Cleanroom Data Readiness
Use this guide before SemiOps outputs are used in shift review, ISO evidence preparation, SMT quality analysis, or maintenance planning.
Readiness Flow
Prerequisites
| Requirement | Why it matters |
|---|---|
| Source owner | Each source needs a reviewer for field meaning and sync behavior. |
| Cleanroom hierarchy | Readings need cleanroom, zone, equipment, and sampling context. |
| Data quality review | Trends and assessments depend on current, aligned readings. |
Inputs
| Source | Readiness checks |
|---|---|
| Cleanroom master data | Cleanroom ID, class target, area, pressure zones, and equipment links. |
| Particle readings | Particle size, count, sample point, period, unit, and instrument context. |
| Pressure and environment | Differential pressure, temperature, humidity, timestamp, and sensor location. |
| SMT records | Line, station, product context, OEE, defects, units, and period. |
| Utility readings | Power, cooling, load, cost assumptions, and meter scope. |
| Maintenance records | Filter pressure drop, equipment status, schedules, alerts, and work feedback. |
Procedure
- List source systems and source owners.
- Map cleanroom, line, utility, and equipment identities.
- Validate particle, pressure, temperature, humidity, vibration, and power units.
- Check source freshness and rejected records.
- Mark each dataset as ready, ready with limitation, or blocked.
- Record limitations in the SemiOps review package.
Expected Output
The team should have a clear readiness record showing which SemiOps views can support operations, which can support evidence review only, and which are blocked.
Validation Checklist
- Cleanroom and equipment identities match source data.
- Sensor timestamps and units are consistent.
- SMT records include line, product, and period.
- Utility scope and assumptions are visible.
- Data limitations are recorded before review.
Failure Handling
| Symptom | Response |
|---|---|
| Sensor timestamps are misaligned | Check timezone, sampling interval, and connector sync history. |
| Particle data lacks sampling context | Keep ISO assessment as draft evidence until context is confirmed. |
| SMT defect data is incomplete | Do not compare DPMO until count, unit, and opportunity assumptions are reviewed. |
| Maintenance data is stale | Record freshness gap and check source connector status. |