
Amazon and FedEx Trial Quantum Optimization for Last-Mile Logistics

August 21, 2024
In a move poised to redefine the future of urban delivery systems, Amazon and FedEx have jointly confirmed trials of quantum optimization software aimed at solving last-mile delivery challenges. The announcement, made on August 21, 2024, confirms that both logistics giants are now piloting Zapata AI’s Orquestra platform—a hybrid quantum-classical computing system—across major cities in North America.
The trials represent one of the most commercially significant deployments of quantum technology in operational logistics, not in a research setting but on the streets of working urban environments. The focus is clear: optimize last-mile delivery routes, package sorting, and delivery batching using quantum-enhanced decision-making, while reducing carbon emissions and increasing speed.
Quantum in the Wild: From Theory to Delivery Trucks
Quantum computing, long a topic of academic discussion and specialized enterprise experiments, is now entering the logistics mainstream. Unlike previous lab-bound quantum demonstrations, these Amazon-FedEx trials involve real-world constraints: traffic congestion, fluctuating package volumes, unpredictable weather, and customer delivery preferences.
Zapata AI’s Orquestra platform is running in hybrid mode, combining classical solvers with quantum-inspired variational algorithms. The aim is to rapidly solve problems such as:
Which packages should be grouped together?
What’s the most fuel-efficient and time-optimal route?
How should urban zones be divided for simultaneous dispatches?
“This is quantum not as hype, but as operational logic,” said Christopher Savoie, CEO of Zapata AI. “We’re showing that quantum-classical hybrids can make a dent in one of the toughest logistical challenges: last-mile delivery.”
Pilot Locations: North America as Testbed
The quantum trials are being conducted in four North American cities:
FedEx: Memphis (HQ and major logistics hub) and Toronto (urban density + international delivery patterns)
Amazon: Los Angeles (traffic-heavy metropolitan grid) and Seattle (Amazon’s hometown with advanced fulfillment infrastructure)
These cities were selected for their diverse delivery environments—ranging from high-density urban streets to sprawling suburban zones, and even international customs routes in the case of Toronto.
“Every city has its own delivery DNA,” noted Rachel Mendez, logistics innovation director at FedEx. “What works in Seattle won’t necessarily work in Memphis. Quantum helps us adapt at speed and scale.”
The Last-Mile Bottleneck: A Known Industry Pain Point
Last-mile delivery—the final leg of a package’s journey from distribution center to the recipient—is the most expensive and inefficient part of modern logistics. According to a 2023 McKinsey report, last-mile operations can account for up to 53% of total shipping costs, largely due to vehicle idle times, failed deliveries, and route redundancy.
Moreover, last-mile emissions make a disproportionate contribution to carbon footprints, especially in cities with high delivery density. Optimization efforts to date have relied on classical computing models that struggle with combinatorial explosion as package numbers rise.
Quantum algorithms offer a compelling alternative. By evaluating millions of route and load combinations simultaneously, they can find high-quality solutions in a fraction of the time required by traditional methods.
Inside the Orquestra Platform: How Hybrid Quantum Works
Zapata AI’s Orquestra platform is not a pure quantum system—it uses a hybrid approach that balances classical processing power with quantum-enhanced subroutines. These subroutines include:
Variational Quantum Eigensolvers (VQEs): repurposed to minimize route costs
Quantum Approximate Optimization Algorithm (QAOA): adapted for dynamic routing clusters
Quantum-inspired tensor networks: used for batching packages with shared delivery constraints
The system integrates with existing route management software and APIs, allowing Amazon and FedEx to test results in real-world delivery flows without overhauling infrastructure.
“We’re not replacing classical tools—we’re supercharging them,” said Dr. Huda Ramez, principal engineer at Zapata AI. “Quantum enhances speed, adaptability, and resource use.”
In many cases, quantum-enhanced solutions are benchmarked side-by-side against traditional algorithms, allowing for direct comparisons in key metrics like:
Total delivery time
Fuel consumption
Failed delivery rates
Algorithm runtime
Sustainability Goals and Carbon Footprint Reduction
Both Amazon and FedEx have made public commitments to sustainability, including net-zero carbon goals by 2040 and 2045, respectively. Last-mile delivery is a major hurdle to those goals.
