saidbleik/batchai_mm_ad

In this walkthrough I show how an end-to-end anomaly detection system can be implemented for IoT use cases. The scenario is intentionally kept simple for illustration purposes and to allow generalizing to different scenarios in industry. The solution is built on Microsoft's Azure stack and includes multiple cloud services that allow handling data streaming, data processing, model training/predicting, and data storage. The main component here is Batch AI, a cloud service that enables users to submit parallel jobs to a cluster of high performing virtual machines. The business problem addressed in this walkthrough is: monitoring sensor measurements of multiple devices and predicting potential anomalies that might lead to failures across these devices. A manufacturing plant has sensors attached to its machines.