Receive and read
Ingest RFQ emails and attachments, extract requirements, compute geometry from STEP or DXF when possible, and surface ambiguities rather than guessing.
/01 · Executive summary
The spreadsheet is not the problem. Hidden versioning and disconnected learning are.
Situation
Dajcor quoting combines RFQ interpretation, profile geometry, reference tables, spreadsheet formulas, and estimator judgement. The supplied files show the process clearly, but the knowledge is split across copied workbooks and people.
Risk in the status quo
Cost factors drift across template copies, overrides are hard to audit, and actual press performance rarely improves the next quote through a structured feedback loop.
What Webisoft proposes
Preserve the deterministic cost model, automate intake and geometry where confidence is high, place explicit review gates around judgement, and connect quoted assumptions to production actuals.
Start with evidence, then expand.
Validate the model and workflow in a proof of concept, run a Tier 1 and Tier 2 pilot in shadow mode, then scope the production system from measured results.
/02 · What we learned
Most of the quote is deterministic. The valuable judgement is concentrated.
The supplied sample set separates naturally into six input classes: geometry, customer requirements, reference tables, cost factors, process planning, and pricing decisions. Geometry, tables, formulas, recovery, and output generation can be tested. Throughput, exceptions, and final price still need Dajcor judgement.
10quote folders reviewed
13worksheets mapped
6cost-factor generations found
123/123current-template cells reproduced
Source: supplied Dajcor quoting automation working document, dated October 16, 2026.
- One quote record carries every value with its origin: computed, extracted, predicted, typed, or approved.
- The cost engine remains deterministic and can reproduce a historical quote from versioned inputs.
- Low-confidence geometry, process-plan suggestions, and pricing exceptions stop at explicit human gates.
- Production outcomes connect back to the quoted die, press, recovery, and throughput assumptions.
/03 · Recommended approach
Automate the repeatable work. Make judgement visible.
The proposal follows the seven pipelines documented in the supplied blueprint, organized into a buyer-friendly workflow that estimators can review from intake through output.
Plan and calculate
Run the tested cost engine, suggest press and throughput from historical evidence, calculate recovery, and show the nearest comparable runs beside the recommendation.
Approve, send, learn
Route exceptions, set the commercial price, generate outputs, and compare quoted assumptions with production actuals so each new quote starts from better evidence.
Design principles
- No black-box price prediction
- Versioned cost factors and reference tables
- Source and confidence visible on every important input
- Shadow mode before customer-facing automation
Not assumed in this proposal
- Autonomous pricing for complex programs
- Unreviewed OCR from scans or photographs
- ERP field mapping before M2M access is confirmed
- Production rollout before proof-of-concept gates pass
/04 · Interactive quoting prototype
Use the sample quote, then change the assumptions.
The workspace below rebuilds the strongest complete example in the supplied set: Sun North part 67740000. It uses the sample's real quote inputs, the documented 9 inch recovery method, the supplied cost factors, and the worksheet result as the baseline. Change length, weight, throughput, cavities, quantity, or price to see the quote recalculate.
Dajcor QuotingInteractive mockup · 67740000
Sun North · SN727X1031
StageReceive
Weight1.0954 kg/m
Recovery84.7%
Pieces / hr420.1
Unit cost$41.84
Price$46.00
Net profit7.0%
Stage 1 · Receive
One RFQ record, every source attached.
The supplied email asks for SN727X1031 at a minimum order of 50 pieces and says annual demand is about 300 pieces. The customer print provides the finished length and weight per foot.
Files received
MSGFwd: Pricing for other productRFQ context
PDFSN727X1030 SN727X1031 drawingsCompetitor print
PDFSN727X1031 model sheetProfile data
Routing
Tier 2 review
Print-derived geometry and a customer order below the standard pack require confirmation.
Needs review
Customer
Sun North
Source: supplied RFQ email.
Customer part
SN727X1031
Source: supplied RFQ email.
Requested quantity
50 pieces
About 300 pieces per year.
Stage 2 · Read
Every important value carries its origin.
The system separates customer facts, worksheet defaults, calculated values, and missing data so the estimator can confirm the profile before cost calculation.
Alloy and temper
6063-T5
Worksheet default
Finish
Mill
Worksheet default
Finished length
3,683 mm
RFQ
Weight per metre
1.0954 kg/m
Converted from 0.736 lb/ft on the supplied customer print.
Print derived
Geometry route
CAD not supplied
Request STEP or DXF to establish area and perimeter. The print-derived weight can proceed only after the weight check.
Weight check · Human gate
Confirm the current weight per metre or edit it in the Plan stage. The cost model updates immediately.
Stage 3 · Plan
Change the physical and process assumptions.
The recovery calculation follows the supplied 9 inch runout method across billet lengths from 21 to 42 inches, using the best considered result and the documented safety deduction.
Custom inputs
Supplied sample: 3,683 mm
Print-derived sample value
Observed 9 inch sample range: 1,450 to 2,200
Supplied sample: 2
Calculated process plan
Best billet
31 in
Computed
Recovery
84.7%
Computed
Finished weight
4.034 kg
Computed
Pieces per hour
420.1
Computed
0%Recovery100%
Plan check · Human gate
Confirm 2 cavities and 2,000 kg/hr before pricing.
