SimILS: a simulation-based extension of the iterated local search metaheuristic for stochastic combinatorial optimization

Abstract

Iterated Local Search (ILS) is one of the most popular single-solution-based metaheuristics. ILS is recognized by many authors as a relatively simple yet efficient framework able to deal with complex combinatorial optimization problems (COPs). ILS-based algorithms have been successfully applied to provide near-optimal solutions to different COPs in logistics, transportation, production, etc. However, ILS is designed to solve COPs under deterministic scenarios. In some real-life applications where uncertainty is present, the deterministic assumption makes the model less accurate since it does not reflect the real stochastic nature of the system. This paper presents the SimILS framework that extends ILS by integrating simulation to be able to cope with Stochastic COPs in a natural way. The paper also describes several tested applications that illustrate the main concepts behind SimILS and give rise to a new brand of ILS-based algorithms.