mindweave's picture
chore: change license from CC-BY-4.0 to CC-BY-NC-4.0
3985db2 verified
metadata
license: cc-by-nc-4.0
task_categories:
  - tabular-classification
  - tabular-regression
language:
  - en
tags:
  - smart-factory
  - sensor-data
  - industry-4-0
  - industrial-iot
  - predictive-maintenance
  - telemetry
  - synthetic-data
  - mindweave
  - time-series
  - iot
  - test-data
  - edge-analytics
  - anomaly-detection
  - manufacturing
  - machine-monitoring
pretty_name: IoT Sensor Telemetry (Synthetic) (Free Sample)
size_categories:
  - 1K<n<10K
configs:
  - config_name: anomaly_events
    data_files: data/anomaly_events.csv
    default: true
  - config_name: sensors
    data_files: data/sensors.csv
  - config_name: telemetry_readings
    data_files: data/telemetry_readings.csv

IoT Sensor Telemetry (Synthetic) (Free Sample)

This is a free sample with 5,003 rows. The full dataset has 50,006 rows across 3 tables.

High-frequency telemetry from a simulated smart factory operating three CNC lines, a finishing cell, and a predictive-maintenance program over six months. Covers temperature, humidity, pressure, and vibration sensors sampled on a rolling 5-minute schedule with realistic shift patterns, machine assignments, maintenance alerts, and operational state changes.

Includes two injected anomalies: a progressive calibration drift on sensor 3 during month 4 and a sudden spike pattern on sensor 1 that signals an equipment failure event. Useful for time-series analytics, anomaly detection, edge telemetry pipelines, and Industry 4.0 monitoring demos.

Sample tables

Table Sample Rows
anomaly_events 2
sensors 1
telemetry_readings 5,000
Total 5,003

Full dataset

The complete dataset includes all tables with full row counts:

Table Full Rows
anomaly_events 2
sensors 4
telemetry_readings 50,000
Total 50,006

Formats included: CSV, Parquet, SQLite

Get the full dataset on Gumroad

About

Generated by Mindweave Technologies -- realistic synthetic datasets for developers, QA teams, and data engineers.

Every dataset features:

  • Enforced foreign key relationships across all tables
  • Realistic statistical distributions (not uniform random)
  • Temporal patterns (seasonal, time-of-day, day-of-week)
  • Injected anomalies for ML training and anomaly detection
  • Deterministic generation (same seed = same output)

Browse all datasets: https://mindweavetech.gumroad.com