When faced with buying more square-footage for your warehouse or distribution center, the best place to start shopping is within those four walls. Increasing storage capacity without increasing the building footprint usually requires less capital spending and results in a smaller building that is easier to manage and maintain, requires less travel time for workers, and is blessed with lower heating and cooling costs over the long term.
The economics become more obvious when the cost-per-square-foot for an addition is high. Consider a minus ten-degree frozen food warehouse costing about $80 per square foot to construct, as is typical in many parts of the country. That’s the initial cost. A major food distributor has calculated that the refrigeration costs for the facility will equal the construction costs over a 14-year time-span. These refrigeration costs are almost directly proportional to the roof’s square-footage exposed to sunlight. When the footprint of a freezer can be kept unchanged as more internal capacity is developed, refrigeration costs won’t increase.
In spite of these and other reasons to improve existing capacity prior to adding to the warehouse, and in spite of the generally agreed upon principle that adding to a warehouse should be the last and least desirable alternative, it is a principle that is often overlooked because of the assumption that “all that can be done has been done to improve storage capacity.”
Start with Basic Calculations
To begin, determine the current density of storage per square foot of storage space. For most operations with mixed goods, I prefer to use cu-ft per sq-ft. Whether cartons, bags, bicycles, or fur coats, the total cu-ft of inventory is divided by the space (sq-ft) that is used just for storage. Obviously if all cartons are dimensionally the same, we could use cartons/sq-ft. But that seldom applies.
For most firms the density is easily calculated. The cu-ft of inventory is simply the extension of total item inventory multiplied by product dimensions. Instead of multiplying dollars per carton times the number of cartons per item to arrive at total inventory cost, we use cu-ft per carton.
As an example, let’s suppose that product inventory is 400,000 cu-ft and total storage area (including aisles) is 114,000 sq-ft. Our resulting storage density is 3.50 cu-ft per sq-ft. Next, measure the clear height that is available Let’s assume we have a typical space that provides 23’-4” clear under the sprinklers. We have 3.5 cu-ft per sq-ft density in 23’-4” vertical height. Another way to look at this is to envision that if all product (without pallets and aisles) were stacked in the 114,000 sq-ft, the merchandise would rise off the floor 3.5 feet.
Using 3.5 feet of the 23’-4” vertical height leaves us with 15% utilization of the liquid cube available. Is 15% good or bad? It’s bad! Is 15% typical? Yes. Have most warehouse operators performed this test? No.
Benchmark your Calculations
Ratios of output per unit of input are useful when setting performance goals, guiding decisions, making comparisons with accepted standards, trying new ideas, and learning from experts in the field. It applies very well to the quest for improving storage capacity.
For a warehouse that is similar to a public storage warehouse with full pallets in and full pallets out, one very successful national firm requires designs that yield 33% utilization.
For a warehouse similar to a grocery distributor with mainly case picking, a major firm requires designs that yield 23% utilization. Most of this firm’s competitors have the previously mentioned 15% utilization. Going from 15% to 23% is not an 8% increase. It is a 53% increase! Increasing utilization to 23% in our 114,000 sq-ft example would raise inventory capacity from 400,000 cu-ft to 613,000 cu-ft.
Note: If the owners of our example warehouse were to assume "nothing can be done to improve the internal capacity of their warehouse," and construction of an addition were undertaken, they would need to add 61,000 sq-ft to get the additional 213,000 cu-ft of storage (at 3.5 cu-ft per sq-ft). The construction cost of this much space far exceeds the cost of internal improvements that are available. And the internal improvement costs can be depreciated over ten years, unlike construction costs.
Neither of these two national firms will share the calculations behind these data because it’s too valuable. It is difficult to find books on the subject. Warehouse management systems software (WMS) and supply chain strategies often take center stage.
Know Thy Inventory
Match storage design characteristics with your inventory characteristics.
If there is high volume inventory of relatively few different items (SKUs), high-density storage designs such as deep racking and floor storage rows can be used per item. If there are a lot of items and very little inventory per item, low-density and high selectivity designs such as bin shelving and selective pallet racking are usually needed.
But when we have a mix of both, it is important to analyze the inventory and break it down using a spreadsheet to sort the inventory into storage type categories. Too much high-density storage will be inefficient and result in unused capacity. Too little high-density storage will cause operating problems and again, a loss of capacity.
Fig. 1 shows a 2-deep x 3-high storage rack over a shipping dock. These structures feature wire decks and netting on the back-side. They are convenient and can be built to be safe. This method offers storage in a most convenient place that is often over-looked. In this photo, the lift truck used to access the rack is a deep-reach stand-up machine.
