期刊名称：International Journal of Computer Information Systems and Industrial Management Applications
印刷版ISSN：2150-7988
电子版ISSN：2150-7988
出版年度：2013
卷号：5
页码：704-713
出版社：Machine Intelligence Research Labs (MIR Labs)
摘要：this paper presents the metrics used for measuring performance of optimization algorithms in dynamic environment. Real world is dynamic in nature so problems in it are mostly dynamic where the optimal solution changes over time. Algorithms proposed for solving such problems must be able to adapt change and produce enough diversity to locate the new or changed optimal solution. Evolutionary algorithms are more suitable to find solutions to dynamic problems as compared to Classical Optimization techniques which sequentially search for the solution and are based on differential equations. Evolutionary algorith ms evolve with each generation and can find multiple optima in parallel. Different performance metrics have been used to measure the performance of multi modal and multi objective optimization algorithms in static environment. However, performance measurement in dynamic environment is difficult and complex as traditional performance metrics of static environment like mean fitness, best current fitness, and speed of convergence are not relevant in dynamic environment as fitness of the optima may change over time. A number of techniques have been proposed to find solutions to multi-modal and multi objective problems in dynamic environment. The paper explores the common performance metrics and experimental framework used to measure the performance of algorithms in dynamic environment.
