KELAS DIMULAI DALAM
Kamis | 20.00 - 22.00 WIB
Master AI, Transform Industries
Mau implementasi AI di industri manufaktur?
📦 di Kelas ini, kamu akan belajar mengimplementasikan AI secara langsung di sektor manufaktur — dari membangun sistem prediksi perawatan mesin hingga membuat agen AI berbasis RAG untuk industri.
Kenapa Kamu Harus Mengikuti Kelas Ini?
📊 AI mulai jadi tulang punggung industri manufaktur.
• McKinsey: AI bisa tingkatkan profit hingga 39%.
• WEF: 80% manufaktur top sudah pakai AI, tapi kekurangan talenta jadi tantangan.
🚨 Engineer paham AI makin dibutuhkan — tapi masih langka.
Materi apa saja yang akan kamu pelajari?
7 Pertemuan Intensif
Kelas Interaktif
Live Coding
• Overview of AI applications in manufacturing.
• Benefits of AI in efficiency, quality, and cost reduction.
• Real-world use cases and success stories.
• Understanding predictive maintenance and its importance.
• Data sources: IoT sensors, PLC/DCS systems.
• AI techniques for predictive maintenance.
• Building a predictive maintenance pipeline.
• Role of computer vision in quality control.
• Object detection models (e.g., YOLO, Faster R-CNN).
• Segmentation models for defect detection (e.g., U-Net, Mask R-CNN, SegFormer).
• Training and deploying detection-based models.
• Overview of anomaly detection in manufacturing.
• Latest anomaly detection techniques.
• Comparison of detection-based vs. anomaly detection approaches.
• Training anomaly detection models with real-world manufacturing data.
• Importance of demand forecasting in manufacturing.
• Machine learning techniques for forecasting:
• Time series models (e.g., ARIMA, LSTM).
• Hybrid models combining time series and tabular features.
• Key metrics to evaluate forecasting accuracy.
• Introduction to Retrieval-Augmented Generation (RAG).
• Setting up a vector database (e.g., pgvector, Pinecone, FAISS).
• Chunking manufacturing knowledge data for retrieval.
• Implementing knowledge upload and retrieval pipelines.
• Integrating knowledge retrieval with SQL databases.
• Using an LLM to query and interact with SQL databases for manufacturing tasks.
• Building a system to interpret user queries, generate SQL commands, and retrieve data.
• Deploying the SQL-powered RAG agent.
Kevin Arnandes
