Creating a Safety Culture in the Cobot Era of Manufacturing
Automation is indelibly tied to the future of manufacturing. According to figures compiled and released by Association for Advancing Automation (A3), the number of robots sold in North America last year surpassed all previous records. Customers purchased 34,904 total units totaling $1.896 billion in total sales. Automotive-related orders were down compared to the previous year (-7.3 percent in units and -3.8 percent in dollars), but non-automotive orders fueled the rise in 2017 with 20.5 percent growth in units and 7.3 percent in dollars.
Preventing the unthinkable takes a team approach.
The robotics industry also set records for North American robots shipped in 2017: 33,575 robots valued at $1.938 billion shipped to customers last year, up 8.7 percent in units and 6.9 percent in dollars over 2016. Non-automotive related shipments again provided the growth, the highest being plastics and rubber (up 59.6 percent), metals (53.9 percent) and food and consumer goods (44.2 percent).
Yet, sometimes the unthinkable happens. According to a 2017 lawsuit filed by the husband of Wanda Holbrook, a robot unintentionally put Wanda in harm’s way in a tragic accident that ultimately took her life. Holbrook worked as a maintenance technician at Ventra Ionia, in Ionia, Mich., which specializes in welding and stamping truck bumpers and trailer hitches.
By its very nature, the closer collaboration of workers and robots (“cobots”) raises safety issues, a basic one being: can the robot, by itself, assure having a safe application? “Not possible,” says Roberta Nelson Shea, global technical compliance officer, Universal Robots.
In addition to her compliance duties, Shea chairs the International Standards Organization (ISO) committee updating robot safety standards. “You cannot have good safety function without advanced planning and engineering,” she says. “Yes, systems have gotten more complex, but they also now have safety functions that did not previously exist. Systems are getting more reliable and lessening exposure of personnel who would otherwise be tending or fixing machinery.”
Shea describes four types of safety-related collaborative robot operations:
- Safety‐rated monitored stop, where a stop is assured without removal of power.
- Manual (“hand-guided”) actuation and control of the robot by the operator during automatic operation.
- Speed and separation monitoring (SSM), where external safety devices such as safety laser scanners or safety vision systems control speed.
- Power- and force-limited (PFL), where safety functions control speed, torque and motion so impact will cause no injury.
How do organizations start addressing robot safety? “First you need to document a risk assessment – even if it’s only a concept,” Shea answers. Required factors include:
- Describe the function. What is it that needs doing?
- What type of robot gripping or end‐effector is best suited to complete the task?
- Where will this happen?
- What is the initial layout for the application? Consider motions, paths, speed, operator location, contact potential, hindrances to access, etc.
Safety standards define two types of potential contact between a person and any part of a robot system. “Quasi-static” contact is where a body part can be clamped between a moving part of the robot system and another fixed or moving part of the robot cell. “Transient” contact is where the body part is not clamped and can recoil from the moving part of the robot system.
“The big deal is auto-resumption of operation,” says Shea. PFL collaborative applications that comply with safety standards are typically going to “low-payload, low-speed” applications. PFL applications often use a safety scanner to run the application at a higher speed than allowed by the standards language when there is no operator or other human intrusion. Intrusion triggers will reduce speed, which complies when the time needed to slow down is less than the time to stop. Safely slowing down the operation also requires less floor space necessary for a complete stop.
“When at reduced speed settings that comply with safety standards, there is no requirement to stop operation in a PFL application,” Shea adds. “A person could interact with the robot system, subject to the risk assessment. When all people leave the scanner field, full speed and operation is permitted to automatically resume.”
Ultimately, production efficiency, productivity and profits are nothing without functional and assured safety. Best-in-class companies realize driving employee behavior with safe processes, procedures and technologies enable all workers to go far beyond simple compliance and deliver improved productivity and greater efficiencies in addition to dramatically lower injury rates.
Safety culture needs to be seen as a required corollary to production efficiency.
In a world where cobot use in manufacturing production is expected to quadruple by 2025, addressing safety effectively will only grow in importance. “The ongoing challenge is making the complexity invisible and the use easier and easier,” concludes Shea.