Design of a forward-pick area

Purpose

The forward-pick or fast-pick area is the most expensive and important area within the warehouse: It is the most conveniently located; it may have the most expensive equipment, such as pick-to-light systems; and it is where most of the labor is concentrated, particularly in small parts picking. Therefore it is essential to get the maximum value from this area.

To get maximum value from this area requires deciding:

• How big the fast-pick area should be; and
• How much of each sku to store in the fast-pick area

• If the fast-pick area is too large then there is more travel per pick, which means less savings per pick; but if the fast-pick area is too small then fewer sku's can be stored there and they will need to be restocked more often.
• If too much of a sku is stored in the fast-pick area then this leaves less space for other sku's; and if too little, then this sku must be restocked too often.

Case #1: An office-products wholesaler

This distributor receives electronic orders until early evening and then picks and ships in the night so that the product is at the customer's site at the start of business the next day. Because response time is critical, this DC set up a fast-pick area.

Data files

Figure 1: Bin-shelving filled with small parts that must be picked and shipped quickly.

Most items are relatively small (staplers, toner cartridges, pens, etc.) and are stored in bin-shelving. Here is a small sample of the sku's that we will use for this exercise (ASCII text format). The data fields are as follows:

• Sku ID: The unique identifier for this sku
• Length of the storage unit, in inches
• Width of the storage unit, in inches
• Height of the storage unit, in inches
• Pieces per case (that is, the standard number of pieces in each storage unit)
• Pieces sold during recent reporting period
• Picks (number of times this sku was requested by a customer) during recent reporting period
• Maximum number of storage units likely to be held in the warehouse

The fast-pick area for these sku's will be bin-shelving, 5 shelves to a section, with each shelf 12-3/4 inches high x 18 inches deep x 41.25 inches wide. Picks in the fast-pick area are estimated to average 0.5 minutes each, compared with 1 minute each in if picking from bulk storage. Restocks are estimated to average 1.75 minutes each.

Questions

Use the fluid model to answer the following questions. Unless instructed otherwise, ignore the issue of whether the amount to be stored is physically realizable.

Assume that you are considering several options for configuring the fast-pick area: 5 sections of shelving, 10, 15, and 20. Assume that over this range, the economics of pick and restocks do not change significantly. However, because of the shapes of the storage units, you expect to actually fill only about 60% of the available volume (the rest being lost to imperfect fit of boxes in the shelves).

• For each of the options (number of sections of shelving) estimate the total pick-savings, the total restock costs, and the total net benefit of storing all these sku's in the fast-pick area.
• For each of the options: If you can choose any subset of sku's to go into the fast-pick area, which should they be and in what amounts? What is the total pick-savings, the total restock costs, and the total net benefit?
• Revisit your solutions to the two previous questions: How many sku's were stored in amounts less than a single storage unit? How many were stored in amounts exceeding their maximum on-hand inventory? How does the quality of your solution change if you were to round each volume stored to one that is a multiple of the volume of a storage unit (so that the solution will correspond to some integer number of cases)?
• For each of the options, estimate how much an additional section of rack would be worth to you.

While you can solve a small problem likes this by hand, it makes more sense to use a spreadsheet or other programming environment; or you may want to use the Warehouse Science carton-slotting program.

Case #2: A tire warehouse

This tire distributor stores all product on pallets. Each sku may be stored either in bulk storage, which is 5-high floor-stack in a separate warehouse, or in the fast-pick area, which is 1-deep pallet rack. All pallets are of identical dimensions; but different sku's may have different numbers of pieces per pallet.

Figure 1: Bulk storage, from which the fast-pick area is restocked.

Figure 2: The fast-pick area, in which picking is concentrated.

Data files

Here is a list of the sku's in the warehouse and something about their activity: It is an ASCII text file.

Each row corresponds to one sku and the columns correspond to the following fields:

• Sku ID
• Average # piece picks
• Average # quantity (in pallets) sold as pieces
• Average # full-pallet picks
• Average # quantity (in pallets) sold as full-pallets
• The minimum # pallets to be stored in the fast-pick area (if any at all)
• The maximum number of pallets expected to be on-hand at any time

Restocks of the fast-pick area average 3.5 person-minutes per pallet. Picks from the fast-pick area average 0.45 minutes per pick-line for less-than-pallet quantities and about the same per pallet for full-pallet picks; picks from bulk average 2.34 minutes per pick-line for less-than-pallet quantities and about the same per pallet for full-pallet quantities.

If a sku is stored partly in the fast-pick area and partly in bulk storage, then all of its piece picks must come from the fast-pick area and all of its full-pallet picks must come from the bulk storage area.

Questions

• Which sku's should be in the fast-pick area and in what amounts? Answer this question for a fast-pick area of 2,000 pallet positions. How many person-hours of labor are expected to be required to move the product?
• How does your answer above compare to the policy of storing the minimum allowable amount of your most popular sku's in the fast-pick area?
• How does your answer compare to the alternative of storing everything in bulk storage so the fast-pick area can be absorbed by manufacturing? How much would this cost (in labor hours)?

While you can solve a small problem likes this by hand, it makes more sense to use a spreadsheet or other programming environment; or you may want to use the Warehouse Science pallet-slotting program.