Robotic Warehousing: A New Market Opportunity for Robot Manufacturers and Integrators?
by Paul Kellett
, AIA/MCA/RIA Director – Market Analysis
Robotic Industries Association Posted 02/07/2011
For the complete version of the 16-page market study, RIA members can proceed directly to downloading the study at no charge. It is also available to non-members for purchase. The Free PDF download and purchase instructions are available at the end of this article.
This, and additional market opportunity studies, are examples of the numerous benefits an RIA membership provides.
For information on becoming an RIA member and, thus obtaining the study at no cost, click: RIA Member Benefits & Application.
According to a new ATC study on robotic warehousing, a revolution is underway in order fulfillment and automation is enabling it. In response to competitive market pressures for shorter cycle times and greater accuracy, warehouses are increasingly using automation. Shorter cycle times are arising because retailers are asking for increasingly smaller and more frequent shipments. Instead of asking for full pallets, they are asking for a few cases. This necessitates more case and piece picking in the warehouse, the most labor intensive part of distribution. As a result, most consumer goods manufacturing companies are feeling the pressure to better utilize their warehouse assets.
Robotic material handling, which has existed in production for quite some time, is finding an ever larger home in distribution. At a growing rate, vision-guided robots are moving into the warehouse, becoming indispensible instruments in the supply chain. And increasingly these robots are mobile, unlike material handling robots in production, which are primarily stationary. As robots move from production into distribution and become more mobile, new market opportunities are arising for automation companies.
To determine the market opportunity represented by robotic warehousing, the new study began with definitions of material handing and warehouse functions. Material handling (MH) is the movement, storage, control and protection of materials, goods and through the processes of manufacturing, distribution, consumption and disposal.
A warehouse (or distribution center) is an essential link in a logistics system. Typically, products enter the warehouse in bulk after completing their transit from production facilities. At the receiving point, products are identified and sorted by type and number and then moved to an assigned, recorded location in the warehouse where they are stored until ordered. In the order fulfillment process, products are retrieved from storage and brought to locations within the warehouse were they are identified, “picked” and packed to assemble complete orders, which are recorded. The orders are then transported to a shipping point within the warehouse, where shipping information is recorded. On a product level, the basic functions of warehousing thus include receiving, identification and sorting, dispatching to storage, placing in storage, retrieval from storage, order picking, packing, shipping and record keeping.
Picking is a particularly important warehouse function for automation. It involves taking different products in different quantities from different parts of the warehouse and combining them for an order. An order is thus typically a request for products that are manufactured in different factories on different production lines. It is much akin to grocery shopping, where needed items are gathered from different aisles of the store.
Robotics in the warehouse involves stationary articulated robots, gantry robots, robotic arms and mobile robots.
1. Stationary Articulated Robots
In a warehouse, stationary articulated robots perform primarily palletizing and depalletizing. Importantly, warehouses present different challenges than in food and consumer goods and manufacturing, where stationary robots have long been used for homogenous palletizing. In a warehouse, pallets are usually built to order and consist of multiple SKUs (stock keeping units) and have various sizes. In addition to the variability inherent in warehouse palletizing, robots are also faced with the need for speed, accuracy, avoidance of product damage and the highest shipping density combined with pallet load stability. Stationary robots are increasingly meeting these challenges in the receiving/replenishment and order fulfillment/shipping areas of the warehouse.
In the receiving/replenishment areas, robots perform layer handling to break down homogeneous pallets and either build multi-SKU pallets to send to shipping or position a layer of cases for transport to a singulator, after which cases are transported to storage. In the order fulfillment/ shipping areas, robots perform palletizing. Here robots build pallets that are typically heterogeneous (mixed SKU) for direct to store shipping where products placed on the pallet may be destined for a single or multiple stores.
It is interesting to note that within a mixed item palletizing work cell, robots may be entirely stationary or mounted on robot transport units to move between build pallets for greater flexibility. (Since they are restricted to the work cell, these mounted robots are not truly mobile.)
2. Gantry Robots
Another type of robot used in warehousing is the gantry robot, a relatively new development in the area of automatic case picking. Gantry robots are bridge-like structures, moving horizontally back and forth along a set of overhead tracks that can span large areas in the warehouse. Attached to the gantry are typically some form of vacuum head which can lift and transport individual or groups of cases to a takeaway conveyor for sorting and shipping. Gantry robots are particularly useful as high-SKU, high-rate case and layer picking applications.
