[Paper Review] PlatEMO: A MATLAB Platform for Evolutionary Multi-Objective Optimization

Ye Tian, Ran Cheng+2 more|View|Jan 4, 2017

Advanced Multi-Objective Optimization Algorithms91 references241 citations

TL;DR

PlatEMO is a MATLAB-based open-source platform that bundles 50 MOEAs, 110 MOPs, and multiple performance indicators with a GUI for easy benchmarking and extension. It supports running experiments, result export, and straightforward incorporation of new algorithms, problems, and metrics.

ABSTRACT

Over the last three decades, a large number of evolutionary algorithms have been developed for solving multiobjective optimization problems. However, there lacks an up-to-date and comprehensive software platform for researchers to properly benchmark existing algorithms and for practitioners to apply selected algorithms to solve their real-world problems. The demand of such a common tool becomes even more urgent, when the source code of many proposed algorithms has not been made publicly available. To address these issues, we have developed a MATLAB platform for evolutionary multi-objective optimization in this paper, called PlatEMO, which includes more than 50 multi-objective evolutionary algorithms and more than 100 multi-objective test problems, along with several widely used performance indicators. With a user-friendly graphical user interface, PlatEMO enables users to easily compare several evolutionary algorithms at one time and collect statistical results in Excel or LaTeX files. More importantly, PlatEMO is completely open source, such that users are able to develop new algorithms on the basis of it. This paper introduces the main features of PlatEMO and illustrates how to use it for performing comparative experiments, embedding new algorithms, creating new test problems, and developing performance indicators. Source code of PlatEMO is now available at: http://bimk.ahu.edu.cn/index.php?s=/Index/Software/index.html.

Motivation & Objective

• Provide a comprehensive, up-to-date MATLAB platform for evolutionary multi-objective optimization (EMO).
• Enable easy benchmarking, comparison, and statistical reporting of MOEAs on a wide range of MOPs.
• Offer an open, extensible framework where researchers can add new MOEAs, MOPs, operators, and performance indicators.

Proposed method

• Incorporates over 50 MOEAs and 110 MOPs within a MATLAB-based environment.
• Uses a two-class architecture (GLOBAL and INDIVIDUAL) to decouple algorithms, problems, and operators.
• Provides a GUI and a text-based interface to configure experiments and export results (Excel/LaTeX).
• Builds on modular file structure with separate folders for Algorithms, Problems, Operators, and Metrics.
• Employs non-dominated sorting techniques (ENS-SS for 2 objectives, T-ENS for more than 2 objectives) to improve efficiency.
• Supports adding user-defined parameters via a standardized interface and documentation comments in function heads.

Experimental results

Research questions

• RQ1 Can PlatEMO provide a comprehensive, up-to-date library of MOEAs and MOPs for EMO research and benchmarking?
• RQ2 How does PlatEMO's architecture facilitate extensibility and reproducible experimentation in EMO?
• RQ3 What are the usability and performance benefits of PlatEMO's GUI compared to script-based experimentation?
• RQ4 How can researchers add new algorithms, problems, operators, and performance indicators to PlatEMO?

Key findings

• PlatEMO includes 50 MOEAs and 110 MOPs across multiple problem suites and performance indicators.
• The platform provides a GUI for one-click experiment setup and Excel/LaTeX export of statistical results.
• PlatEMO emphasizes modularity, enabling easy extension of MOEAs, MOPs, and operators with open-source MATLAB code.
• It implements efficient non-dominated sorting methods (ENS-SS for two objectives and T-ENS for more than two) to speed up evaluations.
• The architecture uses two main classes, GLOBAL and INDIVIDUAL, to decouple algorithm, problem, and operator components.