Local Search and Constraint Programming

TL;DR: Real-world combinatorial optimization problems have two main characteristics: they are usually large, and they are not pure, i.e., they involve a heterogeneous set of side constraints, so the algorithmic approach requires flexibility, and this flexibility can be guaranteed by Constraint Programming.

Abstract: Real-world combinatorial optimization problems have two main characteristics which make them difficult: they are usually large, and they are not pure, i.e., they involve a heterogeneous set of side constraints. Hence, in most cases, exact approaches cannot be applied to solve real-world problems, whereas incomplete algorithms, and among them Local Search and Metaheuristic methods, have proved to obtain very good results in practice. Moreover, real-world applications typically lead to frequent update/addition of constraints, thus the algorithmic ap-proach requires flexibility, and this flexibility can be guaranteed by Constraint Programming.