Exploration-exploitation tradeoffs in metaheuristics: Survey and analysis

Metaheuristics (MHs) have been established as a family of the most practical approaches to hard optimization problems. Metaheuristic (MH) algorithm is a high-level problem-independent algorithmic framework that provides a set of guidelines or strategies to develop heuristic optimization algorithms. Many different kinds of MHs (e.g. genetic algorithms, tabu search, simulated annealing etc) were proposed during last several decades. Most of MHs focused on experimental studies and applications. It is well known that a suitable and reasonable tradeoff between exploration and exploitation (T: Er& Ei) is crucial for their success, and having a great effect on global optimization performance, e.g., accuracy and convergence speed of those algorithms. But rigid and useful theoretical study is rare up to date. A systematic analysis and a detailed survey of this problem were presented in this paper. From a system's perspective, it shows that combining MHs with problem instances' key properties, algorithm characters and human intelligence is a right way to deal with this difficulty.