calculate the landed cost for shipment 3:

    (No Inventory Cost (items/are procured) + Outbound Handling Cost + Final Leg Cost) / 4 Units

    If this equation is used, the landed cost for shipment 3 is $3.33, as shown in the following calculation:

    ($3 (per shipment) + $7 (ground shipping to customer from DC3)) / 3 Units = $3.33

The calculations for shipment 1, shipment 2, and shipment 3 are added together in order to determine the landed cost to fulfill option 2, which is $121.

Ship complete order

This parameter ensures that all product lines in the promising inquiry request are either completely scheduled or not scheduled at all. Lines could, however, be sourced from different shipping locations.

Ship order from single ship node

This parameter ensures that all product lines in the promising inquiry request are either completely scheduled or not scheduled at all. It also ensures that the complete request is sourced from a single node on a single date. This is a super set of "ship complete order" and when this parameter is set, a "ship complete" is assumed.

Ship complete line

This parameter ensures that every product line on an individual line basis is either completely sourced or not sourced at all. However, lines could be sourced from different shipping locations. The difference between this rule and the "ship complete order" rule is that this rule does not enforce that all lines of the request are completely sourced. One particular line can be sourced while another line of the same request could be backordered.

Ship line from single ship node

This parameter ensures that every product line in the inquiry request is either completely scheduled or not scheduled at all. It also ensures that each individual line is sourced from a single node on a single date. This is a super set of "ship complete line" and when this parameter is set, a "ship complete line" is assumed. However, this rule does not enforce that all lines are shipped from the same node. A particular line may be completely shipped from node 1 while another line could be completely shipped from node 2.