Resumo

CABALLERO-VILLALOBOS, Juan Pablo  e  ALVARADO-VALENCIA, Jorge Andrés. Greedy Randomized Adaptive Search Procedure (GRASP). Ing. Univ. [online]. 2010, vol.14, n.2, pp.275-295. ISSN 0123-2126.

A GRASP algorithm was implemented in a common spreadsheet for single machine scheduling total weighted tardiness problem, and was tested with OR-Library instances. Results were compared with optimum or best known schedules for each instance, yielding less than 1% of difference in 93% of the cases, which results in an excellent tradeoff among results quality, computational effort and implementation easiness. Local search was performed on the post-optimization phase, based on dominancy rules, which yielded even better results with little implementation effort.

Palavras-chave : Production programming; metaheuristics; weighted tardiness problem; GRASP.

        · resumo em Português | Espanhol     · texto em Espanhol     · Espanhol ( pdf )