Packaging World

Have a kid wondering what to major in at school? Here's a thought-provoking Artificial Intelligence thought from AI trailblazer NVIDIA's Rev Lebaredian from Rockwell Automation's #ROKevents Automation Fair this morning, who spoke with Rockwell CEO Blake Moret. Lebaredian said, “Up until pretty recently when people would come up to me and ask, 'what should my child study in college?’ Typically, I’d tell them computer science, at least as a minor. Because no matter what you do, it would be very useful to be able to understand how a computer works and translate what you need into a computer program to help you build tools for that thing you're interested in. Nowadays, I don't think that's good advice. What's more valuable is having that domain experience. So I tell them, ‘have them go study material science or, or physics, fluid dynamics, or pharmaceuticals and medicine, things that are important in the real world around us. And go deep in that, because we're essentially going to have an unlimited number of [artificial] computer scientists and computer programmers to help you there. Of course, there's still going to be a need for computer scientists like me to build the, the systems that will produce the virtual computer scientists we can all take advantage of. But the value of the domain knowledge is now far, far greater than the knowledge on how to program a computer, or how to build a computing system. That's going be taken care of for everyone [by AI]. So we humans can go deeper into each domain.” "It's really unique in the history of computing, in that the more advanced this technology gets, the easier it is to use. That hasn't been true for most of the other technologies that preceded it. Using natural language to to help with programming is unusual ... up until recently, if you want to write a computer program tell a computer what to do, you have to study a lot, you have to go do a lot of work to understand how a computer works, how computer languages work, and how to make them do these things correctly. But we're in an era now where we can speak to a computer with natural language, tell it what to do, and it translates our intentions into an actual program. It's not perfect yet, but it's actually pretty good already, and it's getting better and better, and it's not going to be long until essentially every person on earth is going to be a computer programmer, because they can program just by instructing a computer in the way they talk to any other, any other human, any other agent that walks from amongst us." More: https://lnkd.in/g-KWKFnP

It was an exciting day yesterday at the AIPIA & AWA Alexander Watson Associates Smart Packaging Congress as Johnson & Johnson Innovative Medicine and Merck Healthcare shared details on their work with digital display labels for clinical trial drugs. #smartpackaging #clinicaltrialpackaging #digitaldisplaylabels #smartpackaging #connectedpackaging

An audience question put a bow on a fascinating panel discussion at Rockwell Automation's #ROKevents Automation Fair today. The discussion featured Danone #packaging and #processing automation engineers explaining their use of Rockwell's always-evolving FactoryTalk AssetCentre tool. The discussion worked working through future features, layers of tech being added, and what brand owners and CPGs might want to see from a potential #AI integration with the tool. The audience member asked: "Regarding AI, I’m seeing a lot of different applications, where each one has some kind of intelligence built into it, or has a lot of predictive features. But is there a central application that could hold all that information and maybe give you some insights of what it’s seeing?" Answered Jon M. (Jon Mills) at Rockwell: "One of the realities of the generative AI that’s powering a lot of these things is that what I would consider to be a turn-key solution doesn’t exist yet for solving problems. So you see AI popping up in all of these tools, but its expression is often very specific to that tool. The reason for that is because you have to interact with the technology in a way that really drives it into trying to solve a specific problem. So having a one-size-fits-all AI solution that is able to interact with multiple systems and try to solve multiple types of problems? I still think there’s probably a technology gap that prevents us from making too much progress in that direction, yet." Added Danone's Jerel Williams: "Thinking about it like we did with robotics development, you have repetitive tasks. I imagine the first step we’re going to see with AI is that we’re going to have to generate a bunch or rungs that are just tedious work, and it’s going to take all that out. The first step of its evolution is going to be doing the simple tasks and correlating varying simpler things together. And as we get used to it, we’re going to see where its limitations are, time will start to push it to the next level. As industry evolves, AI is going to evolve as well." "And you see that same type of evolution in similar product offerings that are out there in the industry," Mills concluded. "I’ll pick on [a code-writing AI assistant product]; it’s basically a coding assistant that lives right inside of your IP. When it was first released, it was basically only able to complete the line of code you were starting. But now, they've released an update that allows you to edit across multiple files, and give it very vague instructions to go out and use chain of thought-style interaction for it to perform more and more complex tasks. So the idea of AI evolving over time, I think is definitely occurring." More here: https://lnkd.in/gGz_6_XU

Shown at the AIPIA & AWA Alexander Watson Associates Smart Packaging Congress in Amsterdam today, a new reusable packaging collection system puts the reverse logistics and cleaning in the hands of the venue in which it's used. The technology provider, Cuploop, is targeting closed-loop environments currently, such as universities and sports venues, but hopes to make it ubiquitous in cities one day.

