Quantitative Analysis and Performance Study of Ant Colony Optimization Models Applied to Multi-mode Resource Constraint Project Scheduling Problems

Abstract

Constraint Satisfaction Problems (CSP) belongs to this kind of traditional NP-hard problems with a high impact in both, research and industrial domains. However, due to the complexity that CSP problems exhibit, researchers are forced to use heuristic algorithms for solving the problems in a reasonable time. One of the most famous heuristic al- gorithms is Ant Colony Optimization (ACO) algorithm. The possible utilization of ACO algorithms to solve CSP problems requires the de- sign of a decision graph where the ACO is executed. Nevertheless, the classical approaches build a graph where the nodes represent the vari- able/value pairs and the edges connect those nodes whose variables are different. In order to solve this problem, a novel ACO model have been recently designed. The goal of this paper is to analyze the performance of this novelty algorithm when solving Multi-Mode Resource-Constraint Satisfaction Problems. Experimental results reveals that the new ACO model provides competitive results whereas the number of pheromones created in the system is drastically reduced.

Antonio Gonzalez-Pardo

Associate Professor

Lecturer at the Computer Science Department. Main research interests are related to Computational Intelligence and Metaheuristics applied to Social Networks Analysis, and the optimization of graph-based problems.