Hartford Business Journal

HBJ101424UF

Issue link: https://nebusinessmedia.uberflip.com/i/1527718

Contents of this Issue

Navigation

Page 24 of 35

HARTFORDBUSINESS.COM | OCTOBER 14, 2024 25 FOCUS | MANUFACTURING far from universal. A recent survey by the Connecticut Business & Industry Association found that of the Connecticut manufacturers that are planning to make changes to their workforce in the near future, 43% expect to add more employees, while 18% plan to invest in automation to streamline operations and increase efficiency, and 9% are considering adopting artificial intelligence appli- cations to enhance capabilities and drive innovation. But those who have made the switch are often evangelists for this new way of doing business. Shelley Fasano, the CEO of Dymotek, an injection molding specialist in Ellington, told the audience at a recent manu- facturing summit in Hartford that auto- mation is now at the forefront of any new customer relationship. "We're starting at the beginning with every new product launch," she said. "Before we would start by saying 'we'll throw labor at it and then we'll automate over time.' And now we will not take in business that we cannot automate, because we will not be successful." Man & Machine Sam Greenbank, a manufac- turing applications engineer at the Connecticut Center for Advanced Technology (CCAT) in East Hart- ford, said the concep- tion of what robots can do, and where they can be useful, has changed considerably over the last two decades. "Most automated projects or robotic processes in the past were based around a hard automation or fixed automation," Greenbank said "That's characterized by a machine that's built to do a specific task. If that task changes, you need to rebuild the system." These were the classic robotic assembly lines that became typical in the automotive industry in the 1980s. Today, robots are built with adapt- ability in mind, Greenbank said. "Instead of doing one task and one task only over the entire lifespan of the robot," he said, "the idea is instead to make the robot as adaptable as possible so that it can be used for one task, or it can be used for a different task every day." That means they can be used for short-run production of just a few hundred, or even a few dozen, parts or packaging tasks, before being quickly and easily reprogrammed. That kind of low-volume, high-mix manufacturing is something that's far more typical of the aerospace businesses that form the backbone of Connecticut's small and medi- um-sized machine shops. Even newly adaptable as they are, these kinds of industrial robots are still kept strictly segregated from humans on the factory floor, usually behind cages, because of the danger of injury. But they've been joined in many industries by collaborative robots, like the ones being used by Sargent Manufacturing in New Haven. They're designed to work "alongside of humans, to serve as an assistant, doing jobs that are more human-cen- tric, using human tools, using more human logic, as opposed to comput- erized work code," Greenbank said. They're built with safety in mind. Cobots have sensors and force-sen- sitive feedback designed to stop them if they come into contact with a human colleague, he said. They're also much easier to program — many of them can learn new tasks simply by being physically manipulated by the people they're working with. "The downside to these is, to be safe, they must be slow," said Green- bank. "They're typically not built for the same level of ruggedized infra- structure as industrial robots; their payload is much less, so they're only good for light-duty tasks." Nevertheless, they're becoming more prevalent in industrial settings, doing things like loading parts in and out of machine tools or molds — robotic machine-tending. They're even more common — for big companies ubiquitous — in ware- house settings. Greenwich-based logistics company GXO caused a stir recently with the implementation of a humanoid robot called Digit that it's using to move heavy loads from mobile robotic carts to conveyors at one of its fulfillment centers in Georgia. The company is also piloting two other humanoid robot models for logistics functions alongside its human workforce. CCAT's Greenbank is dubious about the use-case for humanoid robots in manufacturing's more complex environments — at least just yet. "They're coming," he said. "Person- ally, I don't think they're going to be industrially relevant for at least another decade or two. There's too much that they have to solve." Automation ROI Because of the complex and often unique nature of the tasks, most robotics solutions don't come ready off the shelf for manufac- turing operations. That means auto- mating a process can be a lengthy and expensive investment, involving customizing a suite of technologies for the precise use case. CCAT principal engineer Nasir Mannan spends a lot of his time breaking this down for company executives who are trying to make smart choices. Often, he says, firms will bring in the actual parts they need to manipulate to run through CCAT's demonstration set-ups. "They're trying to see what the value is of adding automation," he said. "We have systems here at CCAT that are either existing in place, or we quickly mock up with the robotics that we have so that we can show them what value that robotic system can give them." He can then walk them through an estimated cost and time to implement what they need, and also how long it would take to retrain their workforce around that system. "So, that translates to an ROI," Mannan said. Sam Greenbank Shelley Fasano 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 $35B $30B $25B GDP OUTPUT Connecticut manufacturing GDP output Source: 2024 CBIA Connecticut Manufacturing Report, U.S. Bureau of Economic Analysis, U.S. Census Bureau

Articles in this issue

Links on this page

Archives of this issue

view archives of Hartford Business Journal - HBJ101424UF