The Hidden Scrap Problem on Metal Slitting Lines
We’ve spent years working with metals manufacturers across Europe and the Americas, and there’s a conversation we keep having. A plant manager walks us through their operation, clearly proud of what they’ve built. Good equipment. Experienced people. Then we ask: “What’s your scrap rate on the slitting line?”
The number is always worse than they think.
This isn’t a skills problem. Your planners are not bad at their jobs. Combining orders on a coil is genuinely one of the harder optimization problems in discrete manufacturing, and humans are not built to solve it by hand. Even simple-looking combinations involve hundreds of thousands of valid configurations. In testing, when we give planners five relatively straightforward order sets and ask them to find the best combinations, they get it wrong every single time. Not most of the time. Every time.
The good news is that this is a solvable problem.
Step 1: Filter and Group Compatible Orders (Data Preparation)
The bottleneck is rarely computing power. It’s getting planners to a manageable set of compatible orders before they start combining anything.
Intelligent grouping and filtering does the heavy lifting here. With the right tooling, a planner can screen for grade, temper, and thickness in a few clicks, collapsing hundreds of thousands of theoretical combinations down to a workable set of real ones. That’s not a small thing. The speed at which a planner can isolate compatible orders directly determines how much time they have to optimize, rather than just search.
Step 2: Simulate Yield Impact with Live Visual Layouts
Here’s what I see in most plants: planners work from spreadsheets or legacy ERP screens that give them no feedback on how a configuration will perform until after they’ve committed to it. They find out about the scrap at the end of the run.
Effective decision support does the opposite. A visual layout of how orders sit on the coil, live KPI feedback as the planner drags and rearranges, and the ability to undo a move that makes things worse. It sounds simple. It changes everything. When a planner can see the yield impact of a decision in real time, they stop relying on instinct and start working with data.
Step 3: Leverage What-If Scenario Planning for Custom KPIs
Most planning tools are tuned for a generic case. The moment your operation diverges from that case, which it will, the tool stops helping.
The question planners face is not just “what’s the tightest combination?” It’s “given our current demand, is overproduction preferable to a second setup? Which customer needs to be protected this week?” Those tradeoffs vary by week and by customer, and the optimizer needs to reflect them.
This is where what-if scenario planning earns its keep. Planners who can quickly run and compare two or three optimized scenarios make better decisions than planners who can’t. The ability to tweak the optimizer based on which KPIs matter most right now is not a nice feature. It’s the feature.
A practical note: watch out for tools that perform well on the demo case and fall apart when constraints change. The “100% fit” criterion matters. If the solution can’t capture your actual operating rules, you’re optimizing a simplified version of your problem, not the real one.
Step 4: Combine Optimization Solvers with Human Expertise
Optimization software does not know that your biggest customer will accept a two-day delay this week because their production schedule slipped. Your planner does.
This is not a concession to analog thinking. It’s a design principle. The best planning environments give planners an optimized starting point and then get out of the way so they can apply judgment where it matters. Which orders can absorb a small deviation? Where is the relationship too sensitive to risk? Those decisions belong to a human with context, not to a solver with none.
The goal is not to replace planners. It’s to give them better problems to solve.
Step 5: Integrate Order Combining with Shop-Floor MES & Scheduling
Optimal order combinations are only half the equation. If the routing and scheduling system can’t execute them, the planning work doesn’t translate into results.
An integrated approach connects order combining, route generation, and schedule creation in a single system. The practical value shows up when something goes wrong. If a coil comes off the line substandard, a planner with real-time MES feedback can reroute the affected order before the delivery date is missed. Without that integration, by the time anyone knows there’s a problem, the options are gone.
Disruptions are not exceptional events in metals manufacturing. They’re routine. A planning environment that can’t re-optimize around them in real time is a planning environment that’s always running behind.
Self-Diagnostic: Is Your Slitting Strategy Built to Minimize Scrap?
Take a look at these questions to see where your strategy stands.
| Your planners can isolate and filter compatible orders (by grade, tempet or thickness) in just a few clicks before starting any combination. | Your optimization software allows what-if scenario plannign to easily adjust to weekly constraint changes and unique customer priorities. |
| Planners have a real-time visual layout of coils and live KPI feedback to see the scrap impact of a decision before committing to production. | Your system provides an optimized starting point, but leaves room for planners to apply human judgment and context where it matters most. |
| Your planning environment can automatically re-optimize schedules in real-time when disruptions or standard coils occur on the factory floor. | Order combining, route generation and scheduling are seamlessly connected in a single system with real-time MES feedback from the floor. |
If you answered “yes” to the majority, then your strategy is fully equipped to radically reduce scrap, protect your margins and deliver elite performance.
If you answered “no” to the majority, then it’s time to move past spreadsheets. Your planners need better tools to stop relying on instinct and start optimizing with data.
Overcoming Inertia: Moving from Spreadsheets to Real-Time Optimization
None of this is theoretical. The manufacturers who have done this work have the scrap figures to show for it. The methodology is repeatable and the results are consistent.
The harder question is why more operations haven’t moved. Part of it is inertia. Part of it is skepticism that a software change can really move the needle that much. And part of it is that scrap on the slitting line doesn’t show up as a line item anywhere obvious. It shows up in yield, in rejects, in the margin that’s slightly lower than it should be.
Get started by contacting us for more information.
DELMIA, a Dassault Systèmes brand, connects the virtual and real worlds to drive innovation and sustainability. Powered by the 3DEXPERIENCE platform, our end-to-end solutions integrate virtual twins, industrial AI and augmented reality to optimize manufacturing, supply chains and workforces. We empower businesses to reduce waste and achieve sustainable, customer-focused operations, building a more resilient future.
Prashanth MysoreSenior Director for Strategic Business Development, DELMIA