The competitive landscape presents a constant pressure to lower prices and run more efficiently. When dealing with a single product being made in manufacturing, it is not difficult to identify process improvements that increase efficiency. If several different products are being produced, an improvement to just one production process is unlikely to have a significant impact on ROI.
A high-mix plant produces many different parts, each of which has different processes - such as CNC machining, stamping/forming, visual inspection, and picture documentation - that all require some human involvement (loading/unloading parts, placing parts in packaging, taking and saving pictures, etc). A total overhaul to automate processes is expensive (not to mention daunting) and low or unpredictable volumes for any individual part makes the per-unit cost of such a large capital investment difficult to justify.
Below are some examples of typical high-mix processes, and the challenges of adding automation to them:
CNC machining of different parts of varying lot size.
Fixturing and removing each part from machine vise(s).
Part geometries require wide array of gripping mechanisms, and varying locations for placement in vise.
Visual inspection of parts/assemblies.
Products require pictures taken from various angles and saving as records.
Many configurations of camera locations for each part are difficult to make consistent. Creating fixtures for each type of part can be expensive and ineffective if the part undergoes design changes. Saving and labeling pictures manually can be time-consuming.
Packaging of several sizes of parts into boxes.
Pick up part, place in bins or boxes as required.
Optimal packing configurations may be drastically different depending on part and package sizes.
At the foundation of any high-mix operation is the concept of changeover time. Any endeavor to automate must take into account the non-value-add time associated with changeovers, or eliminate them entirely. With this in mind, here are three paths possible to automate in a high-mix environment.
Traditional (Fixed) Automation
There are nearly endless options available for traditional fixed automation, including industrial robots, gantry-style cartesian robots, conveyors, or even a turnkey machine designed by an integrator. The main takeaway for this category of solutions is that they are typically going to be fixed to one or a handful of products in your mix.
On the positive side, the “tailored” nature of this category of solutions means they tend to be very efficient for the processes on which they are deployed. If cycle times have to be much lower than 2 seconds, then traditional automation may be the simplest path to improving a process.
However, in a high-mix environment, fixed automation generally means a decision between a lower capital investment that only touches a few of your products or a much higher investment for a system that can handle all of your products. Traditional systems tend to be designed for throughput, not rapid flexibility, so tooling changeovers can take considerable time, especially when safety protocols require physical separation from human traffic. The total cost of ownership, when considering the safety equipment, floor space, and setup needed for traditional safety methods, tends to be high. In any case, this lengthens the ROI period longer in a high-mix environment, particularly if volumes are relatively low.
Robots as a Platform (RaaP)
The available RaaP solutions can be as varied as your product lineup, but one thing they have in common is using the robot as part of a system designed to be quickly redeployed as needed. Rather than the traditional deployment paradigm of coupling a robot to one specific process, RaaP enables the robot to go wherever production demands on any given day. The ideal solution in a high-mix environment should have the ability to be quickly deployed from application to application with as little changeover time as possible. This minimizes the non-value-add changeover time that can make traditional automation prohibitively expensive.
One option here is the CrossRobotics UR Kit, which involves a UR Collaborative Robot (CoBot) mounted on a mobile adjustable-height pedestal, with a preinstalled Robotiq Adaptive Gripper. This platform allows rapid deployment of a CoBot while also being able to cater to a wide variety of applications. The robot’s Collaborative capability means that guarding may not be required after an application-based risk assessment. The removal of guarding and other traditional safety equipment greatly enhance the ability of the UR to be easily deployed and moved around the plant, typically with considerably less upfront investment than a non-collaborative solution. The 85mm stroke of the adaptive gripper allows for a wide variety of parts to be handled out of the box, with gripper finger changeovers being done in minutes.
If we need a true multitool, READYRobotics is a recent and exciting entrant to RaaP that offers the TaskMate platform with a UR CoBot at its center.
This platform is intended to augment existing processes as a retrofit - it offers the SwifTool interface end-of-arm tooling, the TeachMate module on the arm to provide pneumatic and signal feedthroughs at the tool flange, and even the pneumatic PedalMate to work with existing foot pedals with no modification to existing equipment. One particularly effective feature of this platform is the R-Align system, which incorporates a probe-and-socket locating mechanism to recalibrate waypoints as the platform is moved around to different locations around the facility.
As an example of why the TaskMate’s features are so powerful, imagine the robot tending several CNC machines. This involves several waypoints for each application, each of which is typically susceptible to error every time the robot is moved to another location. The R-Align system allows the platform to be rolled roughly into position, hand-guiding the end-of-arm probe to the socket mounted on the machine, running the calibration wizard, and… that’s it. All program waypoints are now automatically recalculated relative to the socket location and orientation. What could be a tedious and time-consuming changeover process can now be done in less than 5 minutes and requires no tools.
Both of these platforms can be placed exactly where they are needed to expand existing capacity while allowing quick changeovers, increasing ROI and shortening the payback period on the investment.
High-mix production environments can be difficult to improve upon. The ideal improvement will touch several production lines, but the investment required to accommodate several processes can be expensive enough to significantly decrease ROI. Traditional automation is certainly an effective solution in high-volume or rapid cycle-time production. For a high-mix scenario with a lot of manual involvement, it may be more beneficial to look at CoBot solutions either by themselves or as part of a platform.
Manan is a North Carolina native - born and raised in Cary, eventually moving to Charlotte in 2005 to attend college. He graduated from UNCC with dual degrees in Mechanical Engineering and Physics, later completing his MBA with Wake Forest University. While working for several years as a design engineer, Manan realized his affinity for robots and making processes more efficient. He joined Cross Company in 2017 as a Robotics Technical Specialist, where he uses robots to help people work more efficiently.