Variety may be the spice of life, but too much of either variety or spice leads to imbalance. The inherently variable nature of warehouse and distribution operations is both a curse and a blessing. On the one hand, it is unpredictable, making planning a daily challenge. On the other, it is, as a friend of mine likes to say, “fraught with opportunities.” It is dynamic and, therefore, especially amenable to the concept of continuous improvement.
For many of the same reasons, DC operations sustain the image of being good places to cut costs and to improve performance. Things — customers, products, markets, vendors, and systems — are constantly changing. Because they are interdependent, the mix is never dull. Improvements are always possible and often significant. Looking at strategies for optimizing workflow in the distribution center should be an ongoing process, and it continues to pay dividends over time.
General principles
The first thing most distribution executives will tell you when you begin to examine how and why things are as they are in their operation is “We’re unique.” Despite that widely held belief, some fundamental principles apply widely, if not universally. Five of my favorites are:
- Reduce product handling (labor, damage).
- Eliminate travel wherever possible (labor, time).
- Minimize the number of steps in any process (simplification, ease of completion).
- Eliminate handwritten data of all kinds (errors, delay, communication, dissemination).
- Don’t let people do anything a computer can do better (common sense, competitive advantages).
Although none of these apply to the same degree to every operation, each of them points to opportunities to improve product flow and work activities in every facility I’ve seen in the past 15 years. Their feasibility is a function of customer requirements, resources (sometimes including capital), and management’s will to make the necessary changes.
Below is a list of somewhat more specific items gleaned from recent audits that my firm conducted at three new client companies. While this list is intended to be neither comprehensive nor prioritized, it contains representative material that either bears directly on workflow or will take advantage of workflow conditions to further improve customer service and the cost of operations. It’s a simple way to convey some of the breadth of this topic. Each principle is followed by a comment that explains its application.
Operating procedures
- Stagger work schedules based on workflow; reduce labor and minimize congestion by varying when people start, when they take lunches and breaks, and how long they work.
- Perform all work in a single shift, if possible. This can help accelerate product flow to the areas of need and enhances communication across functions.
- Assign work to individuals, not teams. Individuals will get more done than teams in most warehouse work activities. If you need teams, make them small ones.
- Pre-print labels and documents; avoid any machinery dictating workflow or pace if possible.
- Provide formal training and cross-training for all employees. This will help ensure that best practices are understood and practiced.
Receiving
- Schedule all inbound receiving, based on the nature and size of the load. Dock space utilization will improve using this technique and the intake process will be more efficient.
- Assign loads to doors adjacent to product destinations. This tactic will reduce travel while increasing the availability of material handling equipment.
- Have vendors label and unitize inbound product. This method can significantly reduce handling and improve holding capacity utilization.
- Certify vendors and work inbound shipments according to quality. Certifying vendors can greatly streamline receiving and putaway activities, and inbound shipments that conform to your guidelines will affect almost every downstream warehouse function as well.
Putaway
- Put away everything the same day. There is no substitute for timely processing of inbound deliveries because it affects space, congestion, purchasing, service levels, and, oh yes, sales and profits.
- Route putaway based on destination locations — travel can be greatly reduced when the path is sequenced based on the shortest route for the SKUs in each load.
Productivity management
- Maintain and use standards in all major functional areas to assist management in identifying systemic bottlenecks such as aisle congestion or other workflow problems.
- Periodically review methods to be sure they are relevant, effective, and documented. As products and packaging change, as vendors and customers change, so must your methods change to maintain optimal workflow and performance.
- Measure individuals, not teams. To determine whether a productivity problem is systemic or individual, and therefore to understand how to tackle it, a manager must know how individuals are performing.
- Reward individuals and teams. By combining individual and group or facility elements in the reward calculation, individuals are protected from some variables that they cannot control (such as systemic problems or seasonal work volumes), and the edge is taken off the negative aspects of competition.
Space utilization
- Reslot early and often; place product in the optimal location to minimize handling, travel, and availability.
- Use overhead space (the greatest opportunity in many facilities). This practice can enable the location of “dog” inventory where it does not impede the movement of more productive items.
- Eliminate aisles to add space. By varying storing techniques and fixtures based on various product characteristics (physical traits, velocity, etc.), some aisles can be reduced or eliminated, creating space.
Selection
- “Deliver” work to pickers. Eliminate a central pick-up point for documents; “pickers should pick” and minimize the travel to do so.
- Include pallet-building skills in picker training and slotting strategies. The process of selecting, packing, and shipping goods to fill orders will be streamlined and handling will be reduced.
Quality
- Move from 100% checking to statistical checking. It can pretty readily be demonstrated that error rates do not decline as the extent of checking goes up. Statistical checking reduces labor and delays, and helps identify systemic problems in the workflow that cause errors.
- Fully utilize systems already in place. It is surprisingly common to find operations in which software, best practices, or standard procedures have been neglected, abandoned, or left unchanged and underutilized over time, despite their potential value.
So how valuable is this stuff, really?
