Two years after the first pretrained transformers for time series — TimesFM from Google, Chronos-Bolt from Amazon, Moirai-MoE from Salesforce — we finally have real answers to the question of whether a custom XGBoost pipeline for failure prediction in MES can be replaced by a zero-shot foundation model. This article walks through it without hype: what these models were trained on, the quality they deliver on real compressor sensor data, latencies on Jetson Orin and on a GPU server, when zero-shot is enough, and when fine-tuning is unavoidable — and what this means for the classic six-month ML project in a factory.

Comparison of PyTorch, TensorFlow, XGBoost, LightGBM, and CatBoost in industrial data analysis. The article discusses the use of PyTorch in machine failure prediction, time series data analysis, and compares it with other tools in terms of flexibility, performance, and applications.

Comparison of TensorFlow, XGBoost, LightGBM, and CatBoost algorithms in fault detection in MES systems. Practical applications, time-series and tabular data analysis, and performance evaluation in terms of speed and computational resources.

Practical application of XGBoost, LightGBM, and CatBoost algorithms in MES systems for fault detection. A description of data analysis methods, machine state classification (production, failure, planned, and unplanned downtime), and real-time notifications implementation to minimize downtime.