Although robots are not new to manufacturing, recent advances in technology and the availability of lower-cost products should expand the use of robotics. Combine these trends with other forces, including scarce labor, worker safety, productivity and ergonomics, and robots may become a stronger force in many areas of manufacturing.
Like any new piece of equipment, robots take time to configure and install. However, improved robot control and connectivity have smoothed the process somewhat. In addition, the growing use of simulation, which simplifies and expedites the programming of robot paths, has accelerated the start-up process -- making robots available to more, and smaller, operations.
However, the greater availability of the technology raises other issues in the manufacturing operation. What cannot be overlooked in new installations is the human factor, especially if the robot will automate a manual operation. Communication and training are essential to defuse employee fears about job loss and ensure worker safety around the robot cell. "The impact of robots will be felt plant-wide," according to Robotics for Senior Management, a paper published by the Robotics Industries Association (Ann Arbor, MI).
The paper also notes the necessity of a senior management champion, who can "provide resources as well as moral support" and a systematic approach to implementation. Implementation steps include an initial survey to identify potential applications, qualification to identify the most practical and value-generating applications, prioritization of applications, economic and non-economic justification, application engineering for the robot and its task(s), system engineering to prepare equipment, processes and people for integration of the robot.
PURPOSE-BUILT SYSTEMS
Once the decision is made to use robots, the type of robot needs to be determined. Although general-purpose robots are widely used, many manufacturers are now preferring to use purpose-built systems. Dedicated units have long been used for activities, such as welding and painting, but now are expanding into functions such as palletizing and case packing. "A lot of people do buy a generic robot, but picking an application-suited one requires fewer compromises," explains San Jose, CA-based Joe Campbell, director, strategic alliances at KUKA Robotics Corp. (Clinton Township, MI).
An example of this transition from general-purpose to application-specific designs is a modular, pre-engineered palletizing product from Motoman Inc. (West Carrollton, OH). Designed specifically for high-throughput palletizing, the modular system includes standard components like the 100-kilogram payload SP100X robot or the 300-kilogram payload EPL300 with controller and menu-driven palletizing application software plus an integrated gripper package. Optional pre-engineered modules for infeed conveyors, pallet locators, pallet racks and pallet outfeed conveyors match equipment capability to product handling requirements and simplify customizing or upgrading.
Another system dedicated to palletizing is KUKA's KR 180 PA robot. "We have been palletizing with six-axis robots for about 10 years," reports Campbell. "About 18 months ago we developed a dedicated product specifically for palletizing," he adds.
The four-axis KR 180 PA robot eliminates any features not related to palletizing and operates within a highly optimized work envelope. Even the structure of KR 180 PA is designed to boost the speed and performance of palletizing tasks through the use of lightweight, yet strong, carbon fiber composite components in the arm. "One way to increase speed is to take mass out of the arm," explains Campbell.
The KR 180 PA design has proven to be so effective, KUKA has introduced a freezer version and a medium-duty version, the four-axis KR 100-2 PA, which is capable of handling lighter payloads weighing up to 100 kilograms.
A new version of KUKA's Pallet Pack software delivers added functionality for pallet picking, layer management and depalletizing, and helps manufacturers build the custom pallets customers want. Many retailers require shipments to be delivered a certain way on a certain pallet with a certain pallet pattern. In many cases, this doesn't match how manufacturers palletize product coming off their lines or how other customers want to receive it. So, a robot is used to depalletize and repalletize the product to customer specifications. "Unstacking and stacking operations are a perfect activity for a robot," notes Campbell. Sensors are often used to help register the load.
With robots playing an important role in how customers receive shipments, many manufacturers are moving to integrate enterprise systems all the way down to the robot level. Advances in technology have made a higher level of connectivity possible between robot and back-end systems, such as enterprise resource planning or warehouse management. Robots also are proving suitable for use in the pick-to-order process, especially for mixed-load orders where the warehouse needs to pull a mixed lot of products and stack them on a single pallet. Robotic order picking is especially well-suited for handling heavy products such as beverages and canned goods.
As connectivity between the enterprise and robots increases and the use of application-specific robots becomes more widespread, vendors will continue to develop more and more specific functionality for their products. "All the robot makers are on two development paths," notes Campbell. One focus is the design of application specific robot families with members of different sizes and payload capacities. "A broad product line allows customers and integrators to match the mechanism to the process," he explains.
The second focus, performance improvement, is timeless and universal. "There's always going to be advances related to speed, reliability, flexibility, precision, software and controls," he says.
In addition to a proliferation of application-specific designs, Campbell expects a surge in the use of advanced materials, such as carbon fiber composites, to replace traditional precision-machined aluminum castings. These new materials offer cost advantages, performance enhancements or both, helping to increase the adoption of robots throughout the enterprise.
Also, other advances will focus on controls where intelligence is increasing, form factors are shrinking and connectivity is broadening. "The old way to make a robot go faster was to push more power into it by specifying bigger gear boxes and other components. Increasingly, higher speeds are made possible by making the unit move more efficiently," reports Campbell. For instance, controls are transitioning to open architectures like personal computer platforms. This enhances connectivity with other equipment and plant networks, and opens the door to the use of third-party software products.
"The level of data integration inside factories is incredible," says Campbell. "Controller technology now is thoroughly robust and can happily communicate with any level of intelligent device on the floor," he adds. "As a result, manufacturers are starting to build up a base of experience that will allow them to implement data driven applications in a reliable fashion."
BRIDGE OVER TROUBLED WATERS
However, bridging the gap between the manufacturing world and the data world presents new concerns. Overcoming these concerns is a key to improving the adoption of robotics. In a data-driven system, everything is driven by information moving between databases. As a result, when the data world is integrated with robotics, there have to be contingency and recovery plans for glitches, such as power failures, to ensure data is not lost or corrupted and that the robotic system will power up and resume operation at the appropriate point.
To address this issue and eliminate start-up delays, Motoman's PalletWorld software incorporates special power failure recovery processing and absolute encoders that require no homing operation after power loss.
Another way that software is helping in the adoption of robots is in implementation. More user-friendly software makes installation and setup simpler. Increasingly, the setup of robots relies on simulation to optimize and program work paths. For example, RoboGuide Off-line Robot Simulations software from FANUC Robotics America Inc. (Rochester Hills, MI) provides engineers with the tools needed to develop and test a complete robotic application in a simulation environment, eliminating the time and costs associated with developing a prototype work cell.
"RoboGuide is a great time and money-saving tool for sales, proposal and application engineers who develop robotic systems," said Gordie Geheb, director of process solutions, FANUC Robotics. "To be competitive, companies must utilize tools that quickly provide reliable, accurate information about a robotic application without the physical need and expense of a prototype setup. RoboGuide provides these tools."
With RoboGuide, engineers can import unique CAD models of parts; create a work cell including machines, part transfer devices and obstacles; and teach robot paths to simulate the operation and performance of a robotic application. Reach verification, collision detection, accurate cycle time estimates and other visual system operations are simulated in RoboGuide's graphical virtual environment. Process-specific plug-ins provide simulation for general-purpose material handling applications (HandlingPRO), arc welding (WeldPRO), painting (PaintPRO) and palletizing (PalletPRO).
Similarly, Motoman's MotoPallet PC-based simulation software is closely integrated with its MotoSim offline programming and simulation software. A process programming approach allows the user to think in terms of boxes, pallets and pallet patterns rather than robot, cycle times and payloads, and makes it easy to optimize robot programs based on box size, shape and orientation requirements.