Inside JD.com’s Order Transfer and Fulfillment Workflow

The author uses JD.com as a case study to detail order transfer, fulfillment workflow, and risk‑control processes.

Order Transfer

Order transfer is the planning stage where millions of orders per minute are allocated to appropriate fulfillment channels (PC, app, WeChat, etc.) and then distributed to specific warehouses. It determines which warehouse will produce the order, effectively linking order allocation with inventory status.

In practice, order transfer acts as a distribution mechanism, deciding which warehouse will handle production based on inventory availability.

Inventory Types

1) Primary (Level‑1) Inventory

Initially, inventory was managed at a national level: orders in Beijing checked Beijing’s inventory, Shanghai orders checked Shanghai’s inventory.

As JD.com scaled, this model could not meet the massive order volume.

2) Secondary (Level‑2) Inventory

RDCs serve as central warehouses, while FDCs are front‑line warehouses (e.g., Jinan, Tianjin). JD.com divides the country into seven regions (Beijing, Shanghai, Guangzhou, Chengdu, Wuhan, Xi’an, Shenyang). An order from Jinan first checks local stock; if unavailable, it is fulfilled from the regional hub (Beijing).

3) Support Relationship

Front‑line warehouses stock fast‑moving items for their region, while long‑tail items are supported by the central hub, ensuring coverage across both first‑ and second‑tier cities.

Key Aspects of Order Transfer

The main goal is to create a production plan and quickly route orders to the appropriate production system. Two core systems are involved:

1) Order Transfer Service System

Defines how, where, and when an order is produced. It serves upstream OFW, receives order data, and triggers transfer via middleware. It handles fast‑track processing for in‑stock, non‑appointment orders across self‑operated and POP channels, relying on OFW, distribution configuration, inventory, and large‑appliance scheduling.

2) Order Planning Engine System

Creates production plans to meet fulfillment promises while optimizing cost. It handles non‑in‑stock and appointment orders, processing complex flows such as multi‑warehouse, FDC, and parallel inventory. It depends on distribution center configuration, inventory, product data, large‑appliance scheduling, Promise, VPR, middleware, and OFW‑COI.

Transfer Process

The process first checks inventory, determining which warehouse will produce the order. JD.com distinguishes between front‑end inventory (SKU‑level) and OFC inventory (order‑level). Images illustrate the inventory rules.

Order Fulfillment Workflow

After an order enters the pipeline, the system receives it, invokes splitting and transfer services, and pushes data to nationwide warehouses. It handles order cancellation, modification, returns, and real‑time status feedback to ERP, Promise, and BI systems, while providing operational tools and alerts.

The workflow consists of two main parts: order information back‑transmission and downstream transmission.

Order Status Back‑Transmission

Real‑time status (e.g., printing, packaging, outbound) is sent to ERP, triggering inventory, outbound, tracking services, and customer SMS notifications.

Non‑Status Business Data Back‑Transmission

Details such as package count, weight, carrier, and invoice are sent to WMS, TMS, and DMS systems.

Order Risk Control

Risk control prevents malicious gift‑bundling exploits. Users may exploit system gaps to receive free gifts without paying for the main product, leading to mismatched deliveries and cancellations.

The core strategy is linked cancellation: if a gifted item is received but the main product is returned, the system cancels the related orders based on a recorded “main‑gift” relationship.