AI plans NYK's 100-ship car carrier fleet in 10 minutes, boosting utilization and cutting CO2

AI-Based Car Carrier Allocation System Goes Live at NYK

NYK, MTI, and Grid Co. have released an AI system that automates and optimizes car-carrier allocation. Full-scale use began at NYK in July.

NYK runs more than 100 car carriers, the largest fleet in this segment. Planning those ships has been a manual grind-hundreds of voyages, months out, with shifting constraints. The new system cuts through that load and produces a plan in about 10 minutes after evaluating millions of schedule options.

Why this matters

• General: Faster, clearer shipping plans reduce delays, improve service levels, and manage costs.
• IT & Development: A real example of AI-driven optimization at enterprise scale with complex constraints and changing data.
• Operations: Higher vessel utilization and transport efficiency, with plans that reflect repairs, port risks, and customer needs.

What the system considers

• Customer requirements and booking priorities
• Fleet status, including maintenance and repair schedules
• Port congestion risks and related uncertainties

These inputs feed an engine that scores plans against core KPIs: vessel utilization, transport efficiency, and transport costs.

Speed and scale

The AI evaluates millions of possible schedules months in advance and returns an optimal plan in roughly 10 minutes. This short feedback loop allows teams to test scenarios, align with commercial forecasts, and respond to disruptions without starting from scratch.

Climate and cost impact

The system factors in the use of next-generation fuel ships and carbon pricing. That supports practical emissions reduction while maintaining service reliability and cost control. For context on industry targets, see the IMO greenhouse gas strategy.

How this likely works under the hood

• Data model: Standardized inputs for demand, fleet availability, ports, and constraints.
• Optimization loop: Heuristic/solver-based search across schedule permutations with cost and service objectives.
• Scenario engine: Run "what-if" plans for demand shifts, port delays, and maintenance changes.
• Human-in-the-loop: Planners review, adjust constraints, and approve the final plan.

What teams can apply now

• Centralize booking, fleet, and port data with clear ownership and update cadence.
• Define hard constraints (safety, compliance, capacity) and soft goals (cost, service, utilization).
• Make KPIs explicit and measurable; tie them to plan scoring.
• Keep humans in control: require approval and audit trails for plan changes.
• Continuously backtest outcomes to refine weights and constraints.

Industry context

Automotive logistics depends on precise timing, port throughput, and reliable capacity. Automating plan creation lets teams focus on exceptions and negotiations, not spreadsheet wrangling. For background on NYK's auto transport business, see NYK Automobile Transport.

Want to skill up your team on AI for operations?

Explore practical training and tools for automation and applied AI: Complete AI Training - Automation.

Key takeaways

• AI-generated allocation plans in ~10 minutes replace weeks of manual iteration.
• Plans reflect real constraints: customer demand, fleet maintenance, and port risk.
• KPIs are optimized explicitly: utilization, efficiency, and cost.
• Integrating fuel choices and carbon pricing supports lower emissions.