In a potential picking flow utilizing a gantry robot, for example, loads of SKUs are depalletized and transported via conveyor belt to a point where they can be quickly grabbed by the gantry robot and deposited in SKU columns in a temporary, active inventory location. From there the gantry robot can select cases at high speed based on customer orders and build multi-SKU pallets for transport to a shipping station within the warehouse. Alternatively, the gantry robot could select different SKU cases and deposit them on conveyors for transport to a location in the warehouse where palletization occurs.
3. Robotic Arms
A robotic arm is a robot manipulator, consisting of links connected by joints that allow either rotational motion or translational (linear) displacement and an end effector located at the business end of a kinematic chain that is formed by the links. Robotic arms can be programmed or operated manually and can be fixed or mounted on a moving surface.
Robotic arms have been typically used for industrial applications, but Lightfoot’s Agilis robot illustrates that they also have a role to play in warehousing. The Agilis system consists of a base unit that rides along a track that is situated between two rows of pallet racking. On the base is a vertical mast upon which rides a carriage system. The mast and carriage system are also components of an ASRS (Automated Storage and Retrieval System). Attached to the carriage system is a telescopic, robotic arm that can reach into the racking to pick product, which is identified by bar code scanning. By virtue of the back and forth, horizontal movement of the base unit, the vertical movement of the carriage system on the mast and flexibility afforded by the telescopic arm itself, the Agilis robot is highly maneuverable.
4. Mobile Robots
A Mobile Robot is a more advanced form of AGV (Automated Guided Vehicle), which is a non-stationary robot that is guided by markers, floor wires, vision, or lasers. Like AGVs in general, mobile robots serve to replace human travel with robotic travel in the warehouse. Utilizing machine vision, advances in computer processing power and novel computer algorithms in the fields of multiagent systems and control theory, they are highly flexible and capable of adapting to changing products and different conditions. Today’s mobile robots are thus more software-oriented than older generation AGVs, which are more hardware-centric. Their primary purpose at present is to free up skilled warehouse workers from pulling carts, delivering palletized loads and positioning supplies in tight areas.
Two companies that offer cutting-edge mobile robots are Seegrid Corp. and Kiva Systems.
A. Seegrid Corp.
The company name reflects its focus: “See” (vision-guided) and “grid” (utilization of a 3D grid for navigational purposes). Using optics and sensors, the software in Segrid’s robots builds a 3D map of its operating environment. As a result, the robots can generate their own guide path for the movement of materials from different points in the warehouse and with much less restrictive tolerances for making those moves than traditional AGV technology. Programming Seegrid robots is simple according to the company. The robots are “taught” by workers by “taking them for a ride”; that is, instructing the robot along a desired path.
Seegrid’ mobile robots include two models: the GT3 and the GP8. The GT3 is a “general purpose tugger”, the purpose of which is to automate the movement of flatbed cars and carts. To operate the GT3, the worker punches in the destination on the GT3’s controls and the GT3 pulls its load to the indicated destination.
The GP8 is a general purpose pallet truck. A worker operates the GP8 by positioning its forks under the pallets to be moved and then punching in the intended destination. Thus instructed, the GP8 transports the pallets to the desired location, drops off the pallets and returns to the pre-assigned location to start again.
B. Kiva Systems
Kiva robots (or “drive units” in the company’s parlance) look like orange suitcases on wheels. Their purpose is to bring selves of ordered products to workers at assigned pick stations, thus eliminating the need for workers to walk through the warehouse to retrieve the products. This allows workers to focus on filing orders instead of the time-intensive activities of managing locations and searching for products.
To bring the right product to the right station at the right time, a Kiva robot positions itself under a man-high rack. Its upper plate then spins and elevates, raising the rack off the ground. Able to manage a rack weighing 454 kilograms, the lift mechanism of the current model consists of a ball screw that rotates to raise the rack above it. One DC motor turns the screw and raises the rack shaft 30.5 centimeters into the air. To prevent the rack from rotating while the screw turns, the control system causes the robot to rotate in the opposite direction at the exact speed required to keep the rack motionless. After lifting the rack, the robot then navigates through the warehouse to bring the rack to the correct picking station.
