Constraint Logic Programming for Real-World Test Laboratory Scheduling

Authors: Tobias Geibinger, Florian Mischek, Nysret Musliu6358-6366

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this work, we show how TLSP-S can be solved by Answerset Programming extended with ideas from other constraint solving paradigms. We propose a novel and efficient encoding and apply an answer-set solver for constraint logic programs called clingcon. Additionally, we utilize our encoding in a Very Large Neighborhood Search framework and compare our methods with the state of the art approaches. Our approach provides new upper bounds and optimality proofs for several existing benchmark instances in the literature.
Researcher Affiliation Academia Tobias Geibinger, Florian Mischek, Nysret Musliu Christian Doppler Laboratory for Artificial Intelligence and Optimization for Planning and Scheduling DBAI, TU Wien, Karlsplatz 13, 1040 Vienna, Austria {tgeibing,fmischek,musliu}@dbai.tuwien.ac.at
Pseudocode No The paper provides an algorithmic description for the Very Large Neighborhood Search framework using numbered steps, but it does not present structured pseudocode or an algorithm block labeled as such.
Open Source Code No The paper states: 'All instances are available online2. 2dbai.tuwien.ac.at/staff/fmischek/TLSP' which points to dataset instances, not the source code for the methodology. No other explicit statement or link for the source code of the described method was found.
Open Datasets Yes We evaluate our model on the benchmark instances that were also used in (Geibinger, Mischek, and Musliu 2019b). This data set contains 30 generated instances containing between 7 and 401 jobs. In addition, we use the three real-world instances from (Danzinger et al. 2020). All instances are available online2. 2dbai.tuwien.ac.at/staff/fmischek/TLSP
Dataset Splits No The paper evaluates its model on 'benchmark instances' but does not specify any training, validation, or test dataset splits.
Hardware Specification Yes Our experiments were performed on a benchmark server with 224GB RAM and two AMD Opteron 6272 Processors each with a frequency of 2.1GHz and 16 logical cores.
Software Dependencies Yes As our CASP solver we use clingcon-5 (unpublished as of the writing of this article3). Our experiments were performed on a benchmark server with 224GB RAM and two AMD Opteron 6272 Processors each with a frequency of 2.1GHz and 16 logical cores.
Experiment Setup Yes Each run had a time limit of 1800 seconds. Unless noted otherwise, we used single-threaded configurations and we usually executed two independent benchmarking runs in parallel.