Title: A VNS-GA-based hybrid metaheuristics for resource constrained project scheduling problem
Authors: Kanchan Joshi; Karuna Jain; Vijay Bilolikar
Addresses: School of Management, Indian Institute of Technology Bombay, Mumbai, India ' School of Management, Indian Institute of Technology Bombay, Mumbai, India ' Father Agnel's College of Engineering, Mumbai, India
Abstract: This paper presents hybrid metaheuristics for resource constrained project scheduling problem to minimise makespan. The hybrid metaheuristics is combination of variable neighbourhood search (VNS) and genetic algorithm (GA). Variable neighbourhood search, a local search metaheuristics strengthens the exploration process of GA. The chromosome of the population pool, generated using priority rule-based heuristics and GA operators, are processed using VNS. The schedules from GA and VNS together form the next generation population pool. Thus the hybrid mechanism provides a combination of exploration and exploitation process to achieve the desired objective. The solution scheme is tested using Kolisch library J30 and J60 datasets for multiple resource types and results are compared with the best available solutions. The computational experimentation presents the performance of the proposed hybrid metaheuristics.
Keywords: variable neighbourhood search; VNS; genetic algorithms; resource constraints; project scheduling; hybrid metaheuristics; project management.
International Journal of Operational Research, 2016 Vol.27 No.3, pp.437 - 449
Received: 23 Dec 2013
Accepted: 03 Sep 2014
Published online: 06 Sep 2016 *