Improved Tabu Solution for the Robust Capacitated International Sourcing Problem (RoCIS)

Abstract

AbstractIn this paper the robust capacitated international sourcing problem (RoCIS) is approached. It consists of selecting a subset of suppliers with finite capacity, from an available set of potential suppliers internationally located. This problem was introduced by González-Velarde and Laguna in [1], where they propose a deterministic solution based on tabu search memory strategies. The process consists of three stages: build an initial solution, create a neighborhood of promising solutions and perform a local search in the neighborhood. In this work we propose improving the construction of the initial solution, the construction of the neighborhood and the local search. Experimental evidence shows that the improved solution outperforms the best solutions reported for six of the considered instances, increases by 13.6% the number of best solutions found and reduces by 34% the deviation of the best solution found, respect to the best algorithm solution reported.Keywordstabu searchrobust optimizationrobust capacitated international sourcing problem