Seaports explore AI, but intricate supply chains complicate implementation

At the Port of New Orleans, the arrival of a cargo ship loaded with massive power transformers, wind turbine components, and industrial generators sets off a complex logistical operation, with the challenge of transferring heavy, oversized equipment inland. Traditionally, moving such unwieldy cargo involved extensive manual coordination, static drawings, and weeks or even months of planning. Now, a new partnership between Port NOLA, the New Orleans Public Belt Railroad, and UTC Transoceanic is leveraging artificial intelligence to streamline and accelerate this process.

Launched in late May and currently in its deployment and implementation phase, the AI-driven solution uses digital twins and predictive modeling to map safe transport routes for heavy cargo. Instead of relying on manual inputs and spreadsheets, the application creates a digital model of the rail network, factoring in real-time data such as track shifts caused by temperature changes, according to the Federal Railroad Administration. Once operational, importers will simply enter their cargo’s dimensions and weight. The AI will instantly assess whether the shipment can safely traverse the intended route, flagging any issues such as bridge clearance or maximum weight limits and proposing alternative routes if necessary.

Kimberly Curth, Port NOLA’s press secretary, told Business Insider that the primary advantages of AI integration are “speed and certainty.” The digital twin system allows rapid and accurate assessments that previously required lengthy engineering studies and inter-agency coordination. This marks a significant shift in an industry characterized by slow adoption of new technology due to concerns over cybersecurity, operational risk, and the complexity of integrating cloud-based AI with predominantly on-premise IT systems.

U.S. ports have generally lagged behind other industries in embracing advanced technologies, often running software developed years before the rise of generative AI, according to Rene Alvarenga, vice president of products, AI, and execution visibility at Kaleris. Alvarenga noted that marine terminals tend to be “risk-averse” since any operational downtime for a software overhaul can disrupt critical supply chains. He estimates that marine terminals trail other sectors by about five years in AI adoption.

Despite these hurdles, a handful of U.S. ports are beginning to experiment with AI tools. The Port of Corpus Christi uses digital twins to track vessels, while the Port of Los Angeles has integrated AI into its truck appointment system. In Georgia, port authorities are rolling out AI-based facial recognition for truck drivers and automating gate check-ins with technology from EAIGLE. Still, the uptake remains limited, with most operators “quietly testing” AI solutions rather than adopting them broadly, said Lauren Beagen, founder and CEO of The Maritime Professor and a former Massachusetts Port Authority project manager.

Integrating AI into port logistics also faces challenges beyond technology, largely due to friction at the handoff points between ship operators, terminals, railroads, and trucking companies - each maintaining distinct systems and data protocols. Amir Hoss, CEO of EAIGLE, described these siloes as potentially a larger barrier than the technical limitations of AI itself.

Nonetheless, both industry executives and port officials see significant promise in combining AI-driven insights with human expertise, enabling faster, more precise decision-making for oversize shipments and day-to-day operations. As ports continue to experiment and invest, industry sources predict more widespread, practical use cases will emerge over the next few years, potentially transforming how the nation’s critical infrastructure moves complex cargo from docks to destinations inland.