Abstract
As industries pursue more sustainable and flexible manufacturing strategies, batch processes continue to play a vital role across a wide range of applications. Batch operations offer the ability to handle diverse feedstocks and accommodate varying product specifications. These processes are broadly used in sectors such as pharmaceuticals, specialty chemicals, food production, and bioprocesses, where precise control over reaction conditions and product quality is essential. However, maintaining optimal conditions in a batch process can be challenging due to the minimal opportunities for mid-batch interference. This work focuses on a real industrial batch process that frequently sees batches with poor yields resulting in large financial losses. Despite utilizing a mid-infrared spectrometer analyzing the batch medium in real-time, the reduced product accumulation observed in faulty batches is not evident until over a third of the batch time has passed, by which point the batch is not economically viable to abort. This study applies and compares various machine learning based fault detection strategies, including multiple Principal Component Analysis (PCA) variants and autoencoder variants, to identify faulty batches as soon as possible, since resetting reset the batches earlier can maximize overall productivity. The findings of this study offer faster and more robust fault detection than typical PCA for this industrial batch process, reducing time lost to faulty batches and improving overall productivity. This work supports the transition towards autonomous and digitalized batch manufacturing while providing an in-depth comparison between several online fault detection strategies on real industrial data.