1D/2D Packaging Problem Solver

Description

This is a University assignment for solving both the 1D (Bin) and 2D (Strip) packaging problems. It is written entirely on Swift 3 and it targets MacOS 10.12.

Project Goal

The goal here is to both experiment with Swift 3, write some software for the Mac, and to try to make a tool to visualize 1D and 2D packaging problems. If you need to write algorithms for solving these kinds of problems, I hope this project can serve as a starting stone for you. It's not perfect, but feel free to fork or even submit your own pull requests.

Features

Solves both BPP and SPP problems.
Allows for easy experimentation with a choice from multiple algorithms, maximum time setting algorithm to run and some simple statistics.
Supports parallel execution of the algorithms via GCD.
Easy visualization of the best-found solution for both types of problems.
Easy to extend with new algorithms by extending the HeuristicAlgorithm class and adding an entry in the Algorithms enum.

Future work / improvements

There's clearly the need for a Solution class. Right now everything is moved as an [Int] array and the fact that the Problem class handles all the evaluation screems for a Solution class.
Problem class should get rid of the throw Exception; I implemented this early on and it's become more of a hassle than a utility.
Enable rotation of Rectangle(s); I didn't want to experiment with rotations, but I suspect others might find a need for this.
We need to open up the posibility of using different packing algorithms, allowing the same flexibility that we have for algorithm (meta-heuristic) selection.

Closing comments

Nothing more to say. Feel free to copy/edit/source this code as you see fit.

dinesharjani / sipc