Why Machine-Shop Quoting Tools Fall Short for Forging
Quoting add-ons built for CNC job shops assume cost comes from setup plus cycle time on a machine. A forging quote is driven by different things: the weight and grade of bar stock you must buy, how much of it ends up as flash and scale, how many pieces a die set will survive, how many parts you get per furnace heat, and whether the customer owns the tooling. A closed-die job might run on a 2,000 ton mechanical press while an open-die job runs on a hydraulic press with a manipulator, and the cost model for each is different.
A tool that cannot represent billet yield, die life and heat-treat batches will force estimators back to spreadsheets within a month.
- Material priced by weight and mill minimums, not by cubic inches of stock
- Die cost carried as tooling, with life in pieces and a rework or resink allowance
- Furnace and heat treat cost driven by load size, not part count alone
- Test and inspection charges that scale per lot or per heat, not per piece
The Cost Elements a Complete Forging Estimate Contains
Work through the estimate in the order the metal moves. Start with the finished forging weight from the forging drawing, add flash, connector and scale loss to get input weight, then price the bar or billet at current mill or service-center cost including cut-to-length charges. Next come heating and forging operations priced by press hour or hammer hour with the crew, followed by trim, any cold coining or straightening, heat treatment, cleaning such as shot blasting, and inspection or testing.
Add die amortization and any die maintenance, then outside processing, freight, packaging and the margin. Each element should be traceable to a rate or a past job, otherwise the estimator is just remembering numbers.
- Input weight and material cost with alloy surcharge and mill minimums
- Heating, forging and trim at press-hour rates with expected pieces per hour
- Heat treatment, cleaning, straightening and machining allowance operations
- Die amortization, testing, certification, packaging and freight
Where Estimators Lose Time and Margin
The slow part of a forging quote is rarely the arithmetic. It is finding the answer to questions the RFQ does not state: is this a repeat of a part forged two years ago, what did that job actually yield, did the dies need rework after the first thousand pieces, and did the heat treat vendor change price. When a shop runs JobBOSS, E2 Shop System, Epicor Kinetic or Infor SyteLine, the actual cost data exists, but pulling it means navigating job histories by hand.
Estimators fall back on the last quote, which was itself an estimate, so errors in yield or die life repeat from quote to quote until a bad job exposes them.
- Last quote reused instead of last actual cost and yield
- Scrap and die rework from past jobs not visible at quote time
- Outside processing prices stale by the time the quote is issued
Try it here, free
Turned Part Secondary Operations Cost Worksheet
Add each secondary operation with setup, per-piece price and vendor minimum charge to get the extended cost, cost per part and total lead time.
How AI Drafts the First Pass With a Human Approving
A private AI quoting assistant reads the RFQ package, extracts the grade, specification and quantity, and searches the ERP and document archive for similar forgings by material, weight range and process route. It pulls actual input weight, yield, press hours, scrap and die history from those jobs and drafts an estimate with the sources listed beside each number. The estimator reviews, changes assumptions and approves.
Nothing is written back to the ERP without that approval. This is where an assistant over your own data beats a generic chatbot: it can only cite jobs that actually exist in your history, and it can run on your own hardware if drawings carry ITAR or customer-confidential data.
- Extract grade, spec, quantity, delivery and testing requirements from the RFQ
- Retrieve similar past forgings with actual yield and press hours
- Draft the estimate with source references for each line
- Hold the draft for estimator approval before anything is saved
When Software or AI Is Not Worth It
If your shop quotes a few dozen jobs a year, most of them repeats, a well-built workbook with locked formulas and a log of actual-versus-estimated results may serve you better than any new tool. The same is true if your real problem is that actual costs are not captured at all: an AI assistant cannot retrieve yield or press hours that nobody recorded. Fix job costing first.
Quoting automation pays when RFQ volume is high enough that estimators are the bottleneck, when win rate depends on response time, and when your ERP already holds a few years of trustworthy job history to learn from.
Frequently asked questions
What should forging estimating software calculate?
At minimum: input weight from forged weight, flash and scale loss; material cost with mill minimums; heating and press time; trim; heat treatment; testing; die amortization; outside processing; and freight. It should also store the assumptions behind each quote so you can compare estimated against actual yield, press hours and die life after the job ships.
Can Excel be good enough for forging quotes?
Yes, for low volume and mostly repeat work. A controlled workbook with locked formulas, a rate table and a log comparing estimate to actual is a legitimate system. It breaks down when one person owns the logic, when quote volume grows, or when nobody checks estimates against what the jobs really cost.
Does AI quoting work for custom forgings?
It helps most when you have comparable past jobs. The assistant finds similar forgings by grade, weight and route and drafts from their actuals. For a genuinely new geometry or alloy, it still saves time on reading the RFQ and building the shell, but the estimator must set yield and die assumptions manually.
Send Netray a sample forging RFQ and a few past jobs, and see a sourced draft estimate built from your own history.