Fig. 2 is a drive-in rack that has been retrofitted with push back sleds in a fluid milk warehouse. The push-backs increase the utilization of the racking since all storage levels can be independently filled and unloaded. The original drive-in rack was able to achieve a 60% utilization (occupancy) level. Retrofitting it with push-backs raised the utilization rate to 83%.
Fig. 3 is storage built overhead in one of the center aisles in this warehouse. It extends behind the camera for a long distance. Note that the overhead storage is not just between the adjacent rack rows, but over alternate cross-aisles as well. A deep-reach lift truck is used to access the two-deep pallets that would be inaccessible to a single-reach lift truck. Push-back sleds could also have been used in lieu of deep-reach lift trucks.
Fig. 4 is a seldom seen rack system called the ‘checkerboard’ layout. Looking down-aisle we have selective racks on each side. Above the aisle are shelves occurring at every other bay of rack. The storage that is blocked by the overhead shelves is accessed from the adjacent aisle with a deep-reach lift truck or with push-back sleds. The photo was taken before the rack installation was completed.
Fig. 5 is a model of a similar checkerboard layout to help you visualize how the rack system works. Note that the shelves in the model are located like a checkerboard between adjacent rows. This enables a lift truck to use the opening between shelves to load and obtain the second deep pallet that is blocked by a shelf in the next row. A photo with product makes the layout very difficult to understand.
This is an excellent application in food distribution freezers. Lighting is installed under the shelves, and less lighting is required on the ceiling. The result is more effective illumination, lower wattage costs, and less refrigeration cost to counter heat from light fixtures.
The checkerboard layout yields the benefits of both high SKU count for order picking and high storage capacity for more inventory. An added benefit is that these shelves can be designed to accommodate odd shaped loads. For example, 12’ long carpet rolls can be placed on deep shelving using a normal 10’ to12’ wide aisle. Note, the lift truck does not have to make a 90 degree turn with long loads to store on the shelves, avoiding a much wider aisle width.
Several manufacturers build lift trucks that operate in seven foot aisles, and these are excellent when aisle passing is not necessary, such as in low velocity warehouses. The masts on these machines are able to be rotated ninety degrees from the body of the truck and are well suited to racking layouts. They are not able to go in and out of deep floor storage or drive-in racking from these narrower aisles, however.
Floor storage is probably the least studied yet most used system found in warehouses. This may be due to equipment suppliers not being involved in floor storage layouts. I have observed that most floor storage layouts are "home-grown." There is great opportunity for improvement in these warehouses.
Any time we can stack pallets to three or more levels high, it is difficult to justify racking. These pallets are typically for reserve storage of high volume items, or for bulk warehouses that ship and receive without pallet break-down. Whatever the case, bulk storage of full pallets is able to use deep rows with fewer aisles, resulting in high storage capacity. Instead of less than six cu-ft per sq-ft storage density, we will see densities as high as 10 cu-ft per sq-ft as long as deep rows are maintained at high utilization.
Again, we need to match the depth of rows with the inventory characteristics of each item. The utilization of a warehouse with all row depths the same will most likely be low. Varying the depths from one aisle to another will enable workers to match an item’s incoming quantity of stacks with proper row depth.
The problems with floor storage are typically two-fold:
• Pallets less than full (due to damage or end of production runs) have nowhere to go except in the full pallet rows. They waste space because they are either on the floor with nothing stacked on top, or on top of full stacks necessitating frequent "shuffling."
Fig. 6 shows a rack that is used for partial pallets no longer co-mingled with full pallets in deep rows. The capacity of this finished goods warehouse was increased in total, once this racking was installed. Lift truck handling was greatly reduced. Now on to the other challenge with floor storage:
• Deep rows are left partially utilized as progressively depleted. The remaining two or three stacks of pallets in an eight-plus-deep row need a place to go when incoming pallets are in need of empty deep rows.
Fig. 7 The right side of the aisle has 10-deep pallet-position rows of full pallets stacked three high. All 10 stacks are of same item and date code per row. The left side of the photo shows two-deep rows. Remaining stacks from the right side are moved to these short rows across the aisle when empty deep rows are needed for incoming pallets. Warehouse workers call these short rows “clean-outs.” They are invaluable. They keep the deep rows at high utilization and improve product rotation and warehouse capacity. (Stacks seen in the photo behind the two-deep rows are in deep rows accessible from the adjacent aisle to the left.)
The most important lesson from all this is, warehouse capacity must be measured and tracked. Don’t assume capacity cannot be improved. It usually can.
Alan Greene is a former director of industrial engineering at Super Valu Stores. His responsibilities included engineering of facilities, labor standards and warehouse systems. Visit www.alangreenecompany.com.