www.sciedu.ca/air Artificial Intelligence Research, December 2012, Vol. 1, No. 2 Published by Sciedu Press 185 ORI GI NAL RESEARCH An ABC-Genetic method to solve resource constrained project scheduling problem Vahid Zeighami 1 , Reza Akbari 2 , I smail Akbari 3 , Yevgen Biletskiy 4 1. Department of Mathematics, Shiraz University, Shiraz, Iran. 2. Department of Computer Engineering and Information Technology, Shiraz University of Technology, Shiraz, Iran. 3. Department of Electrical and Computer Engineering, University of New Brunswick, Fredericton, Canada. 4. Department of Electrical and Computer Engineering, University of New Brunswick, Fredericton, Canada. Correspondence: Vahid Zeighami. Address: Department of Mathematics, Shiraz University, Shiraz, Iran. Email: vahid.zeighami@gmail.com. Received: May 13, 2012 Accepted: August 20, 2012 Published: December 1, 2012 DOI : 10.5430/air.v1n2p185 URL: http://dx.doi.org/10.5430/air.v1n2p185 Abstract The aim of this work is to study the effect of hybridization on the performance of the Artificial Bee Colony (ABC) as a recently introduced metaheuristic for solving Resource Constrained Project Scheduling Problem (RCPSP). For this purposes, the ABC is combined with a Genetic Algorithm (GA). At the initial time, the algorithm generates a set of schedules randomly. The initial solution has been evaluated against constraints, and the infeasible solutions have been resolved to feasible ones. Then, the initial schedules are to be improved iteratively using a hybrid method until termination condition is met. The proposed method works by integrating the ABC and GA search processes. The GA method updates schedules by including the best solutions found by the ABC approach. Next, the ABC method picks the solutions found by GA search. The new methodological approach is used by the algorithm to maintain the priorities of the activities in feasible ranges. The performance of the proposed algorithm has been compared against a set of state-of-art algorithms. The simulation results have demonstrated that the proposed algorithm provides an efficient way for solving RCPSP and produce competitive results compared to other algorithms investigated in this work. Keywords Artificial bee colony, Genetic algorithm, Hybrid method, Resource constrained project scheduling problem 1 Introduction The single mode Resource Constrained Project Scheduling Problem (RCPSP) problem has been studied in this work. The RCPSP is known as a NP-hard problem [1] . Different types of algorithms ranging from exact to meta-heuristics have been proposed by researchers, and practitioners to tackle the complexities of the RCPSP problems. The exact methods have difficulties in solving large-scale RCPSP problems [2] . Hence, the use heuristic and meta-heuristic methods are needed. There are different classes of heuristics and meta-heuristics such as; Simulated Annealing (SA), Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), Genetic Algorithms (GA), Bee Algorithms (BA), etc which have been successfully used to solve RCPSP problems. The meta-heuristic methods have the ability to solve large-scale RCPSP because they provide at least one solution from the start of the algorithm. However, they may trap in local optima that result sub-optimal solutions.