v 2.4AI-native warehouse management for factories · zh / en

Your warehouse,
finally articulate.

TALS · WMS is a next-generation WMS purpose-built for discrete manufacturing — ten document types, one workspace, plus the Talbi AI assistant. Every material, every doc, every scan is traceable, explainable, and optimizable.

Deploy · on-prem or cloudTerminals · iOS / Android / industrial PDAIntegration · REST / DB / file exchange
TALS.ai/wms · Workspace / inbox

Workspace

Good morning, Wen · SH-01
Inbound
248
+12% ▴
Outbound
412
+5% ▴
Pending
7
2 urgent
Lines
68
Throughput · 7d+18.4%
Doc no.Type / partyStatusQtyRows
PI-20260509-014Purchase IN · YibangSubmitted1,24012
SO-20260509-208Sales OUT · TmallApproved8208
MI-20260509-003Mfg IN · WO-1142Done5403
TR-20260509-022Transfer · SH-01 → SH-02Partial2205
10+
Doc types · IN / OUT / Internal / Count
99.6%
Scan accuracy · audited
3.2×
Faster doc throughput
28d
Avg time to go-live
Document matrix · 01

Ten document types, one inbox.

From supplier receipts to customer shipments, from work-order picking to transfers and counts — every paper trail in your warehouse, normalized. Each doc has its own fields, state machine, and approval chain — but one UI, one set of shortcuts, one assistant.

PI · 入库
Purchase inbound
Supplier · 3-way match
MI · 入库
Mfg inbound
From work orders
SO · 出库
Sales outbound
Order → pick → ship
PK · 出库
Pick list
Wave / path opt.
SH · 出库
Shipment
Carrier · tracking sync
TR · 内部
Transfer
WH / bin transfer
MP · 内部
Mfg picking
Work order → floor
RT · 内部
Return / RMA
Customer / supplier
SC · 盘点
Stock count
Dynamic / static
DM · 盘点
Write-off
Damage · loss
Talbi · AI assistant · 02

Not a chat box.
A copilot in your warehouse.

Talbi reads every doc, lot, and scan event across your factory. It surfaces anomalies before you ask. It pre-fills 80% of your documents — so you don't fill rows one by one.

T
Talbi
Reading 184 docs · 12 WH · 4.2K rows
Goods on PO-2026-0488 arrived, the paper count looks off — can you check?
Identified. I cross-checked the PO lines in ERP against this morning's 09:42 receipt:
!SKU-9911 · Canvas Tote / L
Plan 240, received 202 — 38 short. Lot LOT-26050901. Paper says 240 — suggest photographing and opening RT return.
⊕ Draft RTNotify buyer
Ask Talbi…
  • 01

    Surface anomalies — proactively

    Shortages, overages, lot misalignment, duplicate scans, suspicious bin moves — flagged as they happen, not at month-end.

  • 02

    Draft 80% of every doc

    Say "draft inbound for PO-2026-0488" — it pre-fills lines, bins, lots, prices. You review and sign.

  • 03

    Every doc is explainable

    Open an RT — Talbi explains why it was created, which PI it links to, the loss, and a suggested owner. Audit-ready by default.

  • 04

    Bilingual zh / en · factory vocabulary

    Quarantine, WIP, packaging, rework — vocabulary trained on factory floor context, not translated from generic logistics.

Module overview · 03

One system,
three audiences — manager, operator, shop floor.

Managers on the workspace. Operators on PDA. The floor on the big-screen wallboard. Three surfaces, one data spine, one assistant context.

Doc lifecycle · 04

From a single line — to a posted ERP voucher.

Every document follows the same five-stage flow: create → submit → approve → execute → done. Standard state machine, configurable approval chain, execution via scan / pick / load / writeback.

01 / CREATE

Create

Manual, ERP / MES sync, or Talbi one-click. Header + lines, field templates per type.

02 / SUBMIT

Submit

Validate required, compute totals, detect shortages and duplicates, version-stamp.

03 / REVIEW

Review

Multi-level approval by amount / category. Talbi marks anomalies — approver clears or rejects.

04 / EXECUTE

Execute

PDA scan / putaway / truck load — every step logged.

05 / DONE

Done

Write back to ERP / MES, post journal, update lots & bins, push to wallboard.

Industry · 06

Three factories, three rhythms.

Order-driven apparel, lot-driven CPG, work-order-driven electronics — each has its own paper trail. We adapt, we don't template.

Apparel

Order-driven · seasonal peaks

Million-SKU complexity, color × size matrix, 30-min packing windows. Our wave picking + path optimization shipped 38% more orders/day during peak.

1.2M
SKU
+38%
Throughput
CPG / Food

Lot-driven · traceability

Raw lot → batch run → finished case, six-layer lot lineage. Write-off + QC quarantine sync — quality incidents triangulated in 3 min.

6 layers
Lot lineage
3 min
Recall scope
Electronics

Work-order driven · kitting

Raw / WIP / FG triple-link, BOM-checked kitting, shortages pushed straight to the buyer's phone.

99.6%
Kit accuracy
−42%
Line stops
Get started

Let Talbi take a look
at your warehouse.

30-minute live demo. We plug Talbi into your sample data and walk through anomaly detection, doc drafting, and PDA scan-in — live.

Book a 30-min demo →Contact us ↗
On-prem availableSOC 2 Type II in progress24/7 zh support