Here is the presentation shown during the client visit and project kick-off (pptx format, 13.3MB).
The company
School Specialty (SSI) distributes school supplies to K-12 teachers throughout the US. Their distribution network includes warehouses in Mansfield, Ohio; Lancaster, Pennsylvania; Greenville, Wisconsin (site of the corporate headquarters); Salina, Kansas; and Fresno, California. Most of their business is concentrated in May and June, when teachers order supplies in advance of the new school year beginning in August.
Small shipments are sent via parcel carrier and larger shipments as freight. UPS manages outbound shipments of both parcel and freight.
During May and June, almost half of the lines are sent as freight, because classrooms are stocking up on supplies. During the rest of the year, orders are smaller and few lines are sent as freight.
Storage
Popular product that moves in large volume is picked from carton flow rack, while slower-moving product is picked from static bin shelving. Pallet rack is used for bulk storage.
Warehouse operations
Each order is broken into pick lines for full cartons and pick lines for less-than-carton quantities. Then the less-than-carton lines are assigned to cartons (that is, a particular box size and a “license plate”). Then a pick-list is prepared for each carton.
Less-than-carton quantities are picked from the flow rack or bin shelving directly into shipping cartons.
Cartons are picked from bulk storage.
Goal of this project
- What is the best mix of carton sizes to reduce shipping costs and costs of boxes and dunnage?
Data
The company data is copyrighted and proprietary. You may use
it for the purposes of this course only. If you would like to use it
for something else, please contact me.
Data will be provided on CD directly to project groups. Here is what to expect:
The Order History represents the peak season (May, June, July and August) and a couple of non-peak months (December and January). This information is segregated by DC and includes the following columns:
- Month: The fiscal month when the order was shopped
- DC: The warehouse code values are: 10 = Greenville, 74 = Lancaster, 75 = Mansfield, 78 = Fresno, 80 = Salina
- Shipper: The WMS Pick ticket number or internal order number
- Order Nbr. The Host Order number dropped to the warehouse
- Sub: The Order number Release number
- Ctn ID: The Container Number there will be multiple lines of the same container based on the items in the container.
- Ctn: The container identifier
- WRKTYPE: The task type utilized by the WMS to separated different work types
- PickLoc: For each container line, indicates the physical location where product was picked in the warehouse.
- Wt: This is the actual weight of the container . The weight amount will be the same on all lines of the container.
- Frt: The recorded Freight cost for each container against values should be grouped at the container level. This would be rack rates in the WMS system.
- City: City where shipped
- St: State where shipped
- Zip: 5 Byte zip where shipped
- SKU: Item identifier on the line = Item_ID from item files
- Nbr: Number of Units sold on the line
- Ctn Count: Total Cartons shipped on the order
- Carrier: The carrier SCAC under which the order shipped.
- PRO: If the carrier was an LTL or TL carrier the PRO number for the order
- Track: UPS or FedEx tracking number from the container if shipped Parcel
- CC: The containerization Category assigned to the items and the container to determine how items are grouped into containers.
The Item Tables includes the following columns:
- DC: Same codes as above
- Item_ID: The SKU items that are part of the shipments sent
- Cartonisation_category: The containerization Category assigned to the items and the container to determine how items are grouped into containers.
- Pack_height: For the selling unit of measure height of a single unit in inches
- Pack_length: For the selling unit of measure Length of a single unit in inches
- Pack_width: For the selling unit of measure Width of a single unit in inches
- Pack_volume: Calculation length x width x height in cubic inches
- FC_Carton_units: The factory carton units for the items.
The Container file lists carton sizes, and includes the following columns:
- Whse: The codes indicate the DC
- Cartonisation_category: The containerization Category assigned both the items and the container to determine how items are grouped into containers.
- Carton Type: The container identifier
- Description: The container description
- Height: Physical Height of the container
- Width: Physical Width of the container
- Length: Physical Length of the container
- Volume: Calculation length x width x height in cubic inches
- Weight: Weight Limit for the container
In addition you will find the UPS zone tables by DC origin and the UPS rate structure for every carton from 1–70 pounds by zone, and estimated annual usage and cost of cartons.
Suggested approach
To reduce shipping costs, pack items into fewer and smaller boxes. You can do this by either choosing a more effective mix of box sizes (including new sizes) or by changing how the SKUs are assigned to boxes. (The latter may be problematical, as we shall discuss.)
First steps
- Read the chapter in the textbook on warehouse activity profiling
and set up your database tools.
- Begin to explore the data. Build a “bird's
eye view” of the warehouse to show the most-frequently
requested SKUs, the SKUs that move in largest volumes, etc. Look for product affinities.
- See the UPS web site for shipping policies and prices.
- Understand the cubing algorithm employed by the WMS. You will likely need to code this logic in order to evaluate effectiveness of collections of box sizes. Suggestion: Java.
- Think about how moving to larger or smaller cartons would change warehouse processes and performance.
- Think about how to search over a collection of box sizes to find the most effective.