8,000 rows · 8 columns

Factory sensor telemetry

Eight thousand readings from twelve machines across three sites, at ten-minute intervals, with drift, seasonal temperature, injected anomalies, and gaps where a gateway dropped offline.

Download CSV 436 KB Open in Explorer Query it in SQL

What is in it, and what is wrong with it

There are two failure modes deliberately hidden in here: a slow bearing-wear drift on vibration_mm_s for two machines, and sharp single-reading spikes that are sensor glitches rather than real events. A good anomaly detector separates them. Timestamps are not evenly spaced — the gateway at site RIV drops out for stretches, so resampling before modelling is not optional.

Schema

Column	Type	Description
reading_at	timestamp	Reading time, UTC. Irregular spacing.
device_id	string	Machine identifier, e.g. MCH-04.
site	category	NOR, RIV, or SLT.
temp_c	float	Machine temperature in Celsius.
humidity_pct	float	Ambient relative humidity.
vibration_mm_s	float	RMS vibration velocity.
rpm	integer	Spindle speed.
status	category	ok, warn, or fault.

First twelve rows

reading_at	device_id	site	temp_c	humidity_pct	vibration_mm_s	rpm	status
2023-01-01 00:10:00	MCH-01	NOR	35.9	45.6	2.859	1769	ok
2023-01-01 00:20:00	MCH-02	RIV	38.37	43.7	3.208	1846	ok
2023-01-01 00:30:00	MCH-03	SLT	37.49	38	2.938	1724	ok
2023-01-01 00:40:00	MCH-04	NOR	38.09	33.1	2.413	1919	ok
2023-01-01 00:50:00	MCH-05	RIV	36.01	47	2.046	1738	ok
2023-01-01 01:00:00	MCH-06	SLT	35.98	44.9	2.036	1819	ok
2023-01-01 01:10:00	MCH-07	NOR	37.29	36.9	2.266	1687	ok
2023-01-01 01:20:00	MCH-08	RIV	36.55	37.9	2.683	1810	ok
2023-01-01 01:30:00	MCH-09	SLT	36.36	44	2.508	1749	ok
2023-01-01 01:40:00	MCH-10	NOR	37.26	42.3	2.233	1788	ok
2023-01-01 01:50:00	MCH-11	RIV	37.82	45.7	3.127	1828	ok
2023-01-01 02:00:00	MCH-12	SLT	37.28	34.2	2.724	1939	ok

Grain

One row per sensor reading.

Things to try

Resample to hourly means and handle the missing windows honestly.
Detect the drifting machines with a rolling z-score; tune the window.
Separate glitch spikes from genuine excursions.
Does ambient temperature explain any of the vibration variance?

Load it

python

import pandas as pd

df = pd.read_csv(
  "https://datanemics.com/datasets/data/sensor-telemetry.csv"
)

Licence

Public domain, CC0 1.0. Synthetic data — no attribution required.