> ## Documentation Index
> Fetch the complete documentation index at: https://docs.garmentflow.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# POM Library

> The workspace's reusable catalog of measurement points — the named points the team grades a style by, each with its tolerance — that grading and fitting specs draw from.

The **POM Library** page — reached from the **Admin & configuration**
sidebar group — is the workspace's reusable catalog of **points of
measurement**: the named points on a garment the team grades and fits
to, such as `Chest`, `Sleeve length`, or `Sweep`, each with the
tolerance the team works to and the order the team likes to see them
in.

Curating the library once means every per-order
<Tooltip tip="The per-order-style POM-by-size reference grid the team grades a sample to — points of measurement down the rows, sizes across the columns, plus a tolerance column.">[grading table](/reference/glossary#grading-table)</Tooltip>
the workspace builds starts from the same palette of points — same
names, same tolerances, same order — instead of being typed from
scratch every time.

## What it is

A single page under **Admin & configuration → POM Library** with two
regions:

* A list of every point of measurement the workspace has on file,
  sortable by the team's preferred order, with an "Inactive" badge on
  any point that has been retired.
* An add / edit modal where you create a new point or revise an
  existing one.

A **POM record** carries four things: the point's English name, an
optional Chinese name, the tolerance the team works to, and the sort
order that controls where it sits on the list. Names are entered
manually — the page does not translate between languages, so the team
types the Chinese name themselves when they want one.

## Why it exists

Grading and fitting work on a fixed vocabulary. Two stages graded
against `Chest` and `CHEST WIDTH` are not graded against the same
point; a workspace that lets every grading table type its own names
is a workspace whose history cannot be read across styles, seasons,
or factories.

POM Library is the team's agreed vocabulary. Once a point is in the
library — with the name the team has settled on, the tolerance the
team works to, and a sort order that matches how the team reads a
grading sheet — every grading table the workspace builds afterwards
picks from that catalog. The team grades the same way across every
style, and a reader looking at two grading tables side by side sees
the same point in the same place with the same tolerance.

## When it is used

* **At workspace start-up**, to seed the library with the points the
  team always grades — chest, waist, hips, sleeve length, sleeve
  opening, body length, sweep, and so on — plus the tolerances the
  team has agreed on for each.
* **When a new point needs to be added** because a style introduces a
  measurement the workspace has not graded before.
* **When the team's tolerance for a point changes** and the library
  copy should reflect the new tolerance going forward.
* **When the team retires a point** that is no longer in active use —
  the point is deactivated so it stops appearing in new grading
  tables, but the points already on existing grading tables are
  preserved.

You reach the page from the sidebar under **Admin & configuration →
POM Library**.

## Dependencies

There are no upstream master records POM Library depends on. The
library is the upstream master itself.

## What depends on it

* The per-order **[grading table](/modules/grading)** reads from this
  library. When the team builds a grading table for a style on an order,
  the points on the table are lifted from the POM Library catalog — and
  any *one-off* point the team adds in the table itself stays local to
  that table (the affordance lives on the Grading grid, not here).
* The **Grading** section of an MO's
  <Tooltip tip="The factory-facing production-meeting-spec workbook generated for a style's MO, filled from a PMS template.">[PMS document](/reference/glossary#pms-document)</Tooltip>
  — the workbook the factory works to — reads from the same per-order
  grading record. See the [PMS Templates page](/admin/pms-templates) for
  how the workbook is filled.

The library is the catalog. The actual per-style measurements live on
the grading tables that draw from it, not on the records here.

## The POM record

Each POM record on the library carries:

* `POM Name` — the point's English name (for example, `Chest`,
  `Sleeve length`). Required. This is the name that appears on every
  grading table that picks the point.
* `POM Name (Chinese)` — the point's Chinese name, optional. Typed
  manually by the team; the platform does not translate from the
  English name. When the workspace runs a grading sheet in a Chinese
  locale, this is the label that shows.
* `Tolerance` — the `±` the team works to for the point, entered in
  inch fractions as a numerator over a denominator (for example,
  `1/4"`). The list shows the fraction next to the point's name so
  the team can read the tolerance at a glance.
* `Sort Order` — an integer that controls where the point appears in
  the list (and, by extension, where it appears on grading tables
  drawn from the library). The team uses the sort order to keep the
  list in the same shape as a graded garment — top of the body to
  the hem — rather than alphabetical.

