Efficient Batch Assignment for Parallel-Machine Production Scheduling

Christoph Fünfzig, Roderich Wallrath, Stefan Hubert

2024

Abstract

In this article, we consider different batch assignment schemes for non-separable, non-preemptive production scheduling on m parallel work stations. Batch assignment is a very important part of production models as batches with identical parameters usually occur in a large number in real-world applications, which causes a large number of symmetric solutions. The common scheme Exactly-n for assignment of n batches to m machines originating from general assignment problems is very inefficient when it comes to scheduling, even with additional ordering for symmetry breaking (Exactly-n Ordered). We define three restricted assignment schemes for the same-parameter batches by forming blocks of size 0 ≤ zi ≤ n on machine i and ordering them between the different machines. We compute bounds for the number of feasible assignments as a measure for the feasible space and give solving times from our experiments with a boolean inference-based solver like the Google CP-SAT solver. We show that with the proposed restricted assignment schemes, production scheduling models result that solve significantly faster than the models with the common scheme Exactly-n.

Paper Citation

in Harvard Style

Fünfzig C., Wallrath R. and Hubert S. (2024). Efficient Batch Assignment for Parallel-Machine Production Scheduling. In Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART; ISBN 978-989-758-680-4, SciTePress, pages 268-274. DOI: 10.5220/0012320800003636

in Bibtex Style

@conference{icaart24,
author={Christoph Fünfzig and Roderich Wallrath and Stefan Hubert},
title={Efficient Batch Assignment for Parallel-Machine Production Scheduling},
booktitle={Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART},
year={2024},
pages={268-274},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012320800003636},
isbn={978-989-758-680-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART
TI - Efficient Batch Assignment for Parallel-Machine Production Scheduling
SN - 978-989-758-680-4
AU - Fünfzig C.
AU - Wallrath R.
AU - Hubert S.
PY - 2024
SP - 268
EP - 274
DO - 10.5220/0012320800003636
PB - SciTePress