Evolutionary Multi-Criteria Optimization: Methods and Applications

Complex problems usually require the simultaneous consideration of multiple performance criteria within multidisciplinary environments. Since the mid-1990s, population-based heuristic approaches have been widely used by researchers in the field of evolutionary multi-criteria/multi-objective optimization (EMO) to address such problems. The EMO research is supported by the rapidly growing number of research publications and the availability of numerous related software tools [see the journal IEEE Transactions on Evolutionary Computation]. Recently, EMO researchers have recognized the need to develop and integrate decision-making into EMO, and the need for cross-fertilization between EMO and the multiple-criteria decision-making (MCDM) communities has become apparent.

This Special Issue aims to bring together both experts and newcomers to discuss new and existing issues in these areas, in particular to continue the integration and blending of ideas between EMO and MCDM researchers, thereby stimulating engagement with the user community. Ten papers from five countries (China, Contributions 1–7; the UK, Contribution 8; Kazakhstan, Contribution 9; Republic of Korea, Contribution 10; and Canada, Contribution 10) have been accepted for publication in this Special Issue.

Theory analysis and new algorithms are practical approaches for solving EMO problems. Combining the evolutionary algorithm with the reinforcement learning algorithm is an efficient approach for solving multi-objective optimization problems. The advantages of evolutionary algorithms and machine learning algorithms could be integrated through combining learning and optimization. A hybrid improved arithmetic optimization algorithm was proposed for solving global and engineering optimization problems in Contribution 6.

Real-world applications are vital for EMO research. A multi-objective evolutionary algorithm was proposed for optimizing standalone hybrid renewable energy system (HRES) configuration problems (Contribution 1). An improved intelligent auction mechanism was utilized to solve a multi-trip, time-dependent, dynamic vehicle routing problem with split delivery (Contribution 2). A multi-kernel learning algorithm with a multi-strategy grey wolf optimizer based on time series (MSGWO-MKL-SVM) was utilized in UAV swarm network reconstruction (Contribution 3). A bilevel optimal sizing and operation method was proposed for solving fuel cell/battery hybrid all-electric ship problems (Contribution 8). Heuristic methods for fuzzy multi-criteria decision-making were proposed to solve the problems of controlling the operating modes of the stabilization column in the primary oil-refining unit (Contribution 9). An equivalent load-based method was proposed to solve time-of-use pricing optimization problems (Contribution 4). An energy–logistics collaborative optimization method was proposed to fully tap the potential of port-integrated energy systems (Contribution 5). An improved Pareto local search-based evolutionary algorithm was proposed for solving the multi-objective shortest-path network counter-interdiction problem (Contribution 7). A multi-modal convolutional neural network integrated with an evolutionary optimization method was proposed to solve the determination of sequential well placement problems (Contribution 10). The EMO methods could be more useful with more applications.

We would like to thank several Academic Editors and the Managing Editor, Ms. Helene Hu, for their excellent assistance with this Special Issue during the entire process. We are equally grateful to all the authors for their valuable contributions and to the reviewers for their constructive suggestions that have greatly helped to improve the quality of the papers presented here. We hope that the reader finds the papers collected in this thematic Special Issue interesting and helpful.