Slotting’s elevated place in an omni-channel world

With the need to fill orders for multiple channels now a reality for many warehouses, attention is shifting to specific tactics that help with multi-channel fulfillment execution. Slotting functionality—software that takes demand data and crunches that against multiple criteria like item dimensions, weight and order velocity to figure out the optimal locations for each stock keeping unit (SKU)—is gaining in importance as part of the multi-channel trend.

More facilities are trying to handle multiple channels in an integrated omni-channel fashion, using a common facility, a common labor force, and to the extent possible, common bin locations, says Ian Hobkirk, founder and managing director of consulting firm Commonwealth Supply Chain Advisors. That is a big change from several years back when the initial impulse for many companies was to outsource budding e-commerce fulfillment to a third-party logistics firm.

“Now more companies are deciding to bring that e-commerce fulfillment in-house and do it all under one roof,” says Hobkirk. “But the challenge when you try to do that is that the different channels have very different order profiles, so that adds complexity.”

One trend among omni-channel operations, says Hobkirk, is to try to have common bin locations, even though the pick, pack and ship processes and automation used from there might vary. As part of this challenge, slotting is key to profiling which SKUs should go into which bins and in what quantities, says Hobkirk. “But the slotting challenge becomes a lot more complex because now I have to look at demand across multiple channels out of one bin and decide things like, ‘how much do I keep on hand to satisfy demand for e-commerce, retail and wholesale, and what is the optimal storage medium?’”

More frequent slotting is part of the answer for omni-channel, say Hobkirk and others, but it’s not the full answer to keeping forward pick areas efficient. Better collaboration around forecasts and promotions is needed, and dynamic replenishment powered by warehouse management system (WMS) software is seen as another way of efficiently moving goods into forward pick areas. For certain, the omni-channel trend is raising the importance of slotting functionality, and the way it is used.

“Omni-channel is completely changing slotting as we know it,” says Peter Schnorbach, senior director of product management for Manhattan Associates. “Particularly for our customer base, which is heavily retail industry, companies are introducing new products almost every single day. So with the increasing number of SKUs, and the intervals at which they are being added, we are seeing a need to be slotting much more frequently.”

What can users of slotting and WMS expect as they face these omni-channel pressures? For one, expect slotting to be used more frequently. Second, slotting may be used on different levels, from quickly finding the optimal spot for new items to reslotting larger zones for maximum picking efficiency. Third, observers say that slotting may need to be combined with dynamic replenishment methods to quickly get forward pick areas stocked up with right goods, especially in the face of unexpected demand.

Frequent slotting
Slotting software is used by companies to establish the best locations for goods in a warehouse by looking at the forecast for SKUs, especially order velocity or how fast a mover an item is, data about the product such as dimensions and weight, and data about locations in the warehouse. It then suggests a plan on where to locate SKUs so that orders can be picked and filled in the most efficient manner.

While that process may sound simple, it can involve complex calculations and multiple criteria. For example, reslotting one SKU often triggers a series or “chain” of moves. “Slotting software is really good at looking at those chains of moves and determining how one move impacts all other locations, then exporting the plan to WMS,” says Hobkirk.

Slotting logic focuses heavily on units of measure, such as items, cases or full pallets, so solutions should be able to generate slotting plans for an item pick area or line as well as a case pick line, adds Tom Kozenski, vice president of industry strategy at JDA. Additionally, a slotting solution’s logic may crunch other factors, such as the safety and ergonomics of the location of heavy items, or keeping items slotted in such a way that they are easy to pick by the “department,” which makes it easier to unpack goods and stock shelves at the store level.

The stickler in using slotting effectively is that it needs reliable forecast data to gauge how fast a mover a SKU will be, or with promotions, what the demand curve will likely be before it “tails” back to normal. “It can be challenging to optimize the facility when you’re not sure what the order profiles are going to look like during the day today,” says Kozenski.

Companies can slot based on history for seasonal goods, but with e-commerce heavily focused on digital promotion campaigns, it’s crucial that merchandising organizations communicate with supply chain and warehouse operations people about expected sales. “Warehouse operations managers are smart enough to know that the products being pushed from a promotion or a discount are going to be popular items for a certain period of time,” says Kozenski. “They may want to react to that information to reprofile their facility.”

