What Data Looks Like for This Category
Technical documentation for production planning and shop scheduling covers sales orders, master production schedules, material requirements plans, weekly production schedules, work orders, material shortage lists, delivery change notices, and more. Data sources include ERP orders and inventory, BOMs and process routes, equipment capacity, staff shifts, work in process, and supplier deliveries. Update cadence varies by business scenario: basic capacity and inventory data is synced daily, production schedules are updated weekly, and updates for documents like delivery change notices and material shortage lists are triggered in real time when orders or exceptions occur. Data fields include order numbers, material codes, process man-hours, equipment load, shift duration, and more, covering all products and processes, and linking resource constraints to delivery commitments.
What Constraints These Characteristics Impose on Deployment and Upgrade
Multiple heterogeneous data sources require deployment to support integration with multiple data sources such as ERP and equipment management systems. Different connection protocols and authentication rules must be configured. Differentiated update frequencies require the system to support flexible incremental sync task configuration, balancing real-time performance and resource usage. Complex, multi-type documents require vector library indexes to cover exclusive fields for different document types such as orders and schedules, to avoid reduced retrieval accuracy. Data scale fluctuates with order volume and exception events. Sufficient storage and computing resources must be reserved during deployment. During upgrades, data sync must be maintained without interruption to avoid disrupting normal shop scheduling operations.
How to Set Configurations
| Configuration Item | Recommended Approach | Basis for This Approach |
|---|---|---|
MONGODB_URI | Enter the connection string for a production environment independent cluster, including database authentication information | Scheduling data involves core business data such as orders and work orders, requiring high-availability non-relational storage to support fast retrieval of multiple document types |
PG_URL | Enter the connection address for a read-only instance associated with the ERP system | ERP order and inventory data must be synced, and a read-only instance avoids performance impact on the business production system |
REDIS_URL | Configure as a local cache cluster, set cache duration to 300-600 seconds | Frequently queried equipment capacity and staff shift data require low-latency access, and caching reduces database query pressure |
DB_MAX_LINK | Set to 20-30 | Meets database connection requirements for processing multiple work orders and rescheduling tasks simultaneously, avoiding service interruptions caused by exhausted connections |
STORAGE_S3_ENDPOINT | Enter the access endpoint for self-built object storage | Production documents require local storage to avoid public network transmission delays and data security risks |
FE_DOMAIN | Enter the private domain for internal access | Used only by internal departments such as shop schedulers and PMC, no public exposure reduces security risks |
Three Common Misconfigurations
- Symptom: Interface calls return 500 errors, and curl tests show no normal response. Cause:
REDIS_URLorPG_URLis not configured correctly, causing failed connections to dependent caches or databases. In FastGPT v4.17.0, this type of error directly blocks pre-data loading for document cross-review. - Symptom: Not all production documents load after offline deployment. Cause:
DB_MAX_LINKis not adjusted to match storage resources in the offline environment, or the single-time document load limit is not restricted, causing memory overflow and failure to complete full document index construction. - Symptom: Scheduling data sync is interrupted after upgrade. Cause: The original
ROOT_KEYconfiguration is not retained, causing the system to fail to recognize original permission rules, blocking sync tasks and preventing normal pulling of the latest ERP order data.
How to Confirm Proper Configuration
- Run the FastGPT built-in database connection test tool to verify the connection status of
MONGODB_URIandPG_URL, confirming no authentication or network errors. - Submit a test work order document, observe the update delay of cached data, and confirm that the
REDIS_URLcache duration meets real-time business requirements. - Upload a test weekly production schedule, verify that the
STORAGE_S3_ENDPOINTstorage and reading functions work normally, and that documents can be indexed correctly. - Access the internal domain
FE_DOMAIN, confirm that the front-end page can normally load the scheduling-related document list with no loading failure prompts.
The values provided are common starting points, and should be evaluated against the target deployment’s own samples.