Manufacturers have spent countless hours refining how they plan and run production. ERP systems, advanced planning and scheduling tools, automation, analytics and AI have improved visibility and control. Yet many of the decisions at the heart of production — what to make, when to make it, in what sequence and with which resources — remain difficult to optimize as operations become more complex.
Production scheduling illustrates this challenge. A plan may need to account for customer demand, equipment and labor availability, materials, changeovers, sequencing requirements and delivery commitments, all while operating conditions continue to shift. Improving one objective can affect another, which means the practical goal is rarely to find a mathematically perfect plan. Manufacturers need a high-quality answer fast enough to keep production moving and respond when reality diverges from the plan.
For years, computational limitations made certain compromises unavoidable. Manufacturers learned to accept schedules that kept production moving, even when those schedules left meaningful efficiencies unrealized. But advances in optimization technology are changing what can be solved within an operationally useful timeframe. Quantum optimization is giving manufacturers a reason to reconsider whether yesterday’s definition of “good enough” still meets today’s competitive standards.
The compromises in a feasible plan
A production schedule may meet the day’s requirements and still carry substantial hidden costs. It may create operational inefficiencies, uneven workloads, idle equipment, late orders, or unused capacity. It may also leave so little flexibility that even a minor disruption requires extensive manual intervention when conditions change.
Complexity can also influence what gets included in the planning process. As more constraints and dependencies are added, solving the full problem can become too time-consuming, leading teams to simplify assumptions, divide larger problems into smaller ones or rely on heuristics and manual judgment. These are practical responses to computational limits, but over time they can shape what an organization believes is possible.
Hybrid-quantum optimization offers manufacturers another way to approach certain highly constrained problems.
The opportunity is not to replace existing manufacturing systems, but to reconsider areas where complexity has historically forced the business to make compromises. In practice, a hybrid-quantum optimization application can work alongside existing ERP, APS and planning systems, with implementation focused on modeling the target problem, connecting the necessary data and integrating the resulting decisions into established workflows.
Rethinking production sequencing at Ford Otosan
Ford Otosan, a joint venture between Ford Motor Company and Koç Holding in Turkey, faced a challenge in production sequencing for its highly configurable Ford Transit line, which includes more than 1,500 vehicle variants. The company worked with D-Wave to develop a hybrid-quantum optimization application for its body shop and address complex, interconnected requirements.
In a benchmark involving 1,000 vehicles and 15,000 constraints, the hybrid approach generated a production schedule in just over three minutes compared with 10 minutes for proprietary classical solvers and more than an hour for an open-source solver. Ford Otosan has since deployed the application in production.
The value of that improvement is not limited to generating a schedule faster. Greater speed and scalability can make it easier to incorporate more upstream and downstream constraints and respond faster when operating conditions change.
BASF, a global chemical company, explored a similar challenge in a different manufacturing environment. In a completed proof of concept, BASF and D-Wave used a quantum-hybrid application to optimize production scheduling and tank assignments at a liquid-filling facility. Compared with an industrial-grade solver, the application reduced projected scheduling time from 10 hours to just seconds. The project also reduced lateness by 14%, setup times by 9% and tank-unloading durations by 18%.
Better decisions can become a competitive capability
The competitive implications extend beyond how quickly a schedule can be generated. Manufacturers frequently face similar pressures around labor, equipment, materials, changing demand and customer commitments, but their ability to make decisions within those constraints can differ significantly.
If one manufacturer can incorporate more of the real operating environment into its planning, adjust schedules more readily when conditions change or find better ways to use existing capacity, those differences can accumulate across production cycles. What begins as an optimization improvement can translate into greater responsiveness and a more flexible manufacturing operation.
This is why some of the most interesting opportunities may be found in processes that already appear to work. A planning method developed around yesterday’s computational limits can remain embedded long after better approaches become available.
Manufacturing will always involve trade-offs and no optimization method eliminates the realities of finite labor, equipment, materials, or time. But as the tools available to manage those trade-offs evolve, manufacturers have a reason to revisit what they have learned to accept. A production plan that was good enough yesterday may no longer be the standard against which tomorrow’s competitors are operating.
To learn more, explore quantum optimization for manufacturing at dwavequantum.com.
