Category: Smart Factory with AI
-
Building an End-to-End Smart Factory AI Pipeline: Case Study and Best Practices
Real-world smart factory AI pipeline: from thermal throttling on edge devices to RL schedulers that violated social constraints. What actually works.
-
Explainable AI for Factory Operations: Building Trust in Automated Decisions
SHAP vs decision trees for factory AI: one gives impressive heatmaps, the other earns operator trust. Real code, real vibration data, real deployment results.
-
Sensor Fusion and IoT Integration in Smart Manufacturing
Comparing weighted averaging vs temporal fusion transformers for sensor fusion in smart factories. Includes LSTM implementation and edge deployment.
-
Object Detection and Tracking: Monitoring Assembly Line Workflows
Comparing YOLOv8 + centroid tracking vs DeepSORT for real-time assembly line monitoring. Which one actually runs on edge devices?
-
Time Series Forecasting for Demand Planning: Why Prophet Fails in Factories
Why Prophet fails in factories and how multi-horizon LightGBM with quantile regression solves demand forecasting for manufacturing.
-
Reinforcement Learning for Production Scheduling: Why DQN Failed and How PPO Saved Our Factory Line
DQN failed on factory scheduling. PPO barely worked. Here's the hybrid RL+heuristic approach that cut tardiness by 40% in production.
-
Digital Twin Technology: Creating Virtual Factory Replicas with Python
Build a digital twin for factory production lines using Python, SimPy, and Kalman filters โ from simulation kernel to live sensor synchronization and what-if analysis.