What the Data for This Category Looks Like
Semiconductor wafer fabrication process documents and exception report data typically originate from wafer fab MES systems, equipment FDC modules, defect detection and metrology devices, wafer map databases, process recipe libraries, and experimental datasets. Daily produced documents include process flow cards, batch operation records, equipment recipes, SPC control charts, exception handling sheets, defect maps, and yield reports. Data updates align with wafer batch transfers, real-time parameter reporting, and daily summary statistics. Fields include nanoscale process parameters, batch numbers, defect coordinates, and yield values. Some documents require coupling data across multiple processes.
What Constraints Do These Characteristics Impose on HTTP Interfaces and External Systems
Data for this category comes from multiple sources with varying update frequencies. Interfaces must support parallel pulling of different data source types via multiple endpoints. Fields for nanoscale parameters and cross-process coupling data have high complexity. Interfaces must support large field transmission and precise field mapping. The uniqueness of each wafer batch requires interface request parameters to bind a unique batch ID to avoid data confusion. Temporary network fluctuations in cleanroom environments may reduce call stability. Interfaces must include basic retry mechanisms. Additionally, documents from ultra-long process chains require association with multi-module data. Interfaces must support batch data pulling and aggregation.
How to Configure the Settings
| Configuration Item | Recommended Practice | Rationale |
|---|---|---|
| Authorization Configuration | Fill in the Authorization field according to the connected system's requirements. Retain or remove the Bearer prefix as needed. | Adapt to external system authentication rules, avoid authentication failures caused by FastGPT's default automatic Bearer supplement. |
| Request Variable Mapping | Bind the wafer batch ID and process node ID to the variables parameter. | Match the field naming conventions of MES systems, ensure requests point to correct production batch data. |
| Response Data Parsing | Extract the responseData field as core business data. | Align with the standard return structure of most external systems, simplify data extraction logic. |
| Timeout Setting | Set to 60-120 seconds. | Adapt to the latency characteristics of wafer manufacturing data pulling, avoid request interruptions caused by excessive data volume. |
| Single Request Data Volume | Limit wafer batch-related data per request to no more than 8000 characters. | Avoid interface overload, adapt to request size limits of most external systems. |
| Error Retry Strategy | Retry 2 times after failure, with a 10-second interval. | Respond to temporary network fluctuations in cleanroom environments, improve call stability. |
Three Common Misconfigurations
- Automatically carrying Bearer tokens during interface calls causes authentication failures. This happens when the automatic token supplement function in FastGPT v4.17.0 is not disabled. The system's default added prefix does not match external system requirements.
- Empty business fields in interface returns prevent extraction of process data. This happens when the wafer batch ID is not correctly bound to the
variablesparameter, causing requests to not point to the correct production batch. - Interface calls time out with a 504 status code. This happens when an insufficiently long timeout period is set, failing to adapt to the pulling latency of large-volume wafer manufacturing data.
How to Confirm Successful Configuration
- Initiate a test request. Verify that the external system receives correct wafer batch data. Check that the batch ID in the request parameters matches expected values.
- Check the
responseDatafield in the interface return. Confirm that expected business document content such as process flow cards and exception handling sheets is included. - Simulate a network fluctuation scenario. Verify that the error retry strategy takes effect, and the interface automatically retries after failure.
- Compare locally stored process recipe data with interface returned data. Confirm that field matching meets business requirements.
The values provided are common starting points and should be evaluated against samples specific to the actual operational environment.