This program isn't theory. It has been delivered on-site to one of the world's largest electronics manufacturers, for production and quality engineering teams working with real MES/ERP data, OEE, and inspection lines. See the anonymized delivered engagements & track record for scope and outcomes.
Most factories already sit on the data they need — MES logs, sensor streams, rejection counts, maintenance records — but it lives in manual Excel sheets and email chains. This program closes that gap. It teaches engineers to make data AI-ready, calculate the metrics that matter (OEE, MTBF, MTTR) directly from raw data, and use machine learning, predictive analytics, and computer vision to catch quality events before they happen. It complements the broader AI training catalog and corporate AI training for team-wide upskilling.
Curriculum
A hands-on 4-day program — concept-first, then live coding on real production data. Modules customise to your lines and data.
Day 1 · Demystifying AI for the Factory Floor
Topics: what AI is and is not in a manufacturing context; ML vs deep learning vs GenAI vs agentic AI (with factory examples); automation vs IoT; RAG, RPA and APIs; Industry 4.0 and 5.0 and where AI sits in smart manufacturing; the analytics maturity model — descriptive → diagnostic → predictive → prescriptive. Honest assessment of your facility's AI maturity.
Day 2 · The Data Journey & AI-Assisted Analysis
Topics: the current workflow (MES/ERP → manual Excel → email) and its failure points; what makes data AI-ready (structured, consistent, timestamped, labelled); data-quality dimensions and cleaning with Pandas and OpenRefine. Hands-on: load a production dataset in Google Colab, clean and visualise it in ~10 lines of Pandas, and auto-explain patterns with a GenAI tool — replacing a 2-hour manual Excel summary.
Day 3 · Machine Behaviour, OEE & Automated Reporting
Topics: reading machine behaviour from raw signals instead of assumptions; calculating OEE (Availability × Performance × Quality), MTBF, and MTTR in Python step by step; building a live OEE dashboard in Apache Superset; using GenAI to auto-generate narrative shift summaries and stakeholder-specific views. Hands-on: spot a machine degrading before its recorded breakdown; query multiple data sources in plain language with LangChain + a local LLM.
Day 4 · Pattern Detection, Predictive/Prescriptive Analytics & Computer Vision
Topics: temporal, spatial, and correlational patterns; clustering (K-Means) and anomaly detection with Scikit-learn; leading indicators before quality events; predictive maintenance (which machine needs attention next), prescriptive recommendations, RUL and regression; the AI tool-selection framework and when not to use AI; computer vision & OCR — AOI enhancement, solder-paste inspection, component verification, and modern open-source models (YOLO, SAM).
What Your Team Learns to Do
Concrete, job-relevant outcomes — not awareness slides
Turn raw data into decisions
- Clean and validate MES/ERP/sensor data
- Compute OEE, MTBF, MTTR in Python
- Replace manual Excel + email reporting
Predict and prevent
- Anomaly detection on production events
- Predictive maintenance & RUL basics
- Prescriptive actions supervisors will use
See with computer vision
- AOI, solder-paste & component inspection
- OCR for labels and barcodes
- Modern open-source models (YOLO, SAM)
Tools & Stack
Accessible and mostly open-source — nothing to procure to get started
Google Colab & Jupyter · Pandas · OpenRefine · Apache Superset · Scikit-learn (K-Means, anomaly detection) · Ollama with Llama 3 · LangChain · GenAI assistants (Claude / ChatGPT) · computer-vision models (YOLO, SAM). Delivered on your data where possible, or on realistic sample datasets provided.
Frequently Asked Questions
Who is it for?
Production, quality, process, and manufacturing engineers, plus plant IT and operations teams in electronics and discrete manufacturing. No prior AI or heavy programming background needed.
What does it cover?
Demystifying AI for the factory, the data journey to AI-ready information, OEE/MTBF/MTTR in Python, AI-assisted and GenAI reporting, pattern and anomaly detection, predictive and prescriptive analytics, and computer vision (AOI, solder inspection, OCR).
Is it hands-on?
Yes — concept-first, then live coding on real or realistic production data using Colab, Pandas, Superset, Scikit-learn, Ollama/Llama 3, and modern vision models. Teams build reports, dashboards, and models they can reuse.
How do we book it?
Email arjun@arjunthakur.dev or book a free intro call with your team size, roles, and goals. You'll get a proposed curriculum, format, and timeline — typically within 48 hours. On-site across India and worldwide, or live online.
Related Training
Past Trainings & Track Record →
Anonymized corporate engagements already delivered — scope, audience, and outcomes.
All Training →
Full catalog of AI and backend training programs and formats.
Corporate AI Training →
Team upskilling for engineering and L&D — build in-house capability.
Agentic AI Training →
End-to-end agent stack with LangGraph, tools, and evals.