Skip to content

Adding A.I. to Supply Chain Management

By M. Lapham Contributing Editor
manufacturing-worker_768x432.jpg
Verusen's A.I. system incorporates data from various sources, including "tribal knowledge" from long-time employees to help companies better manage supply chains.

The bare shelves left in the wake of the pandemic made everyone aware of how fragile a supply chain can be, and how far the effects the disruption it caused can travel. But a new self-learning software promises to prevent this from happening again.

Verusen, an Atlanta-based supply chain materials intelligence provider founded in 2015, has developed an artificial intelligence (A.I.) system that the company says can identify and provide the tools to lower risk, increase inventory visibility, and optimize capital.

Customers are seeing $5 million to $50 million of working capital savings, averaging a nearly 345 percent return on investment, according to Verusen. This is achieved by reducing material redundancy and operational inefficiency, which reduces unneeded spending.

The A.I. takes a large collection of data, including standard information such as how many materials a business uses, geography, and company history. Then the information is converted into a usable language and a model is created.

“It is about building on better people and technology for more accurate and accessible solutions,” said Paul Noble, Verusen’s founder and CEO.

He compares Verusen’s A.I. to “playing the brain.” The system takes in the information, analyzes it, and presents it in a way that can be used practically as it learns.

Improved Visibility

One of the most common problems manufacturers have had in the past is the lack of access of information from one step to the next. Each location in the process, from the plant to the warehouse to processing—and all stops in between—has valuable information, which is not readily available to the various points along the way.

Verusen’s system can be accessed by anyone who needs to see any part of the supply chain, whether it is once a quarter or multiple times a day. This allows companies to quickly address potential supply chain problems, even before they happen.

Many competitors aim to improve productivity and profits through hard data and analytical systems. Where Verusen differs is it uses the same information, then moves beyond the numbers. This could include tapping into an employee who always seems to know the problem before it happens, because he has been working at the plant for 20 years. His experience is valuable input to Verusen.

Information from other clients also can be useful. That may sound like a red flag, but Verusen draws a hard line in what’s used, according to Noble.

Anheuser-Busch, one of its clients, has more than 150 years of experience. The secrets to that longevity and how the company successfully runs its business are never shared with other customers.

However, industry trends and problems that any food and beverage manufacturer might run into are a part of the equation.

Noble, who first saw the need for better communication and tracking in supply chains while working at Sherwin Williams, noted that the trend is accelerating. Verusen has conducted annual surveys the past three years with global supply chain leaders. This year’s study included responses from about 100 businesses, with annual revenues ranging from $1 billion to more than $20 billion. Key findings include:

  • Reducing supply chain risk is the primary concern among manufacturers
  • More than half of the respondents said they need replacement parts weekly or monthly to prevent manufacturing outages
  • 80 percent cannot track their supply chain digitally
  • 43 percent cite lack of inventory visibility as a primary problem in sharing information
  • Three-fourths believe it would take 1-2 years to set up A.I.-driven materials management

The misconception with set-up time plays into something Noble constantly finds himself fighting against. People are conditioned to maintain the status quo and continue to do things the same way. The more time it takes to set up a new system, the more unlikely they are to try it.

Just in A.I. Time

It takes 90 days to fully implement Verusen’s software, Noble said. When the pandemic started, just-in-time (JIT) manufacturing strategies caused severe supply shortages. Some manufacturers started to question the practice, but as things returned to normal so did JIT.

Launching its A.I. platform just before COVID hit gave Verusen a unique look at the shifting relationship and attitudes toward JIT. Noble believes Verusen’s system could find a balance between the two.

“On the whole, people grew to understand the need to take action,” he explained. An integrated approach allows the cost benefits of just-in-time, while providing the inventory cushioning it lacks, which is an advantage of constant analysis and detailed knowledge from across the chain.

“This is what I have been evangelizing,” he said, adding, “not going back.”

Verusen’s initial customers are mostly larger and mid-size manufacturers ($500 million to $1 billion in sales) that operate five or more plants.

The A.I. system starts at $10,000 a month, but varies based on complexity.

Time is money. Knowledge is power. These are cliches for a reason. Accessing those truths is the key to success.

Datanomix, Vallen Partner on Real-Time Factory Analytics

Under a new agreement, Belmont, N.C.-based Vallen Distribution Inc. will offer Datanomix’s software solution to industrial customers as part of their Industry 4.0 and industrial automation initiatives. Datanomix, Nashua, N.H., claims it provides the industry’s only automated production intelligence software platform.

