The difference between 95% and 97% yield on a slitting line does not sound like much. On a million-dollar annual throughput, it is $20,000. On a five-million-dollar throughput, it is $100,000. Most service centers do not track yield at the production-order level, so they never see the loss clearly enough to fix it.
Where Yield Loss Happens
Yield loss in steel processing comes from five sources. Each one is small. Together, they add up.
Edge trim. When a master coil is slit, the edges are trimmed to create clean widths. The trim varies by material gauge and condition, but a typical slit job loses 0.5" to 1.5" per side. On a 60" master coil slit into four 14" mults, the edge trim plus the space between mults (kerf) can consume 3% to 5% of the original width.
End crop. The leading and trailing ends of a coil often have gauge variation, edge damage, or weld marks (for joined coils). These ends are cropped and scrapped. Depending on the coil condition and customer specification, end crop can be 10 to 50 feet of material per coil.
Kerf. The physical width of the cut itself removes material. On a slitting line, kerf is minimal (the spacers between knives are thin). On a shear or plasma table, kerf can be 1/8" to 1/4" per cut. Multiple cuts across a sheet multiply the kerf loss.
Defect rejection. Material with surface defects (scratches, rust, scale, edge damage) below customer specification gets pulled during processing. The rejection rate varies by material source and grade, but even well-managed operations reject 0.5% to 2% of incoming material during processing.
Off-specification output. Material that is processed but does not meet the customer's tolerance requirements. A slit coil with a width tolerance of +/- 0.005" that comes off the line at +0.008" either gets reworked (consuming additional processing time) or reclassified as remnant stock and sold at a discount.
Why Tracking Matters
Most service centers know their overall scrap rate as a percentage of purchased material. Fewer track yield at the production-order level. The difference matters because aggregate data hides the variation.
A service center with a 96% overall yield might have one slitting line running at 98% and another at 93%. The 93% line is losing 5% more material on every job, but the average masks the problem. Without production-order-level tracking, the issue stays hidden.
Production-order-level tracking means recording the input weight and the total output weight (finished product plus remnants plus scrap) for every job. The difference between input and output is the yield loss. When that number is tracked for every job, patterns emerge.
Certain material grades might consistently yield lower (galvanized coil with edge damage from poor mill packaging). Certain operators might produce tighter edge trim (because they set up the slitting line with more care). Certain shifts might have higher reject rates (fatigue, lighting, temperature). These patterns are invisible without the data.
The Input/Output Accounting System
Effective yield tracking requires an input/output accounting system at the production order level. This is simpler than it sounds.
For every production order: record the input material (by heat number, weight, and dimensions). Record all outputs: finished product (weight and count), remnants (weight, dimensions, and new inventory record), and scrap (weight and category). The sum of outputs should equal the input, within measurement tolerance.
Any gap between input and output is unaccounted yield loss. If a 22,000-pound coil produces 20,500 pounds of finished product, 800 pounds of tagged remnants, and 400 pounds of weighed scrap, the numbers add up to 21,700 pounds. The missing 300 pounds (1.4% of input) is unaccounted loss. It might be floor sweepings, unweighed edge trim, or a measurement error. Regardless, it is visible, and visibility is the first step to improvement.
Turning Tracking Into Improvement
Once you have production-order-level yield data, improvement follows a straightforward path.
Benchmark by machine, operator, material type, and customer specification. Identify the combinations that produce the best yields and the ones that produce the worst. Focus improvement efforts on the largest gaps.
Set yield targets by job type. A straightforward slit of 16-gauge HRC from a domestic mill should yield differently than a precision slit of 26-gauge galvanized from an import source. One-size-fits-all yield targets are meaningless. Job-specific targets based on historical data create accountability.
Track scrap value separately. Not all scrap is equal. Edge trim from 304 stainless has significant recovery value. Edge trim from A36 carbon has minimal recovery value. Tracking scrap by material grade ensures that the scrap revenue credited back to operations reflects reality.
Review yield data weekly with the production team. Share the numbers. Celebrate improvements. Investigate declines. When the production team sees that their machine ran at 97.8% last week (above the 97% target) and the other machine ran at 95.2% (below target), competitive motivation kicks in.
Yield management is not glamorous work. It does not show up in marketing materials or trade show presentations. But for service centers processing $5 million or more annually, a 1% yield improvement drops $50,000 straight to the bottom line. That is a better return than most investments a service center owner can make.