Vehicle Routing and Scheduling Breakthrough: GPUs Are The Future Of Routing Optimization

For decades, routing technology has been built on a simple assumption: incremental hardware improvements will continue to deliver better optimization results. That assumption no longer holds. As fleet operations grow more complex and cost pressures intensify, traditional routing engines are struggling to keep pace. 

Now, a new generation of routing platforms is fundamentally changing what is possible. 

The Limits of Traditional Routing and Scheduling Technology

The promise of “more computing power equals better optimization” quietly stopped being true years ago. These traditional systems, which are rooted in CPU-based hardware (Central Processing Units), result in optimization engines that take too long to run, limit multiple scenario testing, and force planners to settle for “good enough” routes. 

Routing and scheduling problems have historically been computationally intensive, testing the limits of even the latest CPUs. Over time, operational complexity exploded. Modern fleet routing must account for a growing number of variables and constraints: driver hours, equipment types, service windows, customer requirements, and tight order cut-off times. Traditional CPU-based solvers, once sufficient, have long since hit a point of diminishing returns, while also being asked to solve for increasingly difficult routing problems. 

To compensate, shippers and fleet operators have responded by adding constraints, taking shortcuts, and instituting manual workarounds to ensure routes can be built and dispatched on time. Fixed routes, restrictive territories, and predefined stop sequences are commonly deployed. 

All of this is occurring in a cost environment that has seen the largest increases in overall costs in twenty years. In the U.S., average fleet cost per mile has increased 43% since 2020, reaching $2.34 in 2025 – and significantly higher for local and regional multi-stop networks. When optimization speed and quality cannot keep up with the growing demands of today’s fleets, significant dollars are spent compensating for the gaps. 

A New Approach: Next Generation Routing and Scheduling Technology Using Graphics Processing Units (GPUs)

The breakthrough comes from rethinking the computational approach entirely. Next generation routing platforms leverage advanced processing architectures that can evaluate thousands of routing scenarios simultaneously, solving complex vehicle routing problems in seconds rather than minutes or hours. 

Chora, developed by Opsi Systems, is the prime example of this new generation of advanced routing solutions. Built natively on GPUs, Opsi uses NVIDIA CUDA and modern optimization algorithms and computational architecture to deliver a fundamental shift in optimization performance. The net result is a platform that can solve even the most complex vehicle routing problems in seconds rather than minutes or hours. 

This is a major advancement that addresses a core limitation that has constrained fleet operations for years: the tradeoff between solution quality and speed. Rather than cutting corners and simplifying assumptions, leveraging the advances in GPUs can solve for all operational requirements in less time and with higher quality results. 

From Tactical Routing to Strategic Advantage

Next generation routing technology – that leverages the advancements in GPU hardware – doesn’t just improve tactical route planning, it reshapes how organizations think about network optimization altogether. 

With near-instant optimization, fleets can: 

• Revisit routing policies and assumptions more frequently 

• Run daily and weekly scenarios that were once considered “too complex” 

• Perform multi-equipment bulk and compartment-based routing 
• Evaluate effectiveness of territories versus a dynamic solution

This then opens the door to more advanced supply chain optimization use cases, including: 

• Directly integrating tactical routing into strategic network design 

• Dynamic fleet versus common carrier decisions 

• Equipment type optimization across large networks 
• Improved distribution center location and flow scenario analysis

In effect, tactical routing becomes part of strategic analysis, rather than a limited input based on a fixed point-in-time. 

The Future of Routing Optimization

For complex networks and fleets, next generation GPU-based routing technology represents a fundamental leap forward, not just an incremental improvement. It removes the long-standing tradeoff between optimization quality and speed-to-completion. With this approach, fleets can generate high-quality routes quickly enough to truly operate dynamically, including evaluating multiple scenarios as part of regular operations. 

Today, only a small number of companies offer this level of capability, but they have reached a limit to what the traditional approach can produce. As cost pressures rise and operational complexity continues to grow, the shift toward this new era of routing platforms is likely inevitable. 

For fleets willing to adapt their processes, policies, and planning workflows, advanced routing and scheduling optimization is not just a technology upgrade – it's a transformative lever that can create a sustained competitive advantage.  

About the Author 

Chris Doersen is a Principal of Client Engagement at JBF Consulting, bringing more than 20 years of experience designing, modeling, and implementing logistics solutions for complex transportation networks. He partners with global shippers to drive efficiency, optimize fleet and carrier operations, and enable technology adoption that delivers measurable impact.

Chris has deep expertise with platforms including Blue Yonder, Descartes, Llamasoft, and Appian, and has led large-scale network design and TMS implementations for leading manufacturers and distributors. Known for his analytical rigor and operational insight, Chris helps organizations turn transportation strategy into sustained results.