Bounded space on-line bin packing

Best is better than first

J. Csirik, D. S. Johnson

Research output: Contribution to journalArticle

35 Citations (Scopus)

Abstract

We present a sequence of new linear-time, bounded-space, on-line bin packing algorithms, the K-Bounded Best Fit algorithms (BBFK). They are based on the Θ (n log n) Best Fit algorithm in much the same way as the Next-K Fit algorithms are based on the Θ (n log n) First Fit algorithm. Unlike the Next-K Fit algorithms, whose asymptotic worst-case ratios approach the limiting value of 17/10 from above as K → ∞ but never reach it, these new algorithms have worst-case ratio 17/10 for all K > 2. They also have substantially better average performance than their bounded-space competition, as we have determined based on extensive experimental results summarized here for instances with item sizes drawn independently and uniformly from intervals of the form (0, u], 0 <u ≤ 1. Indeed, for each u <1, it appears that there exists a fixed memory bound K(u) such that BBFK(u) obtains significantly better packings on average than does the First Fit algorithm, even though the latter requires unbounded storage and has a significantly greater running time. For u = 1, BBFK can still outperform First Fit (and essentially equal Best Fit) if K is allowed to grow slowly. We provide both theoretical and experimental results concerning the growth rates required.

Original languageEnglish
Pages (from-to)115-138
Number of pages24
JournalAlgorithmica (New York)
Volume31
Issue number2
Publication statusPublished - 2001

Fingerprint

Bin Packing
Bins
Experimental Results
Packing
Linear Time
Limiting
Data storage equipment
Interval

Keywords

  • Best Fit
  • Bin packing
  • First Fit
  • On-line algorithms

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software
  • Safety, Risk, Reliability and Quality
  • Applied Mathematics

Cite this

Bounded space on-line bin packing : Best is better than first. / Csirik, J.; Johnson, D. S.

In: Algorithmica (New York), Vol. 31, No. 2, 2001, p. 115-138.

Research output: Contribution to journalArticle

Csirik, J. ; Johnson, D. S. / Bounded space on-line bin packing : Best is better than first. In: Algorithmica (New York). 2001 ; Vol. 31, No. 2. pp. 115-138.
@article{f33631a67c33494185e5b93556f4a063,
title = "Bounded space on-line bin packing: Best is better than first",
abstract = "We present a sequence of new linear-time, bounded-space, on-line bin packing algorithms, the K-Bounded Best Fit algorithms (BBFK). They are based on the Θ (n log n) Best Fit algorithm in much the same way as the Next-K Fit algorithms are based on the Θ (n log n) First Fit algorithm. Unlike the Next-K Fit algorithms, whose asymptotic worst-case ratios approach the limiting value of 17/10 from above as K → ∞ but never reach it, these new algorithms have worst-case ratio 17/10 for all K > 2. They also have substantially better average performance than their bounded-space competition, as we have determined based on extensive experimental results summarized here for instances with item sizes drawn independently and uniformly from intervals of the form (0, u], 0 K(u) obtains significantly better packings on average than does the First Fit algorithm, even though the latter requires unbounded storage and has a significantly greater running time. For u = 1, BBFK can still outperform First Fit (and essentially equal Best Fit) if K is allowed to grow slowly. We provide both theoretical and experimental results concerning the growth rates required.",
keywords = "Best Fit, Bin packing, First Fit, On-line algorithms",
author = "J. Csirik and Johnson, {D. S.}",
year = "2001",
language = "English",
volume = "31",
pages = "115--138",
journal = "Algorithmica",
issn = "0178-4617",
publisher = "Springer New York",
number = "2",

}

TY - JOUR

T1 - Bounded space on-line bin packing

T2 - Best is better than first

AU - Csirik, J.

AU - Johnson, D. S.

PY - 2001

Y1 - 2001

N2 - We present a sequence of new linear-time, bounded-space, on-line bin packing algorithms, the K-Bounded Best Fit algorithms (BBFK). They are based on the Θ (n log n) Best Fit algorithm in much the same way as the Next-K Fit algorithms are based on the Θ (n log n) First Fit algorithm. Unlike the Next-K Fit algorithms, whose asymptotic worst-case ratios approach the limiting value of 17/10 from above as K → ∞ but never reach it, these new algorithms have worst-case ratio 17/10 for all K > 2. They also have substantially better average performance than their bounded-space competition, as we have determined based on extensive experimental results summarized here for instances with item sizes drawn independently and uniformly from intervals of the form (0, u], 0 K(u) obtains significantly better packings on average than does the First Fit algorithm, even though the latter requires unbounded storage and has a significantly greater running time. For u = 1, BBFK can still outperform First Fit (and essentially equal Best Fit) if K is allowed to grow slowly. We provide both theoretical and experimental results concerning the growth rates required.

AB - We present a sequence of new linear-time, bounded-space, on-line bin packing algorithms, the K-Bounded Best Fit algorithms (BBFK). They are based on the Θ (n log n) Best Fit algorithm in much the same way as the Next-K Fit algorithms are based on the Θ (n log n) First Fit algorithm. Unlike the Next-K Fit algorithms, whose asymptotic worst-case ratios approach the limiting value of 17/10 from above as K → ∞ but never reach it, these new algorithms have worst-case ratio 17/10 for all K > 2. They also have substantially better average performance than their bounded-space competition, as we have determined based on extensive experimental results summarized here for instances with item sizes drawn independently and uniformly from intervals of the form (0, u], 0 K(u) obtains significantly better packings on average than does the First Fit algorithm, even though the latter requires unbounded storage and has a significantly greater running time. For u = 1, BBFK can still outperform First Fit (and essentially equal Best Fit) if K is allowed to grow slowly. We provide both theoretical and experimental results concerning the growth rates required.

KW - Best Fit

KW - Bin packing

KW - First Fit

KW - On-line algorithms

UR - http://www.scopus.com/inward/record.url?scp=0242364713&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0242364713&partnerID=8YFLogxK

M3 - Article

VL - 31

SP - 115

EP - 138

JO - Algorithmica

JF - Algorithmica

SN - 0178-4617

IS - 2

ER -