---
title: "Merging Constraint Satisfaction Subproblems to Avoid Redundant Search"
authors:
  - "Javier Larrosa"
doi: ""
url: "https://ijcai.org/proceedings/1997/63"
pdf: "https://ijcai.org/proceedings/1997/0063.pdf"
venue: "Proceedings of the Fifteenth International Joint Conference on Artificial Intelligence"
year: 1997
pages: "424-432"
track: "Constraint Satisfaction"
keywords:
---

# Merging Constraint Satisfaction Subproblems to Avoid Redundant Search

Javier Larrosa

*Proceedings of the Fifteenth International Joint Conference on Artificial Intelligence*, pages 424-432.

## Abstract

When solving a constraint satisfaction problem we may find similar subproblems. Algorithms not exploiting this similarity are condemned to duplicate some work. In this paper we introduce a technique for merging sibling subproblems which avoids redundancy in search associated to their similarity. We show that, whenforward checking is enhanced with this capability, its performance may be increased . Experimental results on hard crossword puzzles support the practical validity of our approach.