An action-space-based global optimization algorithm
for packing circles into a square container
Kun He, Menglong Huang
n
, Chenkai Yang
School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China
article info
Available online 7 January 2015
Keywords:
Global optimization
Circle packing
Action space
Basin-hopping
abstract
This paper proposes an action-space-based global optimization (ASGO) approach for the problem of packing
unequal circles into a square container such that the size of the square is minimized. Starting from several
random configurati ons, ASGO runs the following pot enti al descent method and basin-hopping strat egy
iterati vel y. It finds configurati ons with the local minimum pot ential energy by the limit ed- memory BFGS
(LBFGS) algorithm, then selects the circular items having the most deformations and moves them to some
large v acant space or randomly chosen vacant space. By adapting the action space defined for the rectangular
packing problem, we approximate each cir cular item as a rectangular item, thus making it much easi er to
find compar at iv el y l arger vacant spaces for any given configur ation. The tabu strategy is used t o pr ev ent
cycling and enhance the diversification during the search procedure. Several other strategies, such as
swapping two similar circles or swapping two circles in different quadrants in the container, are combined to
increase the diversity of the configurations. We compare the performance of ASGO on 68 benchmark
instances at the Packomania website with the state-of-the-a rt results. ASGO obtains configurations with
smaller square containers on 63 instances ; at the same time it matches or appr oaches the current best
results on the other five instances.
& 201 5 Elsevier Ltd. All rights reserved.
1. Introduction
The packing problem is a well-known NP -hard problem, and it is
concerned with how to pack a certain number of circles of given radius
inside a container with no overlap. The shape of the container can be a
circle, a re ctangle, a sq uare or a poly gon, and th e it ems can be circles,
rectangles or irregular items. The packing problem has a wide range of
applications in the areas of marine transportation, motor cycle industry ,
material cutting, fashion industry, wireless communication, food
industries, etc. As an NP-har d problem, there exists no exact algorithm
for solving it to optimality in polynomial time unless P¼NP.
If the items to be packed are circles, the problem is called the
circles packing problem (CPP), which has been subject of study by a
wide spectrum of differ ent approaches in the literature. The problem
can be classified into two categories, in accordance with the items
being equal circles or unequal circles. Besides, there is an important
ext ension of the circle packing problem with eq uilibrium constraints
(CPPEC) [1,2] .
For the problem of packing equal circles, most heuristic approaches
are based on a q uasi-ph ysical or q uasi-human method. Lubachevsky
and Graham [3] proposed a billiards simulation algorithm based on
the collision forces among objects: they regarded each cycle item as a
rigid billiard and considered their movements under the collision
force. They also proved that their algorithm could obtain optimal
solutions when the number of circles n eq uals 3kðk þ1Þþ1forany
positive integer k.Grossoetal.[4] presented a genetic algorithm based
on the monotonic basin-hopping (MBH) strategy and on a population
basin-hopping str ategy. Huang and Ye [5,6] reg arded each item as an
elastic circle. They also considered the smooth movement driven by
elastic forces and the violent movement driven by strong repulsive
forces and attractive forces and proposed a quasi-phy sical global
optimization method. By adapting an improved energy-landscape-
paving method (ELP), Liu et al. [7] incorporated a new configuration
update mechanism into the ELP method.
For the pr oblem of pac king unequal circles into a larger container,
the methods can be classified into two categories: constructive
heuristics and global search heuristics. For the constructi ve heuristics,
George et al. [8] formulated this situation as a nonlinear mixed integer
programm ing problem and developed some heuristic procedur es,
including a genetic algorithm and a quasi-random techniq ue. Huang
et al. [9] defined the concept of hole degree in order to select the next
circle to place and applied a lookahead search strategy. Lü and Huang
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/ locate/caor
Computers & Operations Research
http://dx.doi.org/10.1016/j.cor.2014.12.010
0305-0548/& 2015 Elsevier Ltd. All rights reserved.
n
Corresponding author. Tel.: þ86 1520 713 7681.
E-mail addresses: brooklet60@gmail.com (K. He),
vklonghml@gmail.com (M. Huang), y_c_k@163.com (C. Yang).
Computers & Operations Research 58 (2015) 67–74