FedEx’s trials in Memphis have already shown promising early-stage results. Internal estimates suggest a 12–16% reduction in vehicle idle time and a 9% reduction in fuel usage over a one-week period using quantum-enhanced routing.
In Amazon’s Seattle pilot, the company has reported an uptick in on-time delivery rates, especially in zones previously labeled “route-stressed” due to roadworks or variable weather.
“Every minute saved in last-mile logistics is also a drop in emissions,” said Karen Thompson, head of Amazon’s Climate Pledge division. “Quantum optimization aligns with our business needs and sustainability imperatives.”
Operational Learnings: What the Trials Reveal
Beyond sustainability and efficiency, the quantum trials are surfacing valuable operational insights:
Microclustering: Quantum algorithms excel at creating hyper-local delivery zones optimized for both time and customer density, improving multi-drop efficiency.
Real-time Adaptability: Orquestra has been used to re-optimize routes mid-shift based on updated package flows or blocked streets—something classical models struggle with in real-time.
Data Fusion: The hybrid system merges GPS, traffic feeds, package weights, customer preferences, and weather data to create rich input datasets for optimization.
“Quantum isn’t just faster—it’s smarter in how it adapts to noise and change,” said Savoie. “This is the kind of agility logistics has been waiting for.”
Enterprise Signal: Quantum Goes Commercial
Perhaps most significant is the signal this sends to the enterprise world: quantum technologies are moving out of the lab and into the operations dashboard. Zapata AI, which previously partnered with BMW and Andretti Autosport, is now clearly demonstrating quantum’s value in mission-critical commercial logistics.
The FedEx-Amazon partnership with Zapata is also being closely watched by:
UPS, which recently funded a quantum R&D hub at Georgia Tech
DHL, currently running quantum trials in its European automation labs
Maersk and Flexport, investigating quantum digital twins for global freight modeling
“This is a tipping point,” said Dr. Anita Quon, supply chain futurist at MIT. “Enterprises are no longer asking ‘if’ quantum helps—they’re asking ‘how soon can we scale it?’”
Challenges: Talent, Integration, and Scalability
Despite the early promise, scaling hybrid quantum systems for full deployment still faces barriers:
Talent Gap: Few logistics professionals are trained in quantum operations, requiring cross-functional teams of physicists, engineers, and supply chain specialists.
IT Integration: Legacy route management systems need adaptation layers to interact with quantum-enhanced solutions.
Hardware Access: While most hybrid platforms run on simulators or cloud-accessible quantum processors, latency and queue times remain non-trivial.
To address these, Amazon and FedEx are investing in in-house quantum literacy programs, while Zapata continues to build hardware-agnostic solutions compatible with providers like IBM Quantum, Rigetti, and IonQ.
The Road Ahead: From Trial to Transformation
According to insiders at both companies, if pilot results continue to outperform classical models, expanded deployment could begin as early as Q1 2025, including:
Same-day delivery optimization in dense cities like New York and Chicago
Peak-season routing adjustments during holiday demand spikes
Integration with warehouse robotics for synchronized outbound planning
Zapata is also reportedly in talks to extend its platform to drone and sidewalk robot deliveries, hinting at a broader transformation of urban mobility logistics powered by quantum engines.
Conclusion: Quantum’s Last-Mile Moment Has Arrived
What began as speculative research a decade ago is now guiding real-world vehicles through real-world streets. The Amazon-FedEx trials of Zapata’s quantum optimization tools mark a historic step in logistics innovation, with hybrid quantum systems proving their commercial viability in one of the industry’s most complex and costly problems: the last mile.
This isn't just a test of technology—it’s a test of scale, readiness, and competitive advantage. And if early signals hold, quantum will soon be delivering more than just potential—it’ll be delivering packages.
“Last-mile logistics is messy, fast-changing, and data-hungry,” said Savoie. “That’s exactly why quantum is the right tool for the job.”