Here's an eye-opening predictive analytics case study from Rockwell Automation's #AutomationFair #ROKevents today: Maintaining precision in fill levels is crucial for both product quality and cost efficiency. On its 320 jar/min?Hellmann's?mayonnaise #packaging line,?Unilever’s?Pouso?Alegre facility in Brazil faced challenges with its traditional PID (proportional integral derivative) loop system, which relied on feedback from downstream checkweighers to adjust fill levels. This reactive approach led to overfilling, increasing material costs and scrap while reducing overall efficiency, according to Unilever's Jean Carlos Ramos and Denis Castro de Almeida. Rockwell's FactoryTalk LogixAI let them predict fill amounts using what could be called, "a soft sensor approach, so think of it as software as a sensor,"?said?Rockwell's Richard Resseguie. "The whole theory behind this is what if we can develop a soft sensor that's going to predict what the fill amount is going to be in the jar, before you actually go and fill it, based on the process variables." LogixAI's implementation involved integrating data from various stages of production, including upstream filtering and mixing processes ahead of the filling itself. This data was used to train a predictive model that could anticipate the dosed weight based on those variables. “We're doing this directly on the edge, right next to the equipment. That's how we're gathering the data, how we're training the model, and then how we're adjusting it,” Resseguie added. This edge-based approach allowed the company to analyze and predict potential drifts in equipment performance, enabling real-time adjustments at the edge. The predictive model was trained using both historical and live data, which helped in exploring the feature space and determining the relationship between input variables and the desired fill level. “Once we determine which variable is contributing in this use case, we use that prediction now to use that for the next jar and then determine, okay, you're about to overfill by let's say three or four grams. We can make that adjustment and reduce it,” Resseguie said. By reducing the standard deviation of fill amounts across different filler heads, the company was able to operate closer to the target set point, significantly minimizing waste. The process involved selecting key input variables through a correlation matrix and using them to train the model. The representative described the process: "You can drag and drop and select those inputs and then from there you train the model." The integration of LogixAI into the company's existing systems was facilitated by its compatibility with Studio 5000 and ladder logic, allowing for seamless control and automation at the control layer. “You actually have variables that control when you train, when you calculate. So that's what allows you to truly automate this entire process at the control layer,” Resseguie said. https://lnkd.in/g5M7JYdC

Paperization is even making its way into RFID tags. Chloe So of PulpaTronics gave a fascinating presentation today at the AIPIA Congress on her company's development on a metal-free laser-etched paper RFID tag.

Ready to kickstart your automation journey? Join our webinar TOMORROW, November 19, 2024, to discover how robotic palletizing is an excellent entry point into automation. This free event, hosted by Packaging World, will provide a comprehensive guide to implementing your first robotic palletizer, equipping you with essential knowledge and strategies for success. Why Attend? -Identify key factors for selecting the right palletizer for your operation -Learn about facility preparation and layout optimization for robotic systems -Explore integration, software considerations, and workforce training for successful implementation. Don't miss insights from industry leader Steve Koenig. https://buff.ly/3CCFWzA

It was a great start to the AIPIA conference in Amsterdam this morning. LIPTON Teas and Infusions' Alix Courdier discussed his company's use of smart packaging to connect with consumers.

We heard rumblings of a big Paper Snap application announcement, though no details could be shared, when we stopped by the Easysnap Technology booth and spoke to Alessio Bonizzato at PACK EXPO last week. Sure enough, only a few days later, news of the new Paper Snap #packaging being adopted by Mars brand Masterfoods, making the company's popular single-serve tomato sauce packs curbside recyclable in Australia, and improving the consumer experience in single-dose sauce application now that it can be done with one hand: https://lnkd.in/dA6-dJzH