From Pilots to Engines of Transformation

By Amy McLellan

AI is no longer a curiosity; it’s increasingly an engine of transformation. Whether it’s robotic welders at work in Korean shipyards or generative AI tools cutting coding times from ten days to two, the technology is starting to reshape how work in the project cargo and logistics space gets done.

This is, however, just the start of the journey, according to senior executives on an AI in Action: From Pilot to Performance main stage panel at Breakbulk Middle East.

Compared with finance and healthcare, logistics has lagged in AI adoption. Data gaps are often to blame, although certain applications — such as voyage planning and CO₂ emissions calculations — have delivered tangible benefits.

Marco Poisler, adjunct professor at the University of Houston Cullen College of Engineering and COO of global energy and capital projects at UTC Overseas, cautioned against overestimating what today’s systems can do. “What we see now is pattern recognition,” he said. “It's very impressive pattern recognition, but it’s not intelligence.”

The result is a lot of potential use cases and pilots but limited operational rollout at scale.

Panel moderator Dr. Sven Hermann, professor of logistics and supply chain management at Northern Business School, cited an MIT survey that revealed nearly 95% of AI pilot projects fail to reach full implementation. Panelists agreed with this finding.

“The awareness is definitely there, but very few, maybe 10%, go beyond theory into actual implementation,” said Ziad Abourizk, CEO of Logiswift. “One problem is everybody wants to do AI, but they don't know exactly what they are trying to solve with AI.”

Too often, he suggested, companies chase hype rather than return on investment. In part, this reflects the challenges of implementing AI in a way that’s workable and profitable.

“Our industry is a very complex industry, it’s not like generic transportation,” Abourizk said. “Before you deploy AI, you need to really understand the whole supply chain, all the stakeholders, whether you have the data you need, where the pain points are.”

That interconnectedness presents a major hurdle. “We have a lot of integration touchpoints,” said Gautham Krishnan, global category manager and SME for logistics at Flour. He pointed to the fragmented platforms used by different players, noting that without structured, high-quality data, AI cannot deliver meaningful insights.

Yet it is precisely in these complex, data-heavy environments that AI shows the most promise. Abourizk said Logiswift’s platform provides customers with real-time supply chain visibility, enabling them to shift from reactive responses to proactive decision-making. The company is now developing agentic AI for workflow orchestration and pricing.

“We have a long roadmap, and agentic AI is the next phase,” he said, adding that vertical integration across EPCs, forwarders, vessels, ports and trucking companies is essential to make such tools viable.

Health and safety is another promising area. Poisler highlighted work by carriers using “captain’s eye” software that analyses video streams to detect unsafe behavior, such as crew not wearing hard hats. “That’s one of the most exciting things,” he said. “We can use AI not to be a policeman, but to be a better coach on doing things safely.”

Ultimately, the panel agreed, the industry must focus on fundamentals: digitalization, data readiness and workforce skills. Corporate culture and legacy mindsets, meanwhile, also need to change.

“We’ve seen when it comes to any new technology, the learning cycles are always stretched because we are accustomed to doing things a certain way,” Krishnan said.

Given how quickly this technology can open capability gaps between AI haves and have-nots, this resistance needs to change, and fast. As Poisler put it: “The mantra now is that AI will not replace people, but it will replace you if you don't know how to use AI.”

Top photo (L-R): Sven Hermann, Gautham Krishnan, Marco Poisler, Ziad Abourizk. Credit: Ulysses&Crew

Second photo: Marco Poisler. Credit: Ulysses&Crew