Learning-Assisted Congestion-Aware Route Scheduling for Semiconductor Fab Material Control Systems
Abstract
Automated material handling systems in semiconductor fabs are operated by a material control system (MCS) that must schedule a relay route for every transport command online, before execution. This is a data-driven scheduling problem in which route cost is dominated in the upper tail by queueing at heterogeneous, partially observable relay equipment, so route selection requires estimating both delivery time and congestion risk at the decision moment. This paper proposes a transport-network-aware dynamic congestion representation (TN-DCR). Built on a static directed transport graph induced by historically observed relay segments, TN-DCR combines structural route priors, multi-window network-wide congestion context, route-level bottleneck exposure, and an inductive graph-aware route embedding, all constructed under a prediction-time-safety invariant that admits only information observed strictly before the prediction moment. The representation feeds separate queue- and transfer-time regressors and an ordinal multi-label classifier producing calibrated multi-threshold exceedance scores, with an empirical-Bayes stock-key residual correction reducing systematic queue-time underprediction. The predictions serve as costs in a risk-constrained route-scheduling rule that minimizes predicted delivery time subject to a bound on extreme-congestion probability, embedding the learned predictors within a lightweight operations-research decision model. In a controlled closed-loop evaluation, mean delivery time falls by 16.4\% and internal resource waiting time by 22.6\% while throughput remains essentially unchanged.
Related
- High-Probability Nash Regret for Decentralized Learning in Markov α-Potential Games: Episodic and Fully Online Asynchronous Algorithms with Applications to Markov Congestion Games
- Online Packet Scheduling with Deadlines and Learning
- Development of APOS Model Student Worksheet Assisted by GeoGebra on Line and Angle Material Representation in Mathematics in Junior High School
- Efficient Detailed Scheduling of Multiproduct Pipelines using Action-Decoupled Deep Reinforcement Learning
- BOLT: Multi-Bottleneck Bound-Guided Learning for Flexible Job-Shop Scheduling with Time Lags and Resource-Coupled Transport
BibTeX
@misc{yin2026learningassisted,
title = {{Learning-Assisted Congestion-Aware Route Scheduling for Semiconductor Fab Material Control Systems}},
author = {Hao Yin and Meiqi Tu and Anbang Liu and Shaochong Lin and Max Z. J. Shen},
howpublished = {arXiv},
year = {2026},
month = aug,
doi = {10.48550/arxiv.2608.30520},
eprint = {2608.30520},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.30520},
}
RIS
TY - UNPB
TI - Learning-Assisted Congestion-Aware Route Scheduling for Semiconductor Fab Material Control Systems
AU - Hao Yin
AU - Meiqi Tu
AU - Anbang Liu
AU - Shaochong Lin
AU - Max Z. J. Shen
PY - 2026
DA - 2026/08/31
PB - arXiv
DO - 10.48550/arxiv.2608.30520
UR - https://arxiv.org/abs/2608.30520
AB - Automated material handling systems in semiconductor fabs are operated by a material control system (MCS) that must schedule a relay route for every transport command online, before execution. This is a data-driven scheduling problem in which route cost is dominated in the upper tail by queueing at heterogeneous, partially observable relay equipment, so route selection requires estimating both delivery time and congestion risk at the decision moment. This paper proposes a transport-network-aware dynamic congestion representation (TN-DCR). Built on a static directed transport graph induced by historically observed relay segments, TN-DCR combines structural route priors, multi-window network-wide congestion context, route-level bottleneck exposure, and an inductive graph-aware route embedding, all constructed under a prediction-time-safety invariant that admits only information observed strictly before the prediction moment. The representation feeds separate queue- and transfer-time regressors and an ordinal multi-label classifier producing calibrated multi-threshold exceedance scores, with an empirical-Bayes stock-key residual correction reducing systematic queue-time underprediction. The predictions serve as costs in a risk-constrained route-scheduling rule that minimizes predicted delivery time subject to a bound on extreme-congestion probability, embedding the learned predictors within a lightweight operations-research decision model. In a controlled closed-loop evaluation, mean delivery time falls by 16.4\% and internal resource waiting time by 22.6\% while throughput remains essentially unchanged.
ER -