A record is either **active** or **inactive**. An active record is
offered to new grading tables; an inactive record is not.

## Adding and editing a POM

To add a point:

1. Open the add modal from the page.
2. Type the `POM Name` (English). Optionally type the `POM Name
   (Chinese)`.
3. Enter the `Tolerance` as a numerator and denominator in inches —
   the resulting fraction is shown beside the input so you can
   double-check before saving.
4. Set a `Sort Order` for the point.
5. Save.

To edit an existing point, open the row's edit pencil. The same modal
opens, pre-filled. Editing a record updates the library copy; grading
tables already built from this point keep the values they were built
with — the library edit applies to *new* grading tables, not to ones
already in flight.

## Reordering points

The list carries inline up / down arrows on each row. Pressing an
arrow moves the row's sort order one step and persists the change
immediately. The order on the page is the order new grading tables
read the catalog in, so the list is the team's chance to put the
points in the shape they want to read them in.

## Deactivating a POM

The deactivate action on a row is **soft** — it does not delete the
record. A deactivated point:

* Stops appearing in the picker when a new grading table is built.
* Still appears on every grading table that has already picked it
  (those measurements are preserved as they are; deactivating the
  library copy does not strip them out).
* Shows an "Inactive" badge in the library list, so the team sees the
  full history of points the workspace has graded against.

Reactivate the point — through the same row's action — and it returns
to the picker.

There is no hard delete from this page. The library is the team's
audit of every point the workspace has ever graded; retirement is
through deactivation, so the history is never broken.

## Business rules

1. **A POM record carries an English name, an optional Chinese name,
   a tolerance in inch fractions, and a sort order.** Names are
   entered manually; the page does not translate between languages.
2. **Tolerances are in inch fractions.** The library is the catalog;
   when a grading table picks a point, the tolerance carries with
   it.
3. **Sort order is the team's reading order.** The list — and the
   grading tables drawn from it — appear in sort-order sequence, not
   alphabetical.
4. **Deactivation is soft.** A deactivated point stops appearing in
   new grading tables; existing grading tables keep the points they
   were already built with.
5. **There is no hard delete from this page.** Retiring a point is
   through deactivation, so the team's grading history is preserved.
6. **The library is the catalog, not the measurements.** Actual
   per-style measurements live on the grading tables on the fitting
   record, not on the records here.

## Best practices

* **Seed the library deliberately, once.** Settle on the team's full
  list of points — and the tolerance for each — before the team
  starts grading styles, so the first round of grading tables already
  uses the right palette.
* **Keep tolerances consistent across the catalog.** A tolerance that
  differs from the team's house standard for a single point will
  travel onto every new grading table that picks it — set it
  thoughtfully.
* **Set the sort order to read like a graded garment.** Top of body
  to hem, then sleeves, then trims, is a common shape. The list is
  the team's chance to make every grading table readable.
* **Type the Chinese name in deliberately when a workspace runs
  bilingual sheets.** The platform will not invent one — the field is
  optional but worth filling in if the factory reads the sheet in
  Chinese.
* **Deactivate, do not delete.** A retired point still tells the story
  of what the workspace used to grade against; deactivation keeps the
  audit trail intact.
* **Use one-off points on the grading table only when a style truly
  needs a point the library does not yet hold.** If the team starts
  reaching for one-off points often, that is a signal to widen the
  library instead.

## Related pages

* [Setting up your tenant](/admin/setting-up-your-tenant)
* [PMS Templates](/admin/pms-templates)
* [Grading](/modules/grading)
* [Samples & fitting](/modules/samples-and-fitting)
* [Styles](/modules/styles)