Disk packing problem, leveraging AMPL and Matlab.
Coursework in Optimization Models for Big Data 2016/2017
The disk packing problem is a well-known NP-hard problem. It consists in the maximization of the radius and number of non-overlapping disks in the unit square. In this work, we propose two equivalent formulation of the disk packing problem.
Run the standard formulation:
$disk-packing> ampl disk-packing.run
View the output:
$disk-packing> cat out/out-[N]_res.txt
where [N] is the number of disks.
Visualize the result executing the Matlab script disk-packing/disk-packing.m
.
Run the equivalent formulation:
$disk-packing-equiv> ampl disk-packing-equiv.run
View the output:
$disk-packing-equiv> cat out/out-[N]_res.txt
where [N] is the number of disks.
Visualize the result executing the Matlab script disk-packing-equiv/disk-packing-equiv.m
.
Giacomo Marciani, gmarciani@acm.org
Michele Porretta, mporretta@acm.org
The project is released under the MIT License.