A Local Search Modeling for Constrained Optimum Paths Problems (Extended Abstract)

TL;DR

This paper introduces a new local search modeling approach for solving Constrained Optimum Path problems, demonstrating its flexibility and effectiveness through application to the edge-disjoint paths problem with positive computational results.

Contribution

The paper presents a novel, modular local search modeling framework for COPs that enhances flexibility and ease of adding constraints, applied successfully to EDP.

Findings

Modeling is compositional and modular.

Side constraints can be easily integrated.

Computational results demonstrate approach effectiveness.

Abstract

Constrained Optimum Path (COP) problems appear in many real-life applications, especially on communication networks. Some of these problems have been considered and solved by specific techniques which are usually difficult to extend. In this paper, we introduce a novel local search modeling for solving some COPs by local search. The modeling features the compositionality, modularity, reuse and strengthens the benefits of Constrained-Based Local Search. We also apply the modeling to the edge-disjoint paths problem (EDP). We show that side constraints can easily be added in the model. Computational results show the significance of the approach.