“We are excited to welcome Vallen into the Datanomix partners program,” said John Joseph, CEO of Datanomix. “Vallen is a respected brand in the industrial market. They can now have conversations about adding a new layer of value by introducing our LIVE production intelligence software to the portfolio of solutions.”

As part of the reseller program, Datanomix will train Vallen’s metalworking customer-facing team on selling, installing, and supporting its customers.

“Several machine monitoring companies are selling basic utilization services to manufacturers today. We carefully evaluated the contenders and selected Datanomix for its true real-time job insights and ability to translate job performance to business impact,” said Vallen CEO Chuck Delph. “The information presented by Datanomix accelerates time to information, shortens corrective action cycles, and directly impacts decision making at exactly the right time.”

The Datanomix system automates the collection and analysis of manufacturing data and delivers deep insights into production performance in real time and over time, according to the supplier, which added that its “no-operator-input” capabilities eliminate cumbersome analysis and data crunching.

ABB Gets Real with Cloud Collaboration

ABB Group has enhanced its RobotStudio robot programming and simulation software with cloud-enabled functionality. The new RobotStudio Cloud enables individuals and teams to collaborate in real-time on robot cell designs from anywhere in the world and on any device, according to the company.

ABB-RS-Cloud.jpg
ABB's enhanced RobotStudio software enables greater collaboration.

New features such as automatic version control are designed to increase transparency and productivity across teams. Other purported benefits include simplified interfaces and intuitive navigation that allow users of all skill levels to work on robotic projects.

“Web-based tools like RobotStudio Cloud bring a new level of agility and flexibility to manufacturers in how they plan and design their robotic automation solutions,” said Marc Segura, president of ABB Robotics division. “Offering a simplified user experience, RobotStudio Cloud helps to increase collaboration and reduce complexity, enabling both novices and experts to push the boundaries of robotics programming.”

The enhanced cloud-based software can help companies program robots more collaboratively while cutting the time, cost, and disruption associated with physical testing and commissioning, ABB said. New features include version control, which allows users to keep track of changes and have full transparency of any amendments. With full knowledge of who edited the program and when, developers can cut the time needed to resolve errors and performance issues, the supplier noted.

Incorporating virtual controllers, developers have a rapid and powerful robot simulation tool to tune and test programs. By producing an exact digital twin, the controllers give developers confidence that—once installed in the real world—the robot will move as precisely as in the simulation, ABB said. This allows for the fine-tuning and optimization that can help minimize waste or problems when production begins.

The software’s simplified interface makes it easier to produce and change programs, enabling users with little engineering expertise to program robot applications rapidly with minimum effort. Meanwhile, enterprise-grade protection ensures high security and productivity.

Calculating the 'Science Behind the Solution'

COE Press Equipment, Sterling Heights, Mich., is introducing its “Science Behind the Solution” calculators. The software utilities are based on FEA data to prove to customers that proposed coil processing equipment can meet their processing requirements and optimizing various operations, according to the supplier.

COE-Calculator-on-Laptop.jpg
COE Press Equipment's new calculators aid the verification process.

“Since the introduction of dual- and triple-phase steels, we’ve seen customers sold equipment that was promised to process these materials, but unable to deliver,” explained COE President Reid Coe. “Our engineering department built these calculators to verify that our equipment can withstand the forces generated in any particular application so that companies can be confident in their investment.”

COE inputs customer parameters, such as the materials to be processed and required speeds, then a proprietary algorithm indicates the processing requirements of the equipment. COE has calculators that help determine the right processing equipment, as well as ones that can help determine optimal operating parameters.

For example, the COE Coil Calculator requires input of the coil ID, OD, and width. From there, COE calculates the coil weight, run time, and parts per coil, allowing customers to better plan production and quote new work. Fabricators considering an in-house blanking line can use the CTL Calculator to see if there is value to be gained investing in this piece of equipment.

COE is encouraging stampers to ask for actual data from any of their coil processing suppliers that can prove out the performance capability of any proposed equipment. Additionally, COE is offering to work with companies to evaluate other offerings, helping ensure they are getting the right equipment for their processing needs.

  • View All Articles
  • Connect With Us
    TwitterFacebookLinkedInYouTube

Always Stay Informed

Receive the latest manufacturing news and technical information by subscribing to our monthly and quarterly magazines, weekly and monthly eNewsletters, and podcast channel.