{"id":"arxiv/2608.29690","url":"https://arxiv.org/abs/2608.29690","tags":["math.OC","cs.LG"],"score":0,"title":"Transformer-Based Flow Shop Scheduling Using MILP-Generated Training Data","domain":"math.OC","source":"arxiv","authors":["Roderich Wallrath"],"reviews":0,"subject":"math.OC","summary":"Advances in machine learning (ML) have created new opportunities to complement traditional operations research (OR) methods. In particular, transformer models can capture complex interactions in token sequences by mapping tokens into a high-dimensional embedding space and propagating contextual information via attention. This makes them a candidate to model non-permutation flow shop scheduling with secondary resources as a next-token prediction task, where tokens represent job-machine-secondary resource tuples. For training, mixed-integer linear programming (MILP)-generated schedules are…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-08-30","reactions":{}}