Endpoint
Request Body
Asset identifier (e.g., “Motor-01”). Must be 1-100 characters.
Maintenance event classification. Allowed values:Preventive Maintenance:
PREVENTIVE_LUBRICATION- Scheduled lubrication activitiesPREVENTIVE_CLEANING- Scheduled cleaning activitiesPREVENTIVE_INSPECTION- Scheduled inspection activities
CORRECTIVE_BEARING_REPLACEMENT- Repair/replacement of bearingsCORRECTIVE_ALIGNMENT- Alignment correctionsCORRECTIVE_ELECTRICAL- Electrical repairs
STATUS_CALIBRATION- System calibration eventsSTATUS_RESTART- System restart events
Event severity level. Allowed values:
LOW- Minor maintenance or routine activitiesMEDIUM- Standard maintenance requiring attentionHIGH- Urgent repairs or significant issuesCRITICAL- Emergency situations requiring immediate action
Human-readable description of the maintenance activity. Must be 1-1000 characters. Include relevant details such as part numbers, observations, and actions taken.
Event timestamp in ISO 8601 format with UTC timezone (e.g., “2026-01-31T10:30:00Z”). If not provided, defaults to the current server time. Allows backdating for historical entries.
Response
Unique identifier (UUID v4) for this log entry
Asset that was logged
Type of maintenance event that was recorded
Severity level of the event
Timestamp of the event (UTC)
Confirmation message indicating successful log creation
Use Cases
- Ground-Truth Labeling: Record maintenance events that can be correlated with sensor data for supervised ML training
- Bearing Replacement Tracking: Log bearing replacements to correlate with vibration anomalies detected in sensor data
- Lubrication Monitoring: Record lubrication events to track maintenance effectiveness over time
- Failure Analysis: Document electrical repairs and other corrective actions for failure pattern analysis
ML Integration
Maintenance logs stored via this endpoint can be queried alongside sensor data to:- Train supervised models that predict failure modes based on sensor patterns
- Identify early warning indicators in sensor data before failures occur
- Validate anomaly detection algorithms against real-world maintenance events
- Build predictive maintenance models with labeled training data