Supply Chain Integration Solution Between JikeCloud and Kingdee Cloud Flagship Edition

Scenario and Value

In retail and distribution environments where multiple systems operate in parallel, data silos between front-end business systems and back-office financial ERP platforms often cause delays in reconciling sales orders, inventory movements, and accounts receivable. This in turn affects reconciliation efficiency and inventory accuracy. In a typical deployment, JikeCloud handles e-commerce orders and inventory operations on the front end, while Kingdee Cloud Flagship Edition carries financial accounting and supply chain management on the back end. A unified data channel is required to synchronize documents and master data between the two systems.

This solution covers core business domains including sales orders, inbound and outbound inventory, procurement, transfers, inventory surpluses and losses, returns, and master data. Through 31 synchronization and query strategies, front-end business documents flow into the back-office ERP following a defined dependency order, while essential master data such as materials, customers, and suppliers is synchronized bidirectionally. The result is consistent financial accounting, accurate inventory, and end-to-end traceability.

Integration Architecture and Data Flow

The architecture is built around JikeCloud and Kingdee Cloud Flagship Edition, with an integration platform responsible for strategy orchestration, field mapping, code conversion, and exception retry. Data flows fall into three categories:

• JikeCloud → Kingdee Cloud Flagship Edition (A→B): 16 synchronization strategies covering sales orders, inbound and outbound documents, purchase receipts, transfer orders, inventory surplus and loss, assembly and disassembly, return documents, and refund-based receivables, together with master data such as sales channels, warehouses, and suppliers.
• Kingdee Cloud Flagship Edition → JikeCloud (B→A): 4 synchronization strategies, primarily for customer archive query results and material master data sync into JikeCloud product archives.
• Internal queries (INTERNAL): 11 query strategies used to retrieve source documents, related documents, customer archives, bank accounts, non-Qimen return-exchange documents, and other auxiliary data during synchronization.

The strategies have strict execution dependencies:

1. Master data (customers, warehouses, suppliers, materials) must be synchronized before business documents to avoid failures caused by missing codes.
2. The sales order chain follows: sales order → sales outbound document → sales return / financial receivable (refund only). Downstream documents depend on the billno and entry-level links of upstream documents.
3. The procurement chain: purchase receipt → purchase return material document. The return material document requires a two-level cross-reference back to the original receipt.
4. Transfer orders: the business date must come from the inbound document retrieved through the "Query JikeCloud Inbound Document" strategy, rather than the creation timestamp.
5. Returns and refunds: sales return documents depend on already-synchronized sales outbound documents; refund-type financial receivables are generated only after the corresponding sales order has been synchronized.

Interface List

The core synchronization interfaces, organized by data object and direction, are as follows:

• Master data synchronization: JikeCloud sales channel → Kingdee customer; JikeCloud warehouse → Kingdee warehouse; JikeCloud supplier → Kingdee supplier; Kingdee material → JikeCloud product.
• Sales order synchronization: JikeCloud sales order → Kingdee sales order, sales outbound document, sales return document, financial receivable (refund only).
• Procurement order synchronization: JikeCloud inbound document (purchase) → Kingdee purchase receipt; JikeCloud outbound document (purchase return) → Kingdee purchase return material document.
• Inventory synchronization: JikeCloud inbound document → Kingdee other inbound, inventory surplus; JikeCloud outbound document → Kingdee other outbound, inventory loss; JikeCloud transfer order → Kingdee direct transfer document.
• Other business documents: JikeCloud assembly document → Kingdee assembly document; JikeCloud disassembly document → Kingdee disassembly document.
• Internal queries: Kingdee sales order / sales outbound / purchase receipt / customer / bank account queries; JikeCloud inbound / outbound / customer archive / non-Qimen return-exchange / refund / purchase return / combination pack queries.
• Alert push: exception notifications pushed through the enterprise WeChat channel.

Implementation Points

At the field mapping level, each strategy uses a two-layer structure combining a header and entries. The header carries document-level information (document number, business date, currency, exchange rate, customer/supplier code, inventory organization, business type, etc.), while the entries use COLLECTION mappings for source fields such as goodsDetail or goodsDocDetailList. Combination products (product codes starting with ZH) must be decomposed into child material lines at the source end, with quantity and amount allocated according to shareRatio. Salesperson names must have any parenthetical content removed before being written into the target field.

Code mapping is a critical part of the implementation. Store codes must be mapped to Kingdee customer codes, product codes to Kingdee material codes, supplier codes to Kingdee supplier codes, and warehouse codes to Kingdee warehouse codes. All mapping relationships are established in the master data synchronization strategies and referenced directly in subsequent business document strategies through material and customer codes, avoiding redundant maintenance.

Entry-level linking (billentry_lk) depends on AfterTargetGenerate scripts: after the target document is generated, the script matches entry lines between source and target documents based on material code, tax-inclusive price, and quantity, then fills in the link field on the target entries to ensure correct traceability for downstream push-down, return, and receivable documents. For sales returns, the qty must be converted to a negative value and deducted against available quantity. For purchase returns, supplier_number must be validated to match the related purchase receipt.

Best Practices

1. Master data first: During initial rollout, prioritize running the four master data synchronization strategies for customers, warehouses, suppliers, and materials, and use full synchronization for the first load. Afterwards, switch to incremental synchronization using business codes as idempotency keys.
2. Chain-based dependency orchestration: Schedule strategies in the order of "master data → sales order → sales outbound → sales return/receivable" and "purchase receipt → purchase return" to prevent downstream failures caused by missing upstream documents.
3. Centralized code mapping maintenance: Build all code mapping relationships in the master data strategies. Business document strategies only reference codes and do not maintain their own mapping tables, reducing maintenance overhead.
4. Separate handling of special document types: Combination products, returns, refund-only documents, and transfer scenarios must be distinguished within strategies through document type constants and source fields such as tradeType / inouttype, avoiding mixed execution with regular documents.
5. Exception handling and retry: Configure unified retry strategies and alert notifications for scenarios such as missing codes, target document validation failures, and API rate limiting. Push notifications in real time to operations personnel through the enterprise WeChat channel.
6. Idempotency and traceability: All synchronization strategies use source document primary keys as the target document id or sourcebillno, ensuring reruns do not create duplicates. Entry-level links are matched across three dimensions—material, price, and quantity—guaranteeing end-to-end business traceability.

Through this solution, the retail enterprise achieved full data integration between JikeCloud and Kingdee Cloud Flagship Edition across business documents and master data, retaining the flexibility of the front-end business system while ensuring consistent financial accounting and inventory data on the back-office ERP.