Heuristic generation via parameter tuning for online bin packing

Yarimcam, Ahmet; Asta, Shahriar; Ozcan, Ender; Parkes, Andrew J.

Abstract

© 2014 IEEE. Online bin packing requires immediate decisions to be made for placing an incoming item one at a time into bins of fixed capacity without causing any overflow. The goal is to maximise the average bin fullness after placement of a long stream of items. A recent work describes an approach for solving this problem based on a 'policy matrix' representation in which each decision option is independently given a value and the highest value option is selected. A policy matrix can also be viewed as a heuristic with many parameters and then the search for a good policy matrix can be treated as a parameter tuning process. In this study, we show that the Irace parameter tuning algorithm produces heuristics which outperform the standard human designed heuristics for various instances of the online bin packing problem.

Citation

Yarimcam, A., Asta, S., Ozcan, E., & Parkes, A. J. (2015). Heuristic generation via parameter tuning for online bin packing. In 2014 IEEE Symposium on Evolving and Autonomous Learning Systems (EALS) (102-108). https://doi.org/10.1109/EALS.2014.7009510

• Conference Name: IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - EALS 2014: 2014 IEEE Symposium on Evolving and Autonomous Learning Systems, Proceedings
• Conference Location: Orlando, FL, USA
• Start Date: Dec 9, 2014
• End Date: Dec 12, 2014
• Acceptance Date: Dec 9, 2014
• Online Publication Date: Jan 15, 2015
• Publication Date: Jan 15, 2015
• Deposit Date: Jun 27, 2016
• Publicly Available Date: Jun 27, 2016
• Peer Reviewed: Peer Reviewed
• Pages: 102-108
• Book Title: 2014 IEEE Symposium on Evolving and Autonomous Learning Systems (EALS)
• ISBN: 9781479944941
• DOI: https://doi.org/10.1109/EALS.2014.7009510
• Public URL: https://nottingham-repository.worktribe.com/output/741019
• Publisher URL: http://dx.doi.org/10.1109/EALS.2014.7009510
• Additional Information: Published in: EALS 2014: 2014 IEEE International Symposium on Evolving and Autonomous Learning Systems : proceedings. Piscataway, N.J. : IEEE, c2014, p. 102-108. ; ISBN: 9781479944958, doi: 10.1109/EALS.2014.7009510