Toward an automaton Constraint for Local Search

TL;DR

This paper introduces a novel automaton-based constraint for local search, enabling incremental violation tracking and demonstrating practical effectiveness in personnel rostering problems.

Contribution

It presents a new automaton constraint for local search that maintains violations incrementally, improving constraint handling efficiency.

Findings

Automaton constraints are practical for real-life personnel rostering.

The approach is competitive with existing methods.

Incremental violation maintenance enhances local search performance.

Abstract

We explore the idea of using finite automata to implement new constraints for local search (this is already a successful technique in constraint-based global search). We show how it is possible to maintain incrementally the violations of a constraint and its decision variables from an automaton that describes a ground checker for that constraint. We establish the practicality of our approach idea on real-life personnel rostering problems, and show that it is competitive with the approach of [Pralong, 2007].