Quantum Use Case #14: Port and container terminal scheduling

Short answer: Researchers and vendors have proposed quantum computing for container terminal problems such as premarshalling, which means rearranging containers to cut blockages. A D-Wave case story describes a project at the Port of Los Angeles and reports a doubling of equipment productivity, which is the vendor's claim. Academic work describes a hybrid quantum-classical method tested only in computational experiments.

The Problem Domain

Container port operations face significant challenges related to efficiency and resource allocation. The Port of Los Angeles is identified as the busiest port in the United States and the nation’s largest facility for handling shipborne cargo. In recent years, global supply chain challenges have made it essential for ports to operate at maximum capacity with the highest level of efficiency. The core logistical issue involves the transfer of cargo from ships to trucks for inland destinations. According to industry documentation, the Port of Los Angeles handled the equivalent of more than 10 million twenty-foot container units in 2021.

Inefficiencies in this process create a drain on manpower and financial resources. Extra minutes and hours squandered during cargo handling represent wasted capacity. Consequently, there is little room to compromise on efficiency when supply chains for essential goods are under strain. Quantum computing is proposed as a tool for identifying optimal solutions to these complex logistical problems, particularly where classical optimization methods may struggle with the combinatorial complexity of terminal operations.

Proposed Quantum Approaches

Research and industry initiatives have explored two primary avenues for integrating quantum computing into port scheduling: direct optimization of terminal operations and quantum-enhanced deep reinforcement learning.

The first approach involves using quantum computers to optimize terminal operations directly. Industry reports describe an initiative at the Port of Los Angeles, in conjunction with Fenix Marine Services and SavantX, that developed technology using AI and D-Wave’s quantum computer. This initiative aimed to optimize terminal operations by addressing complex logistical constraints. The vendor's case story reports a doubling of cargo handling equipment productivity and the production of more predictable cargo flows. This is the provider's own framing and has not been independently verified. The second approach is detailed in academic literature addressing the container premarshalling problem (CPMP). Premarshalling involves the rearrangement of containers to eliminate blockages and optimize crane movement. A study proposes a model that integrates quantum computing with an improved soft actor-critic (SAC) algorithm within a deep reinforcement learning (DRL) framework. In this hybrid architecture, variational quantum circuits (VQCs) are used to enhance the action selection process for the CPMP. The goal is to ensure container terminal efficiency and productivity by solving the optimization problems associated with container rearrangement and resource allocation.

Validation and Current Status

The status of these technologies varies between industry claims and academic validation. For the Port of Los Angeles initiative, the source material describes the technology as "deployed", which is the vendor's description. The vendor reports a significant increase in deliveries and equipment productivity. However, without independent third-party auditing of the operational metrics, these claims remain specific to the vendor's reporting.

In contrast, the academic approach using quantum-enhanced deep reinforcement learning is explicitly described as being validated through "computational experiments." The study notes that the hybrid quantum-classical algorithm was tested within a deep learning framework to address the challenges of the CPMP. This indicates that, at least for the reinforcement learning variant, the evidence is currently limited to simulation and theoretical validation rather than live operational deployment. Researchers have shown in simulation that variational quantum circuits can enhance action selection, but this does not imply that such systems are currently operational in commercial ports.

The distinction is important: while one source claims a deployed solution with specific productivity gains, another source frames the quantum integration as a research-stage model validated in computational experiments. The latter aligns with the typical trajectory of quantum machine learning research, where proofs of concept are established in simulation before any potential real-world integration.

Operational Considerations

Implementing quantum solutions for port scheduling requires addressing specific logistical constraints. These include container rearrangement, resource allocation, and crane movement optimization. The objective is to eliminate blockages that impede efficient cargo flow. In the context of the Port of Los Angeles, the initiative focused on Pier 3, aiming to reduce the time wasted in the cargo transfer process.

The integration of quantum computing with AI is central to both proposed approaches. In the industry case, AI and quantum computing were used in conjunction to optimize operations. In the academic case, quantum computing is integrated into a deep reinforcement learning framework, specifically enhancing the soft actor-critic algorithm. This hybrid quantum-classical approach is necessary because current quantum hardware may not yet handle the full scale of port optimization problems natively; thus, the quantum component is used to enhance specific subroutines, such as action selection in reinforcement learning.

FAQ

What problem does this use case address?

Container terminal efficiency, in particular cargo premarshalling, where containers are rearranged to remove blockages and speed up crane movement.

What does the Port of Los Angeles case story claim?

According to D-Wave's case story, a project with Fenix Marine Services and SavantX used AI and D-Wave's quantum computer, and reported a doubling of cargo handling equipment productivity. That is a vendor claim and we have not verified it independently.

What does the academic work show?

A study combines quantum computing with an improved soft actor-critic reinforcement learning algorithm for the container premarshalling problem. It was validated through computational experiments.

Is quantum port scheduling a proven technology?

No. Vendor material describes a deployment, while the academic work is research-stage. Treat the productivity figure as a claim, not an established result.

Why combine quantum computing with AI here?

Both approaches described in the sources pair quantum computing with AI methods to tackle the optimization.