Part rationalization
Reducing the number of part references a company buys by finding the ones that are the same, or close enough, and keeping one.
Also called: Part rationalisation, Part harmonization, Spec rationalization, Reference reduction, Part standardization
Most manufacturers buy the same part under several numbers. An M8×25 hex screw in zinc flake might exist as 14 references across three plants: different suppliers, different descriptions, one with a drawing and thirteen without. Each one has its own price, its own stock and its own minimum order.
Rationalization means finding those groups and collapsing them. Fourteen references become two (because one plant really does need stainless). The volumes add up on fewer lines, the price per piece drops, and the warehouse holds fewer bins.
Why it is hard
Part names do not help. “Vis H M8x25 8.8 Zn”, “HEX BOLT M8 X 25 GALV” and “screw, hexagonal, 8x25” are the same part to an engineer and three unrelated strings to an ERP. Matching has to happen on what the part is: its dimensions, material, grade, finish and tolerances, usually read off drawings or spec sheets.
The second difficulty is not technical. Every reference has an owner who had a reason to create it, and switching costs something: a validation, a test, sometimes a customer approval. A rationalization programme that ignores this produces a nice spreadsheet and no change.
What it is worth
On fasteners and standard parts, consolidating volumes typically brings 10 to 25% on the affected lines. The less visible savings are often bigger: fewer references to qualify, stock and audit, and fewer line stoppages because the one variant in stock was not the one on the work order.