Multiobjective evolutionary algorithms: classifications, analyses, and new innovations

摘要: Abstract : This research organizes, presents, and analyzes contemporary Multiobjective Evolutionary Algorithm (MOEA) associated Optimization Problems (MOPs). Using a consistent MOEA terminology notation, each cited MOEAs' key factors are presented in tabular form for ease of identification selection. A detailed quantitative qualitative analysis is presented, providing basis conclusions about various MOEA-related issues. The traditional notion building blocks extended to the MOP domain an effort develop more effective efficient MOEAs. Additionally, community's limited test suites contain functions whose origins rationale use often unknown. Thus, using general suite guidelines appropriate function substantiated generated. An experimental methodology incorporating solution database metrics offered as proposed evaluation framework allowing absolute comparisons specific approaches. Taken together, this document's classifications, analyses, new innovations present complete, view current "state art" possible future research. Researchers with basic EA knowledge may also part it largely self-contained introduction