Meta-heuristic and Evolutionary Algorithms for Engineering Optimization

7 Ant Colony Optimization

Summary

This chapter describes ant colony optimization (ACO). The basic concepts of the ACO are derived from analogy to the foraging behavior of ants. The chapter begins with a brief literature review highlighting the development and applications of the ACO. This is followed by a description of the ACO’s algorithm. A pseudocode of the ACO closes the chapter.

7.1 Introduction

Ant colony optimization (ACO) was introduced by Dorigo et al. (1991, 1996). It attempts to simulate in algorithmic fashion the foraging behavior of ants. Several varieties of ACO algorithms have appeared since its original inception, and those include the elitist ant system (AS) (Dorigo, 1992; Dorigo et al., 1996), Ant‐Q (Gambardella and Dorigo, 1995), ant colony system (Gambardella and Dorigo, 1996; Dorigo and Gambardella, 1997), max–min AS (Stutzle and Hoos, 2000), and the hypercube AS (Blum and Dorigo, 2004). The ACO has solved various types of problems such as vehicle routing (Reimann et al., 2004), project scheduling (Merkle et al., 2002), and open shop scheduling (Blum, 2005). Various types of ant‐based algorithms have found frequent implementations in civil engineering and structural optimization (Christodoulou, 2010; Lee, 2012; Sharafi et al., 2012). Abadi and Jalili (2006) applied the ACO for network vulnerability analysis. Effatnejad et al. (2013) implemented the ACO for determining the feasible optimal solution of economic dispatching. Afshar et al. (2015) wrote ...