This article presents a review of a kind of evolutionary algorithms belonging to lamarckian learning group. It begins with providing background of evolutionary algorithms. Explanation on generic tabu search is given together with reviews of its variation. The reviews cover reactive tabu search (RTS), parallel tabu search (PTS), probabilistic tabu search (PrTS), hybrid tabu search (HTS) and adaptive tabu search (ATS). Finally, the article presents a meta-heuristic approach referred to as the management agent (MA) for search. With the MA, search performance can be improved dramatically as being confirmed by our results of surface optimization problems. In this article the most important patents are also discussed.