A Bi-directional Ant colony algorithm for resource constrained project scheduling

The resource-constrained project scheduling problem is a typical combinatorial optimization problem. An ant algorithm with dual ant colonies is proposed to improve the effective allocation of project resources. The algorithm adaptively adjusts resource allocation according to the pheromone updated by artificial ants employed to search for feasible schedules. Two separate ant colonies are employed. The forward scheduling technique is applied by one ant colony while the backward scheduling technique is applied by another ant colony. The pheromone information of the two ant colonies is exchanged periodically to avoid early local convergence. An experimental testing indicates that the new design of two separate ant colonies with different scheduling techniques helps to improve the performance of resource constrained project scheduling.