When Richard Lebovitz looks at the way most companies manage their supply chains, he sees a $2 trillion problem. That’s the amount of capital he estimates is tied up in excess inventory, shortages, and other inefficiencies — all because supply chain teams are stuck firefighting while executives chase working capital.
“They think they’re aligned, but they’re not,” Lebovitz explained on a recent episode of the Manufacturing Executive podcast. “And then all of a sudden you got the supply chain guys trying to prevent shortages, and you got finance trying to reduce inventory. And it’s not synced up.”
Lebovitz is the founder and chief strategy officer of LeanDNA, an intelligent supply chain execution platform that uses AI and advanced analytics to help manufacturers optimize their operations. With more than 30 years of experience in the field, he’s seen firsthand how those competing priorities can pull teams in opposite directions — and how the right technology can bring them back into alignment.
Part of my passion throughout my career has been seeing this need to make manufacturing work easier and simpler. And that was a lot of the philosophy behind lean.
The key, Lebovitz argues, is to run a supply chain the way a good marketer runs a campaign: short, targeted, and measured. That means using AI not to make decisions, but to surface insights and prescribe actions that human workers can then execute on.
It’s a concept Lebovitz calls “prescriptive analytics,” and it’s a far cry from the predictive analytics that have become table stakes in the manufacturing world. While predictive analytics are all about crunching data to forecast what might happen, prescriptive analytics take things a step further by suggesting what to do about it.
“What we do is we now log all that data,” Lebovitz explained. “So if you think about all the analysis for a part, it could be pages and pages of information. And where we’re seeing a great way to start with AI is helping you understand things, helping you explain things, and taking very complicated data with different patterns and supply chain and making it to where the actions I need to take are clear and understandable.”
The goal, in other words, is not to replace human workers but to empower them — to shift their focus from reactive firefighting to proactive problem-solving. And that, Lebovitz believes, is where the real opportunity lies for manufacturers looking to gain a competitive edge.
If you go back to my earlier comment about what I first started in manufacturing and lean and working with the Japanese, it was all about just simple terms and it’s all about making work easy for people. And a lot of where we think AI can really be effective is how do you make work easy for people?
Of course, getting to that point is easier said than done. As Lebovitz is quick to point out, the AI landscape is changing at breakneck speed — and the risks of getting it wrong are high.
“The mistake we’ve seen is when you try to jump and start using AI to actually make decisions,” he cautioned. “If you jump in there and you have AI making those decisions, that’s where I think people have a problem, right? Because they think that it’s right to get all this positive feedback, and it turns out to either be limited or doesn’t have enough data, and it’s hard to detect that.”
Instead, Lebovitz advises manufacturers to start by using AI to drive the underlying math and algorithms behind their supply chain decisions, while leaving the ultimate call to human experts. From there, they can gradually layer in more advanced capabilities like natural language processing and explainable AI to help workers understand the logic behind those recommendations.
It’s an approach that requires a hefty dose of patience and humility — two qualities that aren’t always in ample supply in the boardroom. But for those manufacturers willing to put in the work, Lebovitz believes the payoff could be transformative.
“To me, it’s probably one of the most exciting times for manufacturing and supply chain,” he said. “When I look ahead the next several years, I think there’s going to be this kind of renaissance or revolution in how these technologies get applied to manufacturing.”
In other words? The $2 trillion problem may not be solved overnight. But with the right mix of human ingenuity and machine intelligence, it might just be solvable after all.
To learn more about how LeanDNA is using AI to help manufacturers optimize their supply chains, visit LeanDNA.com. You can also connect with Richard Lebovitz on LinkedIn or hear his full conversation with Joe Sullivan on the Manufacturing Executive podcast.
