Today
Designers create new part numbers because finding an existing part takes longer than drawing it again. Every new part then costs effort in purchasing, stock and documentation for years.
An example
A designer at Ebert & Lorenz Maschinenbau needs a 120 × 80 × 6 mm bracket in S235 for the HP-250 press. demi finds part 300-4410: used in 14 machines, 300 in stock, with only one hole different (Ø 9 instead of Ø 11 mm).
Your data, wherever it lives
Part master data, stock, purchase prices and where-used lists sit today in the ERP, drawings and 3D models in CAD and PLM, older drawings as PDFs on drives, and every plant keeps its own numbers. demi reads, understands and links them per part across all plants, with your rights. The AI finds similar parts by shape, dimensions and material in seconds, with stock and usage, and engineering decides whether to reuse a part or create a new number.
Where the data lives today
- ERP
- CAD & drawings
- PLM
- PDFs
- Folders & drives
How demi solves it
Three layers in this order: connect the data first, then the rights, then the AI.
- 1 · ConnectData Layer
Data sources in this case: the PLM, ERP, CAD and file share. demi replaces nothing and joins part master data, drawings, attributes, where-used lists, stock and purchase prices into one record per part, across all plants.
- 2 · PermitPermission Layer
Designers see every released part; parts in development stay with their team. If you want, a new part number is issued only after the designer has confirmed the reuse check, and the reason is recorded.
- 3 · ActIntelligence Layer
The agent finds similar parts by attributes and drawing content and ranks them by stock, price and usage, each hit with its drawing and record. It compares dimensions, material and attributes value by value from the master data; the designer decides.
How it runs
- 1
Describe what you need: dimensions, material, function
- 2
demi searches across all plants, systems and drawings
- 3
Compare hits by stock, usage and differences
- 4
Reuse the existing part or create a new number with a reason
What changes
Fewer new part numbers
Less effort in purchasing and stock
Faster design with proven parts
What does it bring you?
Check it with your own numbers
Every starting value is an assumption you can change. The calculation sits right below, step by step.
Your numbers
Result
+180 hfreed per year
Assumptions and calculation
demi finds similar parts by geometry and attributes and shows which are proven; you decide.
30 min → 21 min per case
How we calculate
100 part searches × 30 min × 30% × 1.0 (manual) × 12 months = 10,800 min = 180 h
Value per year, before demi costs
Freed per year+180 h
Assumption: labour cost of €49.50 per hour in German manufacturing in 2025 (Destatis), plus workplace, IT and overheads.
= Value per year, before demi costs
+€11,700
gained per year
180 h ÷ 1,760 h per role = 0.1 full-time roles
An estimate, not a promise. The starting values are assumptions for an industrial company with 150 to 500 people.
Combine with more cases and get the packageQuestions about this use case
Does it search by drawing or only by attributes?
Both. demi uses the attributes in the master record and reads the drawings themselves, such as dimensions, material and title block. Every hit shows the exact differences, so nobody has to guess.
How does it handle parts from different plants?
Parts from every connected plant and system end up in the same data set, even under different numbers. You see where a part is made and where it is in stock. Who may see which plant follows your access rights.
Can we require a reuse check before every new part number?
Yes. You can set a rule that a new number is only created after the reuse check. The search, the hits and the reason for the new part are stored with it.
What if our master data has gaps?
demi fills missing attributes from drawings and data sheets and names the source. Searching by drawing content still finds candidates when values are missing from the master record.
