Two phase predictive approach for supply network optimization
A method, system and computer program product for managing an order in an Omni-channel order fulfillment system is disclosed. A stock-keeping unit (SKU) node level cost is predicted for a plurality of SKUs. The order is received containing one or more SKUs. A candidate list of fulfillment nodes is determined for fulfilling each of the one or more SKUs in the order using the predicted SKU node level cost and a fulfillment node-destination shipping distance. One or more fulfillment nodes are selected from the candidate list, and the order is fulfilled using the selected one or more fulfillment nodes.
1. A computer-implemented method for reducing a cost of fulfilling an order in an Omni-channel order fulfillment system, the method comprising:
predicting, by a processor, a stock-keeping unit (SKU) node level cost for a plurality of SKUs, wherein the SKU node level cost includes a markdown cost for the SKU;
receiving the order that contains one or more SKUs;
determining a master list of fulfillment nodes available for providing the one or more SKUs to fulfill the order;
determining a fulfillment cost for each of the fulfillment nodes in the master list for each of the one or more SKUs, wherein the fulfillment cost for a selected node is a summation of the predicted SKU node level cost and a weighted fulfillment node-destination shipping distance cost;
selecting, from the master list of fulfillment nodes, a candidate list of fulfillment nodes for fulfilling each of the one or more SKUs in the order, the candidate list including a selected number of fulfillment nodes having a lowest fulfillment cost, wherein the candidate list is shorter than the master list;
providing the candidate list and a shipping database to an optimization program to select one or more fulfillment nodes from the candidate list;
causing, by the processor, the one or more SKUs of the order to ship from the selected one or more fulfillment nodes to fulfill the order; and
selecting the one or more fulfillment nodes from the candidate list using a mixed integer optimization program, where using the candidate list in the mixed integer optimization program reduces a processing time for fulfilling the order.
2. The method of claim 1 , wherein predicting the SKU node level cost further comprises estimating business operational costs for the SKU at each fulfillment node.
3. The method of claim 2 , wherein the business operational costs further comprise at least one of: inventory performance costs, load balancing costs, daily operating costs, and costs for shipping from fulfillment node to customer.
4. The method of claim 1 , further comprising determining the candidate list based on a cost computed based on the predicted SKU node level cost and the fulfillment node-destination shipping distance.
5. The method of claim 4 , further comprising selected a relative weight of the predicted SKU node level cost and the fulfillment node-destination shipping distance in determining the candidate list.
6. A system for reducing a cost of fulfilling an order for an Omni-channel retailer, the system comprising:
a memory having computer readable instructions; and
one or more processors for executing the computer readable instructions, the computer readable instructions comprising:
predicting a stock-keeping unit (SKU) node level cost for a plurality of SKUs, wherein the SKU node level cost includes a markdown cost for the SKU;
receiving the order that contains one or more SKUs;
determining a master list of fulfillment nodes available for providing the one or more SKUs to fulfill the order;
determining a fulfillment cost for each of the fulfillment nodes in the master list for each of the one or more SKUs, wherein the fulfillment cost for a selected node is a summation of the predicted SKU level cost and a weighted fulfillment node-destination shipping distance cost;
selecting, from the master list of fulfillment nodes, a candidate list of fulfillment nodes for fulfilling each of the one or more SKUs in the order, the candidate list including a selected number of fulfillment nodes having a lowest fulfillment cost, wherein the candidate list is shorter than the master list;
providing the candidate list and a shipping database to an optimization program to select one or more fulfillment nodes from the candidate list;
causing the one or more SKUs of the order to ship from the selected one or more fulfillment nodes to fulfill the order; and
selecting the one or more fulfillment nodes from the candidate list using a mixed integer optimization program, where using the candidate list in the mixed integer optimization program reduces a processing time for fulfilling the order.
7. The system of claim 6 , wherein predicting the SKU node level cost further comprises estimating business operational costs for the SKU at each fulfillment node.
8. The system of claim 7 , wherein the business operational costs further comprise at least one of: inventory performance costs, load balancing costs, daily operating costs, and costs for shipping from fulfillment node to customer.
9. The system of claim 6 , further comprising determining the candidate list based on a cost computed based on the predicted SKU node level costs and the fulfillment node-destination shipping distances.
10. The system of claim 9 , further comprising selected a relative weight of the predicate SKU node level costs and the fulfillment node-destination shipping distances in determining the candidate list.
11. A computer program product for reducing a cost of fulfilling an order in an Omni-channel order fulfillment system, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform:
predicting a stock-keeping unit (SKU) node level cost for a plurality of SKUs, wherein the SKU node level cost includes a markdown cost for the SKU;
receiving the order that contains one or more SKUs;
determining a master list of fulfillment nodes available for providing the one or more SKUs to fulfill the order;
determining a fulfillment cost for each of the fulfillment nodes in the master list for each of the one or more SKUs, wherein the fulfillment cost for a selected node is a summation of the predicted SKU node level cost and a weighted fulfillment node-destination shipping distance cost;
selecting, from the master list of fulfillment nodes, a candidate list of fulfillment nodes for fulfilling each of the one or more SKUs in the order, the candidate list including a selected number of fulfillment nodes having a lowest fulfillment cost, wherein the candidate list is shorter than the master list;
providing the candidate list and a shipping database to an optimization program to select one or more fulfillment nodes from the candidate list; and
causing the one or more SKUs of the order to ship from to the selected fulfillment nodes to fulfill the order; and
selecting the one or more fulfillment nodes from the candidate list using a mixed integer optimization program, where using the candidate list in the mixed integer optimization program reduces a processing time for fulfilling the order.
12. The computer program product of claim 11 , wherein predicting the SKU node level cost further comprises estimating business operational costs for the SKU at each fulfillment node.
13. The computer program product of claim 12 , wherein the business operational costs further comprise at least one of: inventory performance costs, load balancing costs, daily operating costs, and costs for shipping from fulfillment node to customer.
14. The computer program product of claim 11 , further comprising determining the candidate list based on a cost computed based on the predicted SKU node level cost and the fulfillment node-destination shipping distance.
15. The computer program product of claim 14 , further comprising selected a relative weight of the predicate SKU node level cost and the fulfillment node-destination shipping distance in determining the candidate list.