Every day we make a choice — to recycle, to compost, or to simply discard an item. But regardless of what decision the consumer makes, corporations are one of the biggest contributors to environmental damage.
“Recycling and composting and all that stuff. It’s great, but it’s not where the real problem is, the real problem is the way we make things and move things around. It’s the factories, and supply chains. If you want to reduce waste in the world, that’s where you target.”
Reducing waste is an ambitious challenge to tackle, but it’s one Steve Pratt, Founder and CEO of Noodle.ai, is facing head-on. The strategy: use artificial intelligence and machine learning to help manufacturers and supply chain distributors limit the amount of waste they produce everyday. On this episode of IT Visionaries, Steve explains just how much waste is produced yearly by industrial operations and the monetary ramifications that waste brings. He also explains the way his team is using A.I. to help manufacturers predict hiccups with their machines, and why it’s time for users to overcome their fear of A.I. Enjoy this episode.
Main Takeaways
- It’s Not a Hardware Problem: The biggest problem facing industrial machines remains the software that runs them. In order to cut down on waste, the software must be consistently updated and optimized based on the working conditions.
- Data Library, Data Lake, Data Everywhere: There’s been an explosion of data, and yet corporations still can’t figure out how to put it all to good use.One way to use data is by extracting it from industrial machines, which can be in operation for decades, and then using it to design A.I. solutions based on variables a machine needs to run on.
- You Have to Test Your Models: When you are trying to build an effective modeling solution, you should consistently test your models to understand if your predictive models are working as they should. Make sure you take your time and test over and over again before implementing. A vast majority of A.I. solutions fail because they are not properly tested.
For a more in-depth look at this episode, check out the article below.
Article
Every day we make a choice — to recycle, to compost, or to simply discard an item. But regardless of what decision the consumer makes, corporations are still one of the biggest contributors to environmental damage.
“Recycling and composting and all that stuff. It’s great, but it’s not where the real problem is. The real problem is the way we make things and move things around. It’s the factories, and supply chains. If you want to reduce waste in the world, that’s where you target.”
Reducing waste is an ambitious challenge to tackle, but it’s one Steve Pratt, Founder and CEO of Noodle.ai, are facing head-on. The strategy: use artificial intelligence and machine learning to help manufacturers and supply chain distributors limit the amount of waste they produce everyday. On this episode of IT Visionaries, Steve explains just how much waste is produced yearly by distributors and the monetary ramifications that waste brings. He also explains the way his team is using A.I. to help manufacturers predict hiccups with their machines, and why it’s time for users to overcome their fear of A.I. Enjoy this episode.
One of the ways that his team is looking to accomplish this ambitious venture is through a process he refers to as flow operations, which uses advanced machine learning intelligence, combined with human intelligence to predict problems that might occur throughout the manufacturing process. The idea is that the data the machine generates will help the user and the software that the machine runs on predict when a company is making too much of a product. This will eliminate excess products, which according to Pratt is the biggest cause for waste in the world.
“It’s people and companies sending the wrong materials to the wrong place, at the wrong time,” Pratt said. “There is about an $861 billion problem in the manufacturing, the factories themselves where people or the factory has produced defective goods that are thrown out.”
One of the reasons for a lot of this waste is that the software that runs most industrial equipment has not been updated to account for the conditions they are operating in. This means that a machine that is running in the summer heat is running under the same conditions it operates on in the winter time. Or a machine producing a certain number of materials a day is never sped up or slowed down dependent on the product demand.
“What happens with high-performance computing and lots and lots of data is that you see that these things are not random and that there are patterns,” Pratt said. ”You can actually decode those patterns [and say] given this relative humidity outside and this weather and this raw set of raw materials, if we produce these products, they’re going to be defective, so we need to adjust them in the factory.”
Cutting down these processes is not as big of an obstacle as it may seem. Industrial machines are lined with sensors, which, according to Pratt, should open up millions of first- party data points. That will make setting A.I. barriers and guidelines in place much easier and will help the machines run to their peak performance.
“You’d be surprised how much data is out there and it’s just hidden,” Pratt said. “It’s just in data jail. Most of the world’s factories have sensors that are recording data on what happened in plant operations. There were previous waves of technology that came along, they collected your data and they put it in this dusty library that just sits there. We’ve developed some real breakthrough technology on the data side of extracting those data points to really be able to analyze it and run very sophisticated algorithms on those sensor data to figure out defects, to figure out when the factory is going to break down, so you can repair it ahead of time.”
To hear more about Pratt’s life journey to Noodle.ai and the unique ways they are using data and machine learning to cut down the world’s waste, check out the full episode of IT Visionaries!
To hear the entire discussion, tune into IT Visionaries here.
