Play 68
Predictive Maintenance AI
High✅ Ready
IoT-driven equipment failure prediction with RUL estimation and maintenance scheduling.
Predictive maintenance platform that ingests IoT sensor data via Azure IoT Hub, applies anomaly pattern recognition through Stream Analytics, and uses Azure Machine Learning for remaining useful life (RUL) estimation. GPT generates maintenance window optimization recommendations, spare parts forecasts, and technician dispatch priorities. Built for manufacturing, energy, and infrastructure verticals where unplanned downtime costs millions.
Architecture Pattern
IoT streaming: ML-based RUL prediction, anomaly detection, AI maintenance scheduling
Azure Services
Azure IoT HubAzure OpenAIAzure Machine LearningStream AnalyticsCosmos DBAzure Monitor
DevKit (.github Agentic OS)
- agent.md — root orchestrator with builder→reviewer→tuner handoffs
- 3 agents — Maintenance Builder (gpt-4o), Reviewer (gpt-4o-mini), Tuner (gpt-4o-mini)
- 3 skills — deploy (235 lines), evaluate (111 lines), tune (169 lines)
- 4 prompts — /deploy, /test, /review, /evaluate with agent routing
- .vscode/mcp.json — FrootAI MCP with OpenAI + IoT Hub inputs + envFile
TuneKit (AI Config)
- config/openai.json — gpt-4o for recommendations, mini for sensor triage
- config/guardrails.json — reliability focus, alert thresholds
- evaluation/eval.py — RUL accuracy >85%, False alarm rate <10%
Tuning Parameters
Failure probability thresholdSensor window minutesRUL confidence levelDispatch priority weightsAnomaly lookback hours
Estimated Cost
Dev/Test
$120–300/mo
Production
$3K–15K/mo