AI & Technology
AI handles the pattern-matching: instant quotes, automated DFM screening and supplier matching. Engineers make the judgment calls on complex geometry, tight tolerances and non-standard requirements. Every order gets both.
Four Places AI Works in Every Order
From upload to dispatch, AI and engineers work in sequence, each on the tasks they do best.
AI-Assisted Instant Quote
- Recognises pockets, bores, threads, undercuts and surface finish callouts
- Breaks down material, machining, finish, QC and logistics
- Accuracy improves with historical order data
Automated DFM Screening
- Tolerances tighter than the process can reliably hold
- Walls below safe minimums and deep pockets that limit tool access
- Internal radii too small for standard tooling, and finish conflicts
Supplier Routing
- Matches on process, material, tolerance class, quantity and lead time
- Routes around capacity constraints in real time
- An engineer reviews the match on high-stakes or non-standard orders
Real-Time Order Visibility
- Status by stage: material, machining, QC and dispatch
- Alerts when milestones complete or timelines shift
- Everything visible in your order dashboard
The Right Tool for Each Task
AI is fast and consistent across many similar decisions. Engineers are better at judgment in new situations, so the platform uses both.
Speed and Consistency
- Geometry parsing and feature recognition, thousands of times a day
- Screening against a library of known DFM failure patterns
- Cost modelling across materials, processes and quantities
- Matching standard parts to the right supplier in the right window
- Order tracking, stage updates and ETA calculations
Judgment and Sign-Off
- Novel geometries and edge cases outside the training patterns
- Tight-tolerance and precision work for photonics, quantum and defense
- DFM recommendations that depend on the application
- Non-standard materials, unusual processes and special supplier selection
- Quality sign-off and accountability for documentation
Every Order Improves the Next
Real order results, such as first-article pass rates, actual versus estimated cost and supplier on-time performance, feed back into the models.
What Improves with Each Order
- Quote accuracy: the gap between estimated and actual cost narrows with each order cohort
- DFM detection: new issue patterns are added from real first-article failures
- Supplier matching: on-time and quality data refines routing decisions
- 1Order placedAI quote generated, DFM screen run, supplier matched
- 2Engineer review and productionDFM issues raised, first-article outcomes recorded
- 3Outcome data capturedActual cost, lead time and quality compared with predictions
- 4Models updatedQuote accuracy, DFM detection and supplier matching improve
Illustration of the loop: each pass stands for one cohort of completed orders feeding back into the models.
Frequently Asked Questions
If your question is not here, an engineer can answer it directly.
How does AI-assisted quoting work?
When you upload a STEP or IGES file, the AI parses the geometry, recognising pockets, holes, threads, thin walls and surface finish requirements, and estimates cost from the process, material and complexity. Standard parts get an instant estimate, and complex or precision orders are flagged for engineer review within 24 hours.
What does the AI check during DFM screening?
Tolerances tighter than the process can reliably hold, walls below safe minimums, deep pockets that restrict tool access, internal radii too small for standard tooling, and finish specifications that conflict with the material or process. Each issue comes with a suggested fix before an engineer reviews it.
How does Factorem match orders to suppliers?
The platform uses order data (process, material, tolerance, quantity, lead time and part complexity) to find the best-fit manufacturer. It weighs each supplier's certified processes, material stock, current capacity and quality record on similar parts. Engineers review the match on non-standard or high-stakes orders before release.
Does AI replace the engineer review?
No. AI takes the pattern-matching work (geometry parsing, common DFM checks, cost estimates, supplier matching) so engineers can focus on judgment calls such as unusual geometry, tight tolerances, unusual material combinations and photonics or defense-grade work. Complex and precision orders always get an engineer review.
How does the platform get better over time?
Every order feeds back quote accuracy, DFM flags, first-article outcomes and supplier performance. Quote models improve as estimated and actual cost converge, DFM screening improves as real outcomes are added, and supplier matching improves as quality and lead time data builds up.
Is my design data secure?
Yes. Files are encrypted in transit and at rest, and access is limited to the engineers and suppliers on your order. Customer designs are not used to train shared AI models without explicit consent, and an NDA is available on request.
See the Platform in Action
Upload a part file for an AI-assisted quote, then decide whether you want an engineer review. There is no commitment to order.
