Formulation of Sudoku Puzzle Using Binary Integer Linear Programming and Its Implementation in Julia, Python, and Minizinc

Sudoku is a number puzzle game popular among people with various difficulty levels (easy, medium, hard, and extremely hard). Sudoku can be modeled as a linear programming problem in mathematics, particularly binary integer linear programming (BILP). Completing Sudoku using BILP is quite tricky becau...

Full description

Saved in:
Bibliographic Details
Published in:Jambura journal of mathematics Vol. 4; no. 2
Main Authors: Bukhari, Fahren, Nurdiati, Sri, Najib, Mohamad Khoirun, Safiqri, Nandika
Format: Journal Article
Language:English
Published: Department of Mathematics, Universitas Negeri Gorontalo 25.06.2022
Subjects:
julia
linear programming
minizinc
python
sudoku
ISSN:2654-5616, 2656-1344
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Sudoku is a number puzzle game popular among people with various difficulty levels (easy, medium, hard, and extremely hard). Sudoku can be modeled as a linear programming problem in mathematics, particularly binary integer linear programming (BILP). Completing Sudoku using BILP is quite tricky because it requires many iterations. Therefore, this study aims to analyze the Sudoku problem using the BILP formulation and implement the problem using Julia, Python, and MiniZinc. Out of 15 cases for each difficulty level, Julia performs better than Python and MiniZinc based on computation time. Moreover, Sudoku with easy difficulty levels is solved with a longer computation time than the other three difficulty levels. The computation time for solving BILP is getting faster as the difficulty level of the Sudoku problem increases. This is because Sudoku problems with easy difficulty levels have more known values as clues and generate more constraints than other difficulty levels.
ISSN:2654-5616
2656-1344
DOI:10.34312/jjom.v4i2.14194