iotsensordata.org is a practitioner-focused engineering reference for building, validating, and scaling industrial telemetry and Overall Equipment Effectiveness (OEE) pipelines — the messy, high-stakes data path that runs from a PLC register on the factory floor to a trustworthy availability number on a plant dashboard.

Every guide is written for the person who has to ship the pipeline: production-grade Python and SQL you can copy and adapt, hand-drawn architecture and data-flow diagrams, and a hard focus on the failure modes that quietly corrupt metrics in real factories — clock drift, duplicate MQTT delivery, floating-point accumulation error, out-of-order events, and shift-boundary edge cases.

Read it live → www.iotsensordata.org

Industrial engineers, IIoT developers, manufacturing data analysts, and Python automation builders who are wiring real machines to real analytics — and who need the details that tutorials skip.

The material is organised into three working tracks that mirror the pipeline itself:

Edge gateways, canonical tag dictionaries, MQTT topic design, payload serialization (JSON / Protobuf / Avro / Sparkplug B), time-series persistence, and precision limits. Turn opaque vendor addresses into a clean, ISA-95-aligned data contract before anything downstream depends on it.

Protocol normalization, clock-drift correction, outlier detection, gap filling, store-and-forward edge buffering, and async idempotent persistence — the deterministic, fault-tolerant path that keeps a broker hiccup or a gateway reboot from losing or reordering telemetry.

State machines, microstop thresholds, shift-boundary logic, OEE formula validation, and reporting rollups. Get from noisy machine-state signals to an Availability × Performance × Quality number that survives an audit.

• Decision guides — TimescaleDB vs InfluxDB vs QuestDB, Modbus TCP vs OPC UA latency, JSON vs Protobuf vs Avro, and OEE vs TEEP vs OOE.
• End-to-end recipes — a zero-data-loss edge buffer with SQLite and replay, NTP + PTP clock sync on an edge gateway, and automated OEE shift reports with Python and TimescaleDB.
• Real engineering accuracy — correct MQTT QoS semantics, ISA-95 hierarchy levels, IEEE 754 precision caveats, and tz-aware datetime handling throughout.
• Fast, accessible, offline-ready — a static site with structured data, hand-authored inline SVG diagrams, and full-page accessibility conformance.

This repository holds the full source of the site — content, templates, styles, and diagrams.

• Content lives in content/ as Markdown with front matter.
• Templates & layouts are Eleventy (Nunjucks) in _includes/, with the site model in _data/.
• Styles, scripts, and imagery are in assets/.

npm install
npm run build      # generate the static site into _site/
npm start          # serve locally with live reload

The site is a static build served from the edge via Cloudflare:

npm run deploy     # build + publish

Content and code © IoT Sensor Data. All rights reserved.

Questions, corrections, or suggestions are welcome — open an issue on this repository, or visit www.iotsensordata.org.