What is slitting optimization?

Slitting optimization is deciding how to cut wide metal coils into narrower strips so that as much material as possible ships to customers instead of going to scrap.

It sounds like a simple question. It is not.

Why it is hard

Say you have a 48-inch coil and an order book with a dozen open widths on it. You could run 6 inches eight times and use the coil perfectly. But you probably do not have eight orders at 6 inches. You have one at 5.9375, three at 11.25, one at 7.5, and a handful of others that do not nest cleanly against each other.

Now add the constraints that actually govern the decision:

• Which orders are due, and when
• What is physically in inventory versus what the system thinks is in inventory
• Slitter capabilities and knife setup limits
• Gauge and grade compatibility across orders
• ID and OD constraints
• Whether the job needs multiple passes
• Edge conditioning requirements
• PIW limits
• Which line is free, and for how long
• What is going on the truck Thursday

Every one of those narrows the field. Together they turn a simple-sounding question into a combinatorial problem with millions of valid answers, most of which are worse than the best one.

Why experience is not enough

Good planners develop instincts. They know which orders tend to nest well, which customers will accept a substitution, which combinations have burned them before. That knowledge is real and it is valuable.

But instinct finds a good answer, not the best one. The search space is too large for a person to walk. A planner working from experience might evaluate a few dozen arrangements before settling on one that works. The optimal arrangement might be the four-thousandth.

The gap between those two is scrap. It does not show up as a mistake, because nothing went wrong. The line ran, the orders shipped, the customer was happy. The margin just quietly went to the scrap bin instead of the bottom line.

That is what makes this problem expensive. It is invisible in the moment and only visible in aggregate, at the end of the year, in the yield number.

What optimization software does

An optimization engine evaluates far more combinations than a person can, and it does it against every constraint at once rather than sequentially.

LineSight's engine analyzes millions of configurations and returns the one that produces the most sellable material. Not the one that uses the coil most completely, which is a different and less useful target. The one that puts the most revenue on trucks, factoring in what is actually sold and what would sit in inventory.

Then it hands back a runnable work order, not a theoretical answer.

What it is worth

At a service center processing 100,000 tons a year at $1,000 a ton, a 5% scrap rate means roughly $5 million in material that never becomes product. Moving that rate down by even two or three percentage points is a seven-figure swing.

At Elite Steel, LineSight cut average scrap rate by 2.9 percentage points, which translated to roughly $850,000 in added annualized revenue and a 20% lift in gross profit margin.

Where the term comes from

Slitting optimization sits inside a broader class of problems called cutting stock problems, studied in operations research since the 1930s. The math is well understood. What has been missing in metals is software that applies it to the way a service center actually runs, with the real constraints of a real floor rather than a textbook version of them.