Cover Story: How to choose a robot, figure robotic ROI

Control Engineering Cover
07/15/2015

So you want robots? Make smart robotic investments in a rapidly changing world. Consider the system using critical path analysis including robots, humans, and facility infrastructure as part of the same system. Modern sensor technology and safety systems have evolved to the point where robots can operate outside of cages, safely around people. Focus on the goal, not the hype, while figuring robotic return on investment (ROI).

The race to automate is on, and knowing when to choose what kind of robots is important, especially when there’s lot of hype around the topic. Growing manufacturing competition increases pressure to improve customer experience, time to market, quality, and cost (ETQC). Not getting left behind requires an unprecedented level of agility and adaptability that is difficult if not impossible to achieve using traditional business approaches. To get ahead of the competition and to establish a reputation as a industry leader, continual process innovation is imperative.

Technology is changing at such a rapid rate that it presents a challenge in mitigating obsolescence. What is the best way to find the right autonomous solution for today’s job that will adapt and meet your needs over time?

To mitigate obsolescence, it’s important to find a solution that requires minimal investment in infrastructure changes, addresses the ETQC value chain, takes a “mission planning” approach, and makes interoperability a major priority. Smart investments in flexible automation will need to be a key ingredient in your overall strategy to stay ahead in ETQC.

When investing in the next generation of robotics, heed these do’s and don’ts of making smart automation decisions, including figuring robotic return on investment (ROI).

Seek value

DON’T: Buy robots.

DO: Cut through the hype.

Robots are all the rage right now. But hype-based buying may not be a successful business strategy. No doubt that many of you reading this recently had senior management tell you to “formulate our robotics strategy,” a hint of panic in their voices, “we are falling behind.” The pressure to automate is real, and for many, smart automation will be paramount to survival. Cutting through the hype and focusing on the real goal you are trying to achieve is what you need to do to win. Don’t just “buy robots,” instead think in terms of investing in systemwide capabilities that will bring long-term value and efficiency to your total ETQC efforts.

The right reasons

DON’T: Chase shiny objects.

DO: Appropriately value novelty.

So right up front, I need to acknowledge that using robotics as a “novelty play,” can be part of an overall successful strategy. But proceed with caution. I define a robotics novelty play as when the idea that you are using a robot is more important than the direct value the robot is actually providing. I have heard from many customers that “we want to be seen as innovative,” or “we need to show our investors we are automating.” While there is value in being seen as innovative, that value will be short-lived if the innovative part didn’t happen. Novelty should generally be valued more highly in consumer markets, while concrete process improvement should generally be the primary motivation in industry.

For an example, two of the most successful next generation robotics companies today are iRobot and Kiva Systems.

iRobot’s Roomba vacuum cleaner robot is the most widely sold robots of all time, having sold millions around the world. iRobot wisely makes sure people know it is a robot first and foremost. People respond well to that, some going as far as considering it part of the family. Cottage industries have popped up to provide robot clothing for it. Novelty is a major part of the value proposition, and most still buy traditional vacuums as well. Had iRobot marketed it instead as an “automated floor cleaning system,” it is easy to imagine a less enthusiastic consumer response.

For Kiva, with a clear target in the order fulfillment market, novelty has a much smaller role. Kiva robots bring the shelves holding many Amazon.com purchases to a human for picking and packing, saving massive amounts of time that would otherwise be wasted by human pickers walking from shelf to shelf to fulfill an order. The Kiva implementation is heavy on process improvement, with novelty playing a smaller role.

Consider the system, ETQC metrics

DON’T: Find something to automate.

DO: Consider the whole system.

Novelty value aside, it’s tempting to look at the factory floor, point to specific problems with existing processes, and buy a robot that attempts to overcome that specific problem. But how do you know that adding automation at that point will have any positive impact on the actual metrics you care about? Critical path analysis looks at an existing system and answers the question “where is the critical point in the system that is the current limiting factor in improving ETQC metrics?”

For example, if the goal is improved output of a manufacturing process, making improvements anywhere in the system other than that critical point is a waste of money, as it will have no positive impact on your goals. Because the critical point is ultimately the “rate limiter,” improving the speed anywhere else in the system makes no difference in the outcome. Critical path analysis is a great way to ensure potential improvements to an existing system that is already producing reasonable results will provide the desired ROI.

Of course, critical path analysis only goes so far. The addition of a system-wide planner can make a huge difference in overall performance of an operation. Think of all of the “actors” in a system. Suppliers, people, robots, machinery, and customers all need to be coordinated to achieve optimal results. Just buying robots without a centralized planning and optimization system will likely achieve poor overall long-term performance.