Consumers have been spoiled by competing supply chain managers. A combination of technology and constantly improving practices has taught the public to want things fast, good, and cheap! At the same time, a tough economy has taught supply chain managers to seek a competitive advantage by applying advanced material handling and logistics systems that reduce labor, speed delivery, and increase accuracy.
Since order-picking is typically the area representing one of the largest labor components, material handling systems manufacturers have focused their offerings on integrated systems which reduce human travel and the number of touches associated with picking. In a book my firm was commissioned to author for the Warehousing Education & Research Council (www.WERC.org) entitled, “Pick This! - A Compendium of Piece-Picking Process Alternatives,” we identify and explore 567 piece-picking process alternatives. The sheer volume of differing approaches warrants detailed data analysis and a thorough understanding of goals and objectives before embarking on any picking automation initiative. After all, if you apply automation to a bad process, bad things happen faster.
Pick system automation is tried and true and solutions have been deployed effectively in Europe for decades. European warehousing, distribution, and fulfillment operators embraced these technologies earlier than practitioners in the United States, largely because of their higher labor and land costs and because their economic policies encourage long-term capital investment.
Further fueling the case for automation is the global trend towards higher SKU counts, smaller orders with higher frequency, compressed order cycle time requirements, and a plethora of value-added services which add cost to every order processed.
Not just for the high-margined
Contrary to conventional wisdom, the application of integrated mechanized and automated systems in Europe is not evidenced only in high-tech, high-margin business sectors such as pharmaceuticals, medical, life sciences, semiconductors, precision optics, or advanced ceramics companies.
During a recent trip to Spain, I visited a two-million square foot supermarket distribution facility, which housed one of the most highly automated operations in the world. The installation consisted of more than two-hundred Automatic Storage & Retrieval System (AS/RS) cranes operating in dry, refrigerated, and frozen environments.
An AS/RS is essentially a high-rise rack or shelving system with long, very narrow aisles in which a crane travels horizontally down the aisle on a floor mounted and overhead rail system. The crane is equipped with a carriage mechanism that travels vertically on the crane mast and inserts or extracts unit loads from either side of the aisle automatically. A Warehouse Control System (WCS) keeps track of the location of each SKU and instructs the machine where and when to operate.
In this grocery operation, homogeneous pallet loads (single SKU) are received at the loading dock, their bar-code license plates are scanned, and then the pallet loads are placed on pallet conveyor stations which induct them into the AS/RS where they remain in the proper temperature controlled environment until needed for picking.
When that particular SKU is needed to fulfill order demand, the pallet is automatically extracted by the AS/RS crane and delivered to a de-layering station where an overhead suction device lifts one layer of cases off the pallet. The layer is then deposited onto a wide array of interleaved conveyor belt sections which “singulate” them, forming a train of cases on a narrow belt.
As each case approached the end of the conveyor, it cascades gently onto a low-profile “slave tray” (a uniform plastic carrier) which is fed by an automatic tray de-stacker. Now that each case is traveling in slave trays, it can be inducted into a “mini-load” AS/RS machine which randomly stores each case of product (on its slave tray) in its correct temperature zone.
Once all of the SKUs needed to complete a particular wave of orders reside in the mini-load system, the discrete order picking cycle begins. SKUs (cases) for a particular order pallet are extracted automatically by the system and delivered to an automatic palletizing system which computed the best possible layer and interlocking patterns for the wide variety of case dimensions and weights. The empty slave trays are delivered to an automatic stacker, queuing them for use once again. Once palletized, the unit load is automatically stretch wrapped, labeled, and delivered to the shipping dock in the lane corresponding to the outbound tractor trailer (in reverse stop sequence).
No human touches a case from the time a single-SKU pallet is received at the loading dock until the time the multi-SKU stretch-wrapped pallet is loaded onto an outbound trailer.
As with all automation, there are of course exceptions. Odd or unwieldy SKUs such as toilet paper, bags of rice, etc. are handled manually and circumvent the automated system. While this accounts for all of the case and pallet volumes moving through the DC, there are also some items that even high volume supermarkets do not receive in case quantities and therefore a “piece-picking” component is needed.
Also housed in this facility are a series of “goods-to-man” picking modules (sometimes referred to as dynamic picking). Here, the picker stands at a station where the SKUs to be picked are delivered to him/her automatically. Target totes (shipping containers) are also delivered to the pick station. A pick-to-light/put-to-light system identifies the SKUs and quantities needed for a particular order. Picking personnel simply follow the lighted instructions as the product comes to them. Human pick rates of 400 - 600 lines per hour are possible with these types of semi-automated systems since unproductive travel and identification/decision tasks are removed from the equation.
While the food industry is clearly a low-margin business, the transactional volumes are incredibly high and harsh refrigerated/frozen work environments make it an excellent candidate for automation. This particular company has gained a significant advantage over its competition as a direct result of this monumental automation initiative and they are planning to deploy more such automated operations as part of their growth plan.
When automation's right
Conversely, nearly a decade ago, I led an automation project for a leading United States jewelry company whose impetus for mechanizing was driven by a commitment to quality and service as well as a desire to literally and figuratively keep fingerprints (human touches) off their high value gold, platinum, diamonds, emeralds, and rubies. The resulting quarter-million square foot facility is still one of the most highly automated operations in North America, however, such automation initiatives are becoming more prevalent as pressures to reduce costs, ensure quality, and compress time mount.
Our more recent automation projects (in the US, Canada, Mexico, Europe, and the Far East) in the appliance, apparel, automotive, biotech, catalog (e-tail), communications, computers & peripherals, consumer products, electronics, entertainment (books, CDs, DVDs), food, fragrances, pharmaceutical, publishing, retail, and technology sectors demonstrate the increasingly broad geographic and industry applications of automated material handling technology.
It is important to note, however, that not all operations are good candidates for automation. Some are still best addressed by well-engineered manual processes. Equipping a warehouse operation with picking automation may be justified when a combination of some of the following characteristics are present:
-
High-value inventory (such as jewelry, precision electronic components, etc.)
-
Large SKU populations (particularly with low inventory levels per SKU)
-
Small cube, uniform shape (such as DVDs, cosmetics, medicines, etc.)
-
High order line volume with low units per line
-
Harsh work environment (refrigerated, frozen, etc.)
-
Batch or lot control requirements
-
Kitting, assembly or other Work-In-Process (WIP) buffering/control requirements
-
Value Added Services (VAS) buffering/control requirements
-
The best ways to evaluate the costs/benefits of automation include a thorough historical data analysis (inventory and movement by SKU), supported by forecasting through the design year (typically a five-year horizon). The result of this data crunching forms the foundation for a plan, allowing a comparative analysis among process, systems, infrastructure, and labor alternatives.
It has been my experience that most United States businesses are seeking return-on-investment formulas that yield a three-year simple payback. But it is also important to quantify the financial impact of soft benefits such as improved service levels (shorter order cycle times and higher fill rates as examples). Additionally, costs avoided by implementing automation should be calculated and factored into the equation (such as lost sales, charge-backs, incomplete orders, cancelled orders, error correction, and returns).
Each design option must be ranked based on capacity, capital cost, return-on-investment, internal rate of return (IRR), productivity, space, labor, complexity, risk, flexibility and scalability. Only after this analysis will your plan be complete.
Lawrence Dean Shemesh is president and CEO of OPSdesign Consulting, an independent supply chain consulting organization specializing in warehousing, distribution, and fulfillment operations design. Shemesh has managed hundreds of operations design projects spanning a variety of industries and market channels. He can be reached at 856-797-1933, x101, or [email protected].
