Qeasy Cloud
Get Started

Kingdee Other-Outbound Sync to Jushuitan: A Single-Strategy Field Guide

· 钟家寿· Integration Solutions· 17 views· 4 min read
JushuitanKingdee CloudInventory Sync供应链集成轻易云单策略教程

What This Strategy Solves

In a real retail engagement, store write-offs, batch adjustments, internal requisitions, and transfer-out scenarios generate dozens to hundreds of "Other-Outbound" documents every day. Kingdee Cloud Cosmic records the real cost and inventory ledger on the back-end ERP side, while Jushuitan records SKU availability on the front-end e-commerce and store middle platform. When these two sides are not synced in time, after about three days warehouse and finance will discover the classic mismatch: Jushuitan still shows sellable stock while the ERP ledger has already deducted it—or the reverse. This strategy pushes Other-Outbound documents from Kingdee to Jushuitan incrementally so that the e-commerce-side availability moves in the same direction as the ERP-side book inventory.

Data Flow and Field Mapping

The overall flow is Kingdee Cloud Cosmic (B) → Qeasy middle layer → Jushuitan (A). On the Kingdee side we read the header and lines of the Other-Outbound document after save or audit; on the Jushuitan side we hit the corresponding Other-Outbound API. A typical field mapping is shown below.

DimensionKingdee (source)Jushuitan (target)Mapping notes
Document no.FBillNoexternal_noPass through verbatim for idempotent dedup
Document dateFDatecreatedNormalize to yyyy-MM-dd
WarehouseFStockIdwarehouse_idMaintained via a centralized warehouse mapping table
SKU codeFMaterialIdsku_idItem master sync must precede this strategy
QuantityFQtyqtyNumeric type; mind unit conversion
RemarkFNoteremarkTruncate to avoid rejection
Business typeFTranTypetypeUse type to distinguish write-off / transfer / internal

A pattern we have validated repeatedly: keep mappings such as "warehouse code ↔ warehouse_id" and "material code ↔ sku_id" in Qeasy's central mapping repository. The document stream itself only performs field mapping and cleansing, never hard-codes any identifiers. When a store or warehouse is later reorganized, you change it in one place.

How to Configure It in Qeasy

Inside the Qeasy Data Integration Platform, this strategy is usually split into three stages: source retrieval, transformation/cleansing, and target write.

  1. Source retrieval: call the Kingdee query interface with an incremental cursor on FDate + FBillNo. From the very first version we recommend also carrying FModifyDate, so that future changes to the incremental condition are easier.
  2. Transformation/cleansing: in Qeasy's transformer, rename fields, convert units, and apply null fallbacks. We recommend handling header and lines in stages—land the header first to obtain the target document number, then expand the lines in batch. This greatly reduces pressure on the Jushuitan interface.
  3. Target write: call the Jushuitan Other-Outbound create API, always carrying external_no for idempotency. Failed rows go into the retry queue; do not drop them silently.

The safer rollout is to "trial-run" for one week, let the failure samples accumulate, and only then enable the production schedule.

Implementation Steps

  1. Incremental starting point: pick a historical date (e.g., the first day of the month) as the start, pull a batch of historical documents for reconciliation, and smooth out mapping and unit-conversion issues.
  2. Full-volume trigger: during the initial go-live, push all historical documents after the starting point in one pass; afterwards, sample-check quantities and amounts on both sides manually.
  3. Schedule frequency: in e-commerce scenarios, 15–30 minutes per round is recommended; for pure store write-offs, once per hour is acceptable. Configure the cron expression in Qeasy's scheduler, and stagger the timing so it does not collide with purchase inbound or return strategies.
  4. Monitoring and alerting: in Qeasy, hang two core metrics—failure rate and document lag time—and trigger enterprise IM alerts when thresholds are exceeded.

Pitfalls and Lessons Learned

  • Item master not synced first: this strategy assumes sku_id already exists on the Jushuitan side; otherwise the entire document is dropped. The safe practice is to make item-master sync an upstream prerequisite task.
  • Header and lines submitted together: submitting a large line set in one shot often causes Jushuitan timeouts. Splitting into header-first then lines makes problem documents easier to locate and re-push.
  • Wrong incremental field: using FDate in cross-day backfill scenarios will miss documents; FModifyDate is more reliable.
  • Missing idempotency key: without external_no, network retries create duplicate documents, which is a classic disaster during finance reconciliation.
  • Units and precision: Kingdee commonly uses base units while Jushuitan commonly uses sales units. For multi-unit SKUs the conversion must be explicit; otherwise quantities can differ by orders of magnitude.

When It Applies and When It Does Not

Applies: multi-store / multi-warehouse retail and e-commerce distribution, where non-sales outbound (write-off, transfer, internal requisition) needs to align with e-commerce inventory. Does not apply: pure manufacturing scenarios—where outbound is dominated by sales shipments and should use a sales-outbound strategy—as well as the early stage when either Kingdee or Jushuitan has not yet completed basic-data synchronization.

Original content. Please credit the source when reposting: https://www.qeasy.cloud/insights/solutions/strat-jushuitan-kingdee-cloud-7144-n7926f273-db714bb4

Comments