KPIs & Targets

Foundry and Forging KPIs: Benchmark Ranges and How to Improve Them

Target ranges for the KPIs that decide foundry and forging profitability, how to measure them, and the specific levers that move each toward world-class.

Casting yield is the headline KPI because every point of yield is metal you melt but do not ship. Typical jobbing foundries run 55 to 65 percent, competent production foundries 65 to 75 percent, and world-class ductile iron work reaches 80 to 85 percent on favorable geometry. Measure it per part family, not plant-wide, since a heavy hydraulic block and a thin bracket behave differently. The lever is gating and riser design: switching from natural to applied risering, or adding exothermic sleeves, can lift yield 5 to 10 points by shrinking riser volume while still feeding solidification.

Melt loss is your metal-in-versus-metal-out efficiency in the furnace. Clean induction melting of steel and iron should hold 2 to 4 percent, and crossing 5 percent signals oxidized or dirty charge, excess turnover time, or poor slag practice. Aluminum runs higher at 3 to 8 percent because of dross formation. Track it per heat by weighing charge and tapped metal. Levers include charging clean, dry scrap, minimizing hold time, covering aluminum melts with flux, and reducing the number of remelts, since each pass through the furnace burns another 2 to 4 percent of the metal.

Casting scrap rate separates profitable foundries from struggling ones. World-class sits at 2 to 4 percent, typical operations run 5 to 8 percent, and anything above 10 percent means gas, shrinkage, or sand defects are out of control. Measure it as scrapped weight over total good-plus-scrap weight, and Pareto the defect codes weekly. The biggest levers are metal cleanliness, mold hardness consistency, and pouring temperature control within plus or minus 10 degrees C. Cutting scrap from 8 to 4 percent on a plant shipping 5,000 tonnes a year recovers roughly 200 tonnes of saleable castings.

Sand-to-metal ratio and sand system stability drive green sand quality and downstream defects. A tight ratio of 5 : 1 to 7 : 1 by weight indicates efficient sand use, while ratios climbing past 9 : 1 flag over-mulling or oversized flasks. Compactability should hold 38 to 45 percent and moisture 3.0 to 3.5 percent with low variance. Measure per shift with a compactability tester, not by feel. The lever is closed-loop sand control tied to return sand temperature below 50 degrees C, since hot sand throws moisture and compactability out of band and spikes blow and scab defects.

Die life is the forging equivalent of tool efficiency and it sets tooling cost per part. Hammer dies for carbon steel typically deliver 8,000 to 20,000 hits before resink, precision press dies 10,000 to 30,000, and hot work tool steel like H13 at 48 to 52 HRC extends the upper end. Measure hits to first resink and total hits to scrap separately. Levers that add 20 to 40 percent life include better die lubrication and cooling, tighter billet temperature control, and nitriding the die surface. The Die Life Estimator sets the target hit count you benchmark actual dies against.

Overall equipment effectiveness ties the melt deck, molding line, and finishing into one number. Foundry OEE commonly lands at 45 to 60 percent, and world-class continuous casting or high-pressure molding lines reach 75 to 85 percent. Break it into availability, performance, and quality, because a plant with 92 percent availability and 96 percent quality still lands at 60 percent OEE if performance drags at 68 percent from slow molds per hour. Match molding line rate to melt supply and shakeout throughput so no single station starves the others and drags the whole line down.

Energy per tonne is a cost KPI that also flags process drift. Induction iron melting should sit near 550 to 650 kWh per tonne of liquid; drifting toward 700 signals worn linings, poor charge density, or excess holding. Aluminum reverberatory melting runs 700 to 1,100 kWh per tonne. Meter each furnace and normalize to tapped weight. Levers include charging to full coverage before power-up, insulating ladles, and minimizing idle holding, which alone can waste 40 to 60 kWh per tonne on a furnace left energized between pours.

Improve in sequence, not all at once, and let each KPI verify the last. Stabilize the sand system and pouring temperature first, because they gate scrap; scrap then unlocks true yield, since you cannot trust a yield number computed on parts that get rejected; yield then sizes honest energy and melt-loss targets. Re-baseline every metric quarterly against the ranges above and track the trend, not the single reading. A foundry moving yield from 62 to 72 percent and scrap from 8 to 4 percent typically lifts saleable output 15 to 20 percent from the same melt tonnage with no new furnace capacity.

Published 2026-07-01.