E-commerce sites not only use frequent promotions to keep customers coming back, but most sites have a “new arrivals” section which creates the need to slot those goods back at the warehouse level. As a result, Schnorbach says slotting has shifted from an exercise done monthly, quarterly or seasonally to a much more frequent exercise. “We’ve seen slotting move from periodically trying to get the entire warehouse into full optimization to ‘I need to get new items into the active pick area as quickly as possible so that they are available for sale,’” he says.

While slotting software is offered by some best-of-breed vendors as well as major WMS vendors, in either case, slotting should be well integrated to the WMS, not only to gain SKU data, location data and order history, but so that slotting plans can be sent back to the WMS for execution.

Some end users have in the past made do with spreadsheets to help slot goods, but today a slotting solution with full integration to WMS is seen as necessary to support more frequent reprofiling. With integration, slotting tasks can be interleaved with picking and replenishment tasks, says Kozenski. “When slotting is integrated with WMS, slotting recommendations go into the queue and become directed work,” he says.

One major caveat to more frequent slotting, or for that matter any slotting plan, is that the labor to execute the plan can’t exceed the labor savings gained from more efficient picking after the reslot. Vendors say that because of this concern, slotting plans should be analyzed against the labor involved, which might require running a plan through a labor management system analysis. “You have to be careful about the necessary labor to support reslotting,” says Jon Kuerschner, vice president of supply chain solutions with HighJump Software. “The ultimate goal with slotting is to make picking more efficient, but you can’t do that at an overly high labor cost to make the slotting happen.”

Different slotting levels
One major wholesale distributor that Schnorbach is familiar with is using slotting on distinct levels. The company uses slotting software on nearly a daily basis to slot new products, but also reslot the entire warehouse every year or two. Additionally, they frequently target zones that have fallen below peak picking efficiency for a reslot.

“They have what they call projects for a zone that has fallen out of an optimized state,” says Schnorbach. “They will reslot that zone, and then roll through the warehouse in that way one zone at a time, so they almost always have a project going in some zone of the warehouse. So we are seeing some users layer different slotting strategies together.”

Slotting can also be used in a “what-if” vein when user organizations are contemplating ways to establish new pick areas or rerack a section of the warehouse. Slotting software generally can’t be used as a facility design tool, but some solutions can generate parameters useful to rerack projects. “Our slotting does have some ability to make calculations on how much racking you would need, but it won’t tell you where to put it or how to configure it,” says Schnorbach.

Slotting can be thought of a mid-level, operations planning solution, but it benefits from long-range planning and forecasts at the supply chain level. Kuerschner points out that companies that do a good job of supply chain network design—establishing the best “nodes” in the supply chain—and then do a good job of inventory planning across those nodes, have a much better foundation for slotting. “You have to take a step back and look at supply chain optimization first,” he says.

Hobkirk agrees that communication around demand plans and solid, enterprise-wide inventory planning sets the stage for better slotting. “If a company has invested heavily in good inventory management tools, a lot of these things with slotting are easier because you have more certainty,” he says.

Dynamic replenishment
In the daily grind of warehouse execution there are plenty of surprises that are hard to plan for and reslot around. What if a Web promotion doesn’t get communicated, or is much more popular with consumers than anticipated? Or, what if a wholesaler suddenly decides it wants to place a big order for a certain SKU? When demand is unpredictable, says Hobkirk, the best way to execute may be through better replenishment. “Being efficient may come down to better tactical replenishment,” he says.

One way to improve replenishment, says Hobkirk, is to keep a certain percentage of open bin locations in a forward pick area that can be assigned to any SKU. “Then when there is unusually high demand, I can use those open bins to bring up additional supplies,” he says.

With unexpected big orders for a certain SKU, another replenishment tactic is to bring up cases from an overstock area and bring them to a put system or unit sorter so that picking can be done efficiently without wiping out the inventory for that SKU in the forward pick area, says Hobkirk. However, some legacy WMS systems aren’t able to handle this type of dynamic replenishment because they want to put a SKU in just one location, or they can’t manage the bin-level order allocation rules to allocate picks to the optimal location.

Kozenski agrees that in some situations there is not enough known about the expected demand to reslot, so it comes down to the more dynamic replenishment and wave management through the WMS. “So if I have a have a big influx of orders for a certain item, maybe the WMS will create a movement where I’ll bring a full pallet of that stock over to the pick line and just let a packer just start packing those items away rather than digging them out of the bins,” he says. “Sometimes, it’s about being able to be more reactive, which is what you get today with a real-time WMS.”

Companies mentioned in this article
• Commonwealth Supply Chain Advisors
• Manhattan Associates
• JDA
• HighJump Software