Bayesian dynamic scheduling of multipurpose batch processes under incomplete look-ahead information
The method proposes a Bayesian dynamic scheduling approach for multipurpose batch processes facing disturbances like delays and demand changes, overcoming limitations of existing methods that assume complete information. It learns how disturbances affect operations using a Bayesian Network to guide real-time rescheduling, resulting in statistically better long-term costs and reduced system nervousness compared to periodic strategies.