Evolutionary multiobjective optimization: principles, procedures, and practices

Abstract

Multi-objective optimization problems deal with multiple conflicting objectives. In principle, they give rise to a set of trade-off Pareto-optimal solutions. Over the past one-and-half decade, evolutionary multi-objective optimization (EMO) has established itself as a mature field of research and application with an extensive literature, commercial softwares, numerous freely downloadable codes, a dedicated biannual conference running successfully five times so far since 2001, special sessions and workshops held at all major evolutionary computing conferences, and full-time researchers from universities and industries from all around the globe. This is because evolutionary algorithms (EAs) work with a population of solutions and in solving multi-objective optimization problems, EAs can be modified to find and capture multiple solutions in a single simulation run. In this article, we make a brief outline of EMO principles, discuss one specific EMO algorithm, and present some current research issues of EMO.