Abstract—Variable Neighborhood Search (VNS) is one of the most recent metaheuristics to solve optimization problems. A new variant of VNS is introduced in this article called Recursive VNS (R-VNS). The proposed R-VNS incorporates recursive methods in order to improve both the exploration and exploitation capability of the basic VNS. The experiments show that the proposed R-VNS outperforms the basic VNS by offering better solutions as well as higher convergence rate. The case study considers classical Job Shop Scheduling Problem in order to evaluate both proposed methods.
Index Terms—Job shop scheduling problem, recursive programing, variable neighborhood search.
M. R. Raeesi N. and Z. Kobti are with School of Computer Science, University of Windsor, Windsor, ON N9B 3P4 (e-mail: raeesim@uwindsor.ca, kobti@uwindsor.ca).
Cite: Mohammad R. Raeesi N. and Ziad Kobti, "Recursive Variable Neighborhood Search," International Journal of Machine Learning and Computing vol.4, no. 3, pp. 263-270, 2014.