Is Warehousing a Necessary Evil
Customer requirements are driving warehouse operations of all sizes to improve inventory accuracy, quality and customer service. Manufacturers have mixed feelings about their distribution centers. They dislike these messy appendages to otherwise clean productive operations. They cater to the wacky requirements of crazy sales people, unreasonable customers and money-grabbing financiers anxious to hit quarterly promises to Bay/Wall Street. Non-core to operations, uncertain internal capabilities, changing specifications, add up to uncertainty, and uncertainty precludes investment. Perhaps this is why many manufacturers under-invest financially, technologically, and even intellectually in their distribution centers. And as performance suffers, executives distance themselves.
Sounds like a golden opportunity for outsourcing? Yet many manufacturers cannot find anyone to do the job properly. According to an Accenture survey (quoted in Inbound Logistics), they are uncomfortable with:
For many, it is a zero-sum game. Value is created by the 3PL by using lower wage and benefit packages, and taking lower returns on capital.
Outsourcing warehousing may no longer be a no-win situation. I recently visited a third party operated warehouse operation that challenges conventional thinking about manufacturers’ warehouses and outsourcing, the Schenker Stinnes Logistics operated consumer package goods fulfillment center in Brampton, Ontario.
What is so special about what they have done? Their first step was to segment product flows by handling characteristics:
From the results of this analysis, they focused on how to engineer the processes, with automation and systems where appropriate, to provide new levels of productivity and quality and the truly perfect customer order:
The core mindset is not how to make the systems work, but how to make the processes work, engineering to six-sigma quality (less than three defects per million, or 99.9997 percent).
For case picking, the tasks are driven by multiple-order waves of bar-coded labels that are “slapped” onto cases, as they are picked by order of weight for pallet quality, and placed on a conveyor. The sortation scanner reads two labels – the pick label and the factory placed SPC label that incorporates lot and date information. If there is a match, the cases are sent to the pallet building stations. In order to respect the optimal pallet building logic that will limit transportation breakage, the system recirculates cases until it is their turn to be palletized, driven by algorithms written by the in-house operations research team. This process also allows full-lot traceability by customer and even delivery. And it is efficient. Grocery picking rates will typically be between 150-225 cases/hour; this system routinely provides picking/palletizing rates well above 300 cases/hour (picked, palletized and staged).
Picking by layer. This process will use specially designed hardware to enable full layers to be picked and weighed. The weight parameters at the level of a pallet are significant, so weight is an effective verification tool in this usage. Layer pallets are far more tidy and robust, which further reduces breakage. Finally, productivity that is three to five times the rate of conventional picking is expected. Given these advantages, the manufacturer would be well advised to actively sell pallet layers to their retailer customers.
Warehouse damage is further reduced by the use of flexible modular layouts with automated shuttle high-density storage systems for bulk storage. With slots of up to 25 pallets each, the forklift operator places palettes at the entrance. A self-guided radio shuttle then repositions the palette behind the last palette in the accumulation rack, while the forklift operator is free to work elsewhere. The shuttles can be quickly and easily transferred between different storage racks, by a conventional forklift truck. The system is set up as a FIFO feed from one side, pick from the opposite side. This shuttle-rack system costs four times that of conventional racking, or about twice that of drive-in storage systems. The payback comes from a high rate of capacity utilization and a high rate of accessibility of the goods because each level can take different products. A 300,000 square-foot facility stores the factory output and ships 26 million cases a year. The bulk storage system provides additional value in productivity and safety (the system is unattended and quiet), energy efficiency, and reduces breakage (no forklifts go near the product stored).
Assemble in Warehouse (AIW) or postponement allows plants to manufacture efficiently while special products and promotions can be customized to order offline, in the DC, as and when they are needed, without costly inventory commitment, or interfering with the production dynamics: Costco special packs, pallets that go directly onto sales floor.
On-time delivery is engineered in a traditional route-driven plan, aiming to get all orders picked, packed and staged for the deadline. A dash-board shows when the wave must be complete, and the supervisor can redeploy his crew as the shift progresses. At the end of the picking wave, the team helps palletize, ship and then clean-up, before the next wave is released.
Low cost is achieved through the low-cost design of these high-productivity processes, the sustainability of the quality (no rework), the use of systems that assure sustainability and information to focus improvements, a “cost-plus” financial arrangement that guarantees transparency and, finally, the use of targeted key performance indicators that ensure the control and accountability of the operator to deliver on their service-level promise. Systems, such as the SoliNet WMS used here, also manage the fulfillment to each customer’s requirements, such as minimum shelf life remaining on all products shipped, or fulfillment rules such as fill and kill, hold backorders, hold orders for complete delivery. Some sophisticated ERP packages used by manufacturers cannot manage these rules by customer, or pallet logic (e.g. one product/pallet) to ensure each product’s characteristics (e.g. odor emission and absorption) are taken account of. For example, this ensures that tea is not stored next to perfumed soap, slow-moving products are overridden for new products or promotions, fire-risk product (oil, aerosol) density is taken into consideration such that critical concentrations are not reached that would require expensive in-rack sprinklers. These are the glue that synchronize the processes, and along with the innovative algorithms, constitute the secret formula behind compelling success. Continued low cost is also achieved through modular, easy-to-scale-up design. The warehouse racking and picking can accept a new level that can double throughput with minimal additional capital expenditure, and the facility can be expanded to handle twice that volume again (100 million cases/year).
Trust is further built by managing the operation to an agreed upon service promise, measured jointly by key performance indicators. Predictability of performance, quality and cost become a solid foundation to build a durable relationship. The significant investment made creates a very powerful incentive for the operator to commit and perform.
Eric Dewey, the Vice President of Logistics at Schenker, to whom this operation reports, says, “We have an excellent employee proposition too:
This makes it easier to keep good people.”
This value-based warehousing approach addresses key concerns around outsourcing – control, predictability, cost visibility and control. Mutual dependability is achieved through long-term contracts and significant investments. Even if you do not want to outsource, there is plenty here for you. Doing things right is not easy, but it can be very worthwhile when you consider costs, waste, health and safety, noise, energy consumption and happy customers. There is hidden value in warehouse operations.
There is also tangible value: A study by GMA estimated the supply chain cost/case shipped in error to be US$27. At 26 million cases per year, moving from 98 percent to 99.5 percent accuracy is worth over $10 million. Engineering idiot-proof processes with adapted automation can provide significant improvements to operations and processes, if a systemic approach such as the one illustrated here, is taken.
Eric Dewey sums it up this way: “For some time we have seen large retailers investing in warehouse technology and automation to increase the speed and efficiency of their supply chains. For some of them this has provided a competitive advantage.”
What is stopping manufacturers from doing the same?