Managing Robots

Stephen DeAngelis

December 7, 2012

“In Isaac Asimov’s classic series of short stories I, Robot,” writes Tim Criswell, President of Wynright Robotics, “he lists the first law of robotics as, ‘A robot may not injure a human being or, through inaction, allow a human being to come to harm.'” [“Helping Robots and Humans Learn to Play Nicely Together,” SupplyChainBrain, 16 February 2012] He notes, however, that “in the real world of warehouses and distribution centers, there is no such caveat. As a result, robotics are usually either isolated completely or cordoned off by fences, light curtains or other barriers in order to prevent humans from putting themselves in harm’s way.” Getting robots to work side-by-side with humans without putting them at risk is a challenge that is just now being solved. In a post entitled Meet Baxter — Your New Co-worker, I discussed the fact that the latest trend in robotics is developing robots that can work side-by-side with humans. Right now robots of all shapes and sizes are being developed to perform a myriad of functions alongside humans. If you’d like some idea of the kinds of robots being developed to work closely with humans, click on this link that will take you to a Washington Post slide show called Robot Mania. Criswell discusses some of the challenges involved in making robots safe to operate with humans. He writes:

“While creating a ‘no fly zone’ for humans makes sense from a safety standpoint, and may even be required by Robotics Industry Association (RIA) standards (which OSHA follows), it takes up valuable floor space that could be put to much better use. In short, although robotic isolation works, it’s the equivalent of not allowing any development on the valuable land around airports just in case a plane crashes. The effect in an operating environment is that valuable space is routinely rendered unusable. This approach also gets in the way of efficiency. Currently, when a robot needs help to correct an error, the operator must enter the safety zone through a gate in the physical barrier. Many times entry calls for following a lock-out procedure to ensure the system is not inadvertently re-energized by someone else. This approach limits the way an operator can access the system to correct the error, which adds time to recover, lengthening downtime and reducing overall productivity.”

Criswell believes, as do others, that “rather than building safety zones, the smarter solution is to re-think how humans and robots interact, or should interact, to integrate that first law of robotics as part of the fundamental design of the unit.” He continues:

“This way of thinking will revolutionize supply chain design. In order for humans and robotics to work in close proximity, the robotics must be intrinsically safe. Having sensing devices, whether based on lasers, RFID tags worn by workers or a technology that hasn’t been created yet, will use space more efficiently and allow related tasks in the same area to be performed by the entity (human or robot) that is best suited to it. In the event of an error, it also offers more paths to the robotic equipment (instead of having to approach through a single gate in a physical barrier), which will make recovery quicker, reducing downtime.”

Criswell points out that simply causing a robot to go to “full stop” when a human approaches is not a good enough solution to the problem of humans and robots working together. “As software and sensing technologies improve,” he writes, “new strategies may have the robotics slowing down to safe levels (i.e., those that will not cause damage to humans if there is an impact) when danger is sensed. Once the human has moved a safe distance away, the robot will resume working at its normal pace without the need for a re-start.” Baxter goes a step further and is capable of working side-by-side with humans. That’s the real future. Criswell continues:

“Why the need for such close interaction between humans and robots? It’s all part of the evolution of the modern supply chain and the never-ending battle to improve productivity. The considerations are both practical and economic. It is well-established that robots are the best choice for large-volume, repetitive tasks where they can do 90 percent or more of the work. When there are no decisions to be made, no criteria to be considered other than what was pre-programmed, robots can deliver significant labor savings. In situations where robots can only perform 80 percent of the work, however, the economics break down. If a human operator needs to get in and out quickly to handle the other 20 percent, the difficulties with implementing a safety system can negate any savings from using robots and may even drive the total cost higher.”

Criswell points to “a 2007 study by NASA’s Ames Research Center” that concluded “a team of humans and robots could accomplish more work than either a team of humans or a team of robots working without the other.” He also notes that designing workplace infrastructure is also easier if humans and robots can work safely side-by-side. He concludes:

“Asimov’s vision of humans and robots working side-by-side is no longer the stuff of science fiction. In the supply chain, it’s science fact. Replacing bulky physical barriers with new technologies that make safety intrinsic to the robot not only protects operators and other workers more effectively, it improves productivity. All while allowing the use of robots in situations where it was once cost-prohibitive, helping drive down overall costs while delivering greater ROI on those units.”

