Skip redundant pieces
[an error occurred while processing this directive]


Paper 011-2007
SAS® And Sudoku
Richard A. DeVenezia, Independent Consultant
John R. Gerlach, MaxisIT, Inc.
Larry Hoyle, Institute for Policy and Social Research, Univ. of Kansas
Talbot M. Katz, Analytic Data Information Technologies
Rick Langston, SAS Institute Inc.


Sudoku puzzles were first popularized in Japan and have taken the world by storm. The puzzle consists of 81 cells arranged in a 9x9 grid, that is broken into 9 3x3 grids. Given a partially filled grid, the solver must fill in the grid so that every row, every column, and every 3x3 box contains the digits 1 through 9.

Sudoku solvers can be programmed in the SAS System using a surprising number of methods. This panel discussion will present several approaches to solving Sudoku puzzles, the use of analytic procedures like PROC CLP and PROC
LP, several Data Step approaches, and an SQL based approach. Issues that will be addressed include solution speed, ability to solve all possible puzzles, validation of puzzles, specification of puzzles - the effect of using fewer sufficient rules for the game, generation of puzzles, automating the rating of the difficulty of puzzles.

Powerpoint Slides




Please send any questions about this page to Larry Hoyle at:

[an error occurred while processing this directive]