Let’s suppose that you have the opportunity to move your operation from an essentially manual environment to some level of computer-based activity, using a warehouse management system to provide additional visibility for planning, as well as control and management functions. Although this is not the only way to influence the flow of goods and work through the operation, it is an excellent way to quickly acquire resources for many of the techniques noted above and more. Your WMS need not be a state-of-the-art, two-million-dollar acquisition, but something with the core functionality that is a prerequisite for any software application to compete in this fairly mature market.
In our model, the advent of WMS capability enables management to group or batch pick small orders with a few lines into groups that share some SKUs in common and are picked in the same physical area. This is an important opportunity, as pickers usually constitute the largest work group in the warehouse and in some operations can find that travel accounts for as much as half of their work effort.
On page 40, you will find an illustration of the travel pattern a worker would follow to pick each of five orders in successive trips through this aisle — one trip per order. (See Table 1 in the diagram for the order profile; the second order has all five SKUs and shares at least one of those SKUs with every other order in the batch.) Based on the scale of this layout using 8′ pallet rack bays, travel from start to finish is 64′ per order, or a total of 320 feet (see Table 2 on page 40).
Most fully featured WMS applications today will enable the “batching” of orders of this kind, creating a consolidated pick document (or RF record), directing the picker to make a single trip through the area and pick 15 pieces, the total for all five orders.
Referring again to Table 2, we see that the travel is reduced by 80% through batching. While this is offset by the need to sort the SKUs by order at the end of the pick process, that step is more than balanced by eliminating four out of five trips to pick each order separately.
Clearly, anywhere that travel can be reduced in the warehouse, it should be. For many DCs, another large opportunity is found in the process of moving received product into the facility and storing or shipping it. The travel paths shown in the diagrams on pages 42 and 46 are part of a best-of-all-possible-worlds example of improving putaway travel.
Product A is received in sufficient quantity to meet immediate shipping needs, provide inventory for the pick line, and be stored in reserve locations. In addition, the product needs to be checked by quality control personnel. For our purposes, this product comes from a certified vendor, meaning that although it needs to be checked for compliance, that process need not hold up using or shipping the receipt immediately.
In a manual environment, that product might be handled in four trips by four different people — one, dock to QC; two, QC to reserve storage; three, storage to the active pick location; and four, the active pick location to the outbound dock. Without addressing how they arrived at the point where they picked up the product, the drivers’ aggregate travel would be 324 feet, as shown in the diagram on page 42. This example does not take into account the time between legs of this trip, which will only extend the process.
In an operating environment controlled by a warehouse management system, it is well within the realm of the possible that a single forklift driver could accommodate all four stages of this process in a single trip. With forewarning about delivering a sample of the merchandise to quality control for evaluation, the urgent need for product to ship for one or more “hot” orders staged at the outbound dock, and the opportunity or requirement for more of that product to be delivered to the active pick location, the driver could accomplish all four steps and travel only 196 feet to do it — a reduction of at least 40% in travel, as shown in the diagram below.
Just listen to Doug and Matt
A number of the opportunities described above parallel the experience that Doug Baker, director of distribution, and Matt Fette, manager of warehousing and distribution systems for Cincinnati-based Totes-Isotoner Corp., had when they moved into a WMS-capable operation and a new distribution operation at the same time.
They were both on hand prior to the changes and can vividly recall the differences. And the numbers are impressive. Totes-Isotoner operates a 450,000-sq.-ft. facility staffed with 140 full-time equivalent employees (FTEs) at the low end of a seasonal cycle and up to 365 FTEs at peak. Its January volume may be as low as 30,000 units shipped per day, peaking at 200,000 units prior to Christmas. “After installing Manhattan Associates’ PkMS WMS application, we saw our cost per unit shipped drop by about 70%,” says Baker. “In addition, the cost of our processing of return units was reduced by about 67%.”
Baker notes that the operation was also able to significantly improve inbound flow by “prioritizing the flow of goods from the dock to ‘hot product locations,’ by delivering components directly to the value-added processing area, and by differentiating putaway zones based on specific SKUs.”
Fette provides another example of the dividends realized. “With the addition of a full-case sorting capability, we were able to take advantage of the WMS’s ability to organize a batch pick for an entire wave of work. This enabled us to use a single picker and gain as much as 30% to 50% productivity improvement. This constituted a major change in philosophy, but it was well worth it.”
To put the crown on this example, let’s try to estimate the value of these benefits. If hourly pay for this operation averages about $12 with benefits, an employee’s annual compensation would be roughly $25,000 without any overtime. Even a 50% improvement in productivity for one-third of 140 people (pickers) constitutes $1,155,000 in annual savings. For 365 people over one quarter (the Christmas rush), that reduction in cost means $752,800 less cost every year.
Now what?
In concluding this exploration, I would like to circle back to the beginning and the focus on continuous, small improvements. A WMS implementation and a new facility are not everyday occurrences. However, one of the things some of us learned from Deming and the Japanese in the 1970s was that many small improvements will always win out over a few large ones. In aggregate, they lead to larger savings, however subtle.
Many of the concepts touched on in this article require little or no capital, can be repeated over time with benefit, and are fairly easy to understand and implement. I commend them to you.
Ron Hounsell is vice president of software solutions at Tom Zosel Associates, a distribution and logistics consulting firm based in Long Grove, IL. He can be reached at (847) 540-6543 and at [email protected].