While many people don’t find it difficult to imagine a world in which robots and humans routinely work together, most of us haven’t thought about what that scenario means for management. “We love robots – tireless, productive workhorses of the modern assembly line,” writes Andrew Hill. “But we also hate robots – sinister mechanical simulacra of the human workers they make redundant.” [“Robots put leadership under skills pressure,” Financial Times, 8 August 2011] Hill notes that “over time, greater automation does hand an advantage to employees with more skills.” That fact alone indicates that management styles must change with the times. Hill continues:

“In developed economies, Lynda Gratton writes in her new book The Shift, ‘when the tasks are more complex and require innovation or problem solving, substitution [by machines or computers] has not taken place’. This creates a paradox: far from making manufacturers easier to manage, automation can make managers’ jobs more complicated. As companies assign more tasks to machines, they need people who are better at overseeing the more sophisticated workforce and doing the jobs that machines cannot. … The insight that greater process efficiency adds to the pressure on managers is not new. Even Frederick Winslow Taylor – these days more often caricatured as a dinosaur for his time-and-motion studies – pointed out in his century-old The Principles of Scientific Management that imposing a more mechanistic regime on workers would oblige managers to take on ‘other types of duties which involve new and heavy burdens’. Knowing the challenge does not make it easier to deal with. According to Edward Lawler, the co-author of Management Reset who worked with Ford and General Motors in the 1980s and 1990s as they grappled with technological and process change, carmakers found it hard to give up a command-and-control model of management. … The big question is how easily [companies] will find and develop managers able to oversee the highly skilled workforce that will march with their robot armies.”

Reporters from The Economist agree with Hill that “management thinkers need to ponder more about homo-robo relations.” They write, “Robots have been the stuff of science fiction for so long that it is surprisingly hard to see them as the stuff of management fact.” [“I, robot-manager,” 31 March 2011] The article continues:

“Robots have been doing menial jobs on production lines since the 1960s. The world already has more than 1m industrial robots. There is now an acceleration in the rates at which they are becoming both cleverer and cheaper: an explosive combination. Robots are learning to interact with the world around them. Their ability to see things is getting ever closer to that of humans, as is their capacity to ingest information and act on it. Tomorrow’s robots will increasingly take on delicate, complex tasks. And instead of being imprisoned in cages to stop them colliding with people and machines, they will be free to wander.”

The authors note that many robots “are starting to look less like machines and more like living creatures.” In the case of Baxter, that was deliberate to make human co-workers feel more comfortable working with it. The article continues:

“Until now executives have largely ignored robots, regarding them as an engineering rather than a management problem. This cannot go on: robots are becoming too powerful and ubiquitous. Companies may need to rethink their strategies as they gain access to these new sorts of workers. Do they really need to outsource production to China, for example, when they have clever machines that work ceaselessly without pay? They certainly need to rethink their human-resources policies—starting by questioning whether they should have departments devoted to purely human resources.”

In other words, as more robotic co-workers come on the scene, are HR departments ready “to deal with how to manage the homo side of homo-robo relations”? The article continues:

“Workers have always worried that new technologies will take away their livelihoods, ever since the original Luddites’ fears about mechanized looms. That worry takes on a particularly intense form when the machines come with a human face. … The arrival of increasingly humanoid automatons in workplaces, in an era of high unemployment, is bound to provoke a reaction. So, companies will need to work hard to persuade workers that robots are productivity-enhancers, not just job-eating aliens. They need to show employees that the robot sitting alongside them can be more of a helpmate than a threat. … Employers also need to explain that robots can help preserve manufacturing jobs in the rich world: one reason why Germany has lost fewer such jobs than Britain is that it has five times as many robots for every 10,000 workers.”

The article then raises the question of what department deals with the robo side of homo-robo relations or, just as importantly, robo-robo relations. It explains:

“Robot scientists are tackling more complicated problems as robots become more sophisticated. They are keen to avoid hierarchies among rescue-robots (because the loss of the leader would render the rest redundant). So they are using game theory to make sure the robots can communicate with each other in egalitarian ways. They are keen to avoid duplication between robots and their human handlers. So they are producing more complicated mathematical formulae in order that robots can constantly adjust themselves to human intentions. This suggests that the world could be on the verge of a great management revolution: making robots behave like humans rather than the 20th century’s preferred option, making humans behave like robots.”

Tim Criswell concluded his article with the battle cry, “Humans and robots unite!” As Hill and The Economist point out, just getting humans and robots to get along with each other is going to be a challenge for managers.