Stage 4 · Price
See the cost stack, then make the pricing decision.
The cost engine recalculates recovery-sensitive metal loss, press cost, pack, freight, die amortization, fixed cost per order, commission, and net profit.
Cost per piece
Commercial inputs
Fixed order cost is spread over quantity
Supplied sample: $46.00
Net profit
7.0%
-15%Target 7% to 12%30%
Review flags
Stage 5 · Send
One approved record produces every output.
The prototype keeps the customer quote, engineering data, ERP fields, and audit trail synchronized. Outputs remain illustrative until Dajcor supplies its customer-facing quote format and M2M field mapping.
Generated outputs
PDF
Customer quote documentPrice, quantity, die charge, commercial terms
Format neededENG
Die drawing title blockWeight, area, perimeter, press, cavities
Partially readyERP
M2M quote recordCustomer, part, quantity, price, die and approval data
Mapping neededJSON
Audit recordInputs, provenance, overrides, cost factors, approvals
ReadyCurrent quote
$46.00 / piece
248 pieces, 84.7% recovery, 7.0% net profit.
Approval required
Review the active flags before customer output.
Data and formulas: supplied Dajcor quote simulator, quoting blueprint, pipeline document, and Sun North sample 67740000. This implementation was independently rebuilt for the proposal and is an interactive mockup, not a production quoting system.
/05 · Proposed scope
Validate the full path on real Dajcor quotes.
| Workstream | Deliverable | Accepted when |
|---|---|---|
| Intake and requirements | RFQ record with extracted fields, provenance, variants, and open questions | Dajcor reviewers can trace each important value back to its email, drawing, CAD file, default, or human entry |
| Geometry | STEP and DXF extraction path, plus controlled fallback for text prints and scans | Clean CAD samples reproduce the profile measures used by the quoting model, while uncertain sources stop for confirmation |
| Cost engine | Versioned implementation of the worksheet calculations and cost factors | The agreed validation set reproduces the current worksheet outputs within the agreed numerical tolerance |
| Process-plan assistance | Suggested press, cavities, throughput range, die assumptions, and comparable historical runs | Recommendations show their evidence and remain editable before pricing |
| Review workflow | Tier routing, exception flags, PE-003 prefill, and approval gates | No customer output is generated while required questions, exceptions, or approvals remain open |
| Outputs and feedback | Prototype quote output, engineering fields, ERP mapping draft, and quoted-versus-actual record | Dajcor confirms the output shape and the keys that connect quote, die, order, and production data |
/06 · Roadmap
Each phase ends with a measurable decision.
Week of Oct 26
Readback and scope confirmation
Walk through the reverse-engineered model, CAD extraction, interactive quoting prototype, findings, and required data pull.
About 4 weeks
Proof of concept
Run intake, geometry, the cost engine, and process-plan assistance end to end on selected won quotes with production actuals.
About 8 to 10 weeks
Tier 1 and Tier 2 pilot
Operate beside the normal estimator process in shadow mode, measure acceptance of prefilled inputs, and compare suggested pricing with issued quotes.
In parallel
Close the feedback loop
Join quotes to production actuals by die and order, then calibrate throughput, recovery, and die-life assumptions.
Scoped from pilot
Production build
Complete the review UI, customer quote generation, M2M integration, both-site configuration, and operating controls based on pilot evidence.
All durations and success gates in this roadmap come from the supplied Dajcor working document and remain subject to Dajcor confirmation.
/07 · Evidence already established
The proposal starts from tested understanding, not a blank discovery exercise.
4/4STEP samples matched estimator geometry
2/2clean DXF samples matched estimator geometry
123/123compared cells matched on current templates
5observed paths to weight per metre
The remaining question is not whether the current logic can be represented. It is whether Dajcor's broader archive and production data support reliable assistance for the judgement inputs.
Source: supplied quoting automation pipeline document, including the sample inventory and validation summary.
/08 · Commercial path
Price the next phase only after its inputs are confirmed.
The supplied package does not state a Webisoft fee, payment schedule, or fixed production scope. Rather than invent terms, this proposal separates the engagement into decision-ready stages.
Current
Readback
Confirm the process map, prototype direction, data owners, success measures, and proof-of-concept boundary.
Next
Proof of concept
Issue a fixed milestone proposal after the data pull and validation set are confirmed.
Conditional
Pilot and production
Scope and price from proof-of-concept results, measured workflow performance, integration requirements, and site coverage.
/09 · Data needed
Three inputs unlock the proof of concept.
Customer-facing quote outputs for the supplied samplesNeeded to define output generation and compare internal pricing with what left the business.
Press production data for the corresponding diesNeeded to validate throughput and recovery, the judgement inputs with the greatest price impact.
Die master data with geometry, press, cavities, cost, and lifeNeeded to connect quoting, engineering, and production through a stable identity.
Priority order and rationale come from the supplied Dajcor data-request list.
Confirm the proof-of-concept boundary.
Review · correct · select data · receive priced SOW
The next working session should confirm the workflow, nominate the validation quotes and production records, settle the success measures, and identify the customer quote format and M2M fields. Webisoft can then return a fixed milestone SOW grounded in those decisions.