Oden Technologies, which develops industrial automation and analytics software and data collection hardware, in December announced it has incorporated new machine learning (ML) and artificial intelligence (AI) capabilities into its software.
For more than four years, Oden has been helping companies, including plastics processors, collect equipment data with the goal of reducing waste, increasing production and reducing machine failures. The newly updated software offers customers more sophisticated recommendations for improving efficiency and production, said Willem Sundblad, company co-founder and CEO. He said Oden’s platform is more sophisticated than traditional manufacturing execution system (MES) software.
“ML is a technology that allows users to get more intelligent insights and predictions out of our solution rather than traditional MES functionality,” Sundblad wrote in response to emailed questions about the new technology.
“An example use case: After having collected process data and environmental data (temperature, humidity and dew point in the factory), Oden’s machine learning algorithm can recommend new settings for the operators that increase both quality and output.”
ML allows a system to make predictions and recommendations for the future based on previous data, he said. MES software doesn’t have that type of capability and the architecture it is built on is not suited for it, either.
AI is often classified as a more advanced type of ML, sometimes called deep learning, Sundblad said.
Oden’s technology employs “a novel, patent-pending infrastructure spanning both the cloud and the edge,” the company said. That means some of the data is analyzed on remote servers in the cloud, while other data is analyzed on Oden hardware or devices physically installed on the factory floor.
This means that data collection and analysis are uninterrupted even in the event of an internet outage or other connectivity problem.
“The cloud is more cost- efficient and powerful when it comes to analysis, but it requires internet connectivity,” Sundblad said. “For mission-critical applications, you don’t want to be dependent on the internet. That’s why it’s important that it can work on the edge (on the factory floor) so that the system is always operational, regardless of network connectivity.”
Oden’s ML and AI framework is tailored for manufacturing processes, including extrusion and injection molding. It integrates ML algorithms and data-science tools with data collected from connected machines, operator inputs, work orders, environmental monitors and product specifications. The hybrid cloud and edge infrastructure “delivers the power of the cloud without compromising the requirements of mission-critical applications,” the company said.
The software and hardware continuously monitor the manufacturing process to predict product quality and machine health, detect abnormal behavior, and provide recommendations for improving recipes and process settings.
Manufacturers can use the Oden technology to rapidly prototype and test new ideas for process monitoring and optimization, and, more importantly, deploy them directly into their operations with little to no overhead.
Customers using Oden’s software and hardware have reported up to a 20 percent increase in monthly output and a 50 percent decrease in total scrap, resulting in millions of dollars in savings and additional revenue each year, according to the company.
Deepak Turaga, VP of data science for Oden, said he believes the software’s new ML and AI capabilities will significantly accelerate such gains.
Bruce Geiselman, senior staff reporter
Contact:
Oden Technologies Inc. New York, 800-230-9063,
Bruce Geiselman | Senior Staff Reporter
Senior Staff Reporter Bruce Geiselman covers extrusion, blow molding, additive manufacturing, automation and end markets including automotive and packaging. He also writes features, including In Other Words and Problem Solved, for Plastics Machinery & Manufacturing, Plastics Recycling and The Journal of Blow Molding. He has extensive experience in daily and magazine journalism.
