Datasets:
metadata
license: mit
task_categories:
- tabular-classification
tags:
- predictive-maintenance
- iot
- sensors
- fleet-management
size_categories:
- 1K<n<10K
Predictive Maintenance Engine Sensor Dataset
Engine sensor readings from commercial diesel vehicles for predictive maintenance classification.
Features
| Feature | Description | Unit |
|---|---|---|
| Engine RPM | Engine revolutions per minute | RPM |
| Lub Oil Pressure | Lubrication oil pressure | bar |
| Fuel Pressure | Fuel delivery pressure | bar |
| Coolant Pressure | Cooling system pressure | bar |
| Lub Oil Temp | Lubrication oil temperature | °C |
| Coolant Temp | Engine coolant temperature | °C |
| Engine Condition | Target: 0=Normal, 1=Needs Maintenance | binary |
Dataset Splits
| Split | Samples | Purpose |
|---|---|---|
| train | 75% | Model training |
| validation | 10% | Hyperparameter tuning |
| test | 15% | Final evaluation |
All splits are stratified by Engine Condition to maintain class balance.
Usage
from datasets import load_dataset
dataset = load_dataset("jskswamy/predictive-maintenance-data")
train_df = dataset["train"].to_pandas()
License
MIT License