The company explains the navigation system of its robots as follows: “The robots navigate the warehouse by pointing cameras at the floor that read two-dimensional bar-coded stickers laid out by hand one meter from each other in a grid. The robots relay the encoded information wirelessly to a computer cluster that functions both as a dispatcher and traffic controller. It instructs, for instance robot no. 1867 to bring rack no. 308 to worker no.12 without colliding with robot no. 1433, which is crossing its path. To fulfill an order, a human operator stands at a pick and pack station on the perimeter of the warehouse. Robots crisscross the floor. They can even use elevators to get to a higher level where they find their order. When the robot positions itself in front of a worker, a laser pointer on a metal pole shines a red dot on the product. Once the worker has retrieved the item, the robot departs and another takes its place. Workers watch for the laser dot, pick a product, scan its bar code, throw it in a box and start over again.”
II. The Robotic Warehousing Market
After examining the various forms of warehouse automation, the study investigated the size of the robotic warehousing market, utilizing available data from market research firms and trade associations. The study found that the global MH robotic equipment market is around $3 billion annually and that just a 10 percent growth in the portion of that market resulting from the increased use of MH robots in warehousing would equate to $300 million annually.
Stationary, gantry, robotic arms and mobile robots have already begun to stake their ground in warehousing. As noted earlier, material handling robots have been traditionally used primarily in production for palletizing, depalletizing and packing. Over time, their use has grown to the point where material handling has become the single largest, robotic application. The next logical opportunity for material handling robots is in the warehouse, but the demands of warehousing are different than the factory.
In the factory, pallets of products tend to be homogeneous. Cartons are typically of one size and contain the same product. In the warehouse, the opposite is the case. Many different products come from different factories, and orders which leave the warehouse typically consist of a host of different products that are oftentimes packed in cartons of varying size. This greater variability makes pallet builds in a warehouse thus more challenging than in the factory.
To effectively handle pallet heterogeneity resulting from mixed SKUs, MH robots must be far more flexible and capable than their factory counterparts. They have to have the ability to recognize different sizes and shapes and adjust accordingly when cartons are misaligned. Additionally, they have to handle cartons safely to prevent damage and stack cartons of different size in such a way to maximize space utilization on the pallet and pallet stability. To achieve these capabilities, bar code scanners are not sufficient. Sophisticated software and vision guidance are needed.
The size of the market opportunity for robotics in warehousing will no doubt also be driven in part by the different types of robots that are available for warehousing. In addition to stationary robots, there are gantry robots, robotic arms, AGVs and state-of-the-art mobile robots. This wide range of robotic equipment amounts to a high degree of versatility to perform warehousing functions ranging from palletizing, depalletizing and packaging to intra-warehouse transport of products. Importantly, with robotic mobility designed for the warehouse, the robotics industry can now compete with manufacturers of conveyor systems.
III. Study Conclusion
The study concluded with a prediction: In the not so distance future, we will increasingly find large, centrally-located, fully automated and computer controlled distribution centers, in which robots busily unload incoming pallets from factories, after which other robots hauling the pallets zip across the floor to storage. At the same time, still other robots retrieve products from storage and transport them to picking stations, where stationary, gantry robots or robotic arms select products and build pallets to order for individual retail stores. Once built, these pallets are taken by robotic tuggers or automatic pallet drivers to shipping.
The warehouse of the not so distant future will thus hum with the computer-orchestrated movements of robots performing essential warehousing functions. Robotic, machine vision and motion control companies will realize this future and profit accordingly.
IV. More Information on Robotic Warehousing
For more information on robotic warehousing, be sure to read ATC’s new study, Robotic Warehousing: Are There Market Opportunities for Automation Companies? The study can be downloaded free of charge by RIA members by logging in to the Robotics Online Member Control Panel from the "White Papers and Market Studies" section. Non-members can purchase the study by ordering for $350 from RIA’s Robotics Online Bookstore (Market Studies).
For more detailed information, be sure to visit Automate 2011 where you speak directly with automation vendors. Co-located with ProMat, Automate 2011 will take place on March 21-24, 2011 at McCormick Place North, Chicago, Illinois. For information on Automate 2011, please visit www.automate2011.com.