As "a new frontier in evolutionary computation research", evolutionary transfer optimization(ETO) will overcome the traditional paradigm of zero reuse of related experience and knowledge from solved past problems in researches of evolutionary computation. In scheduling applications via ETO, a quite appealing and highly competitive framework "meeting" between them could be formed for both intelligent scheduling and green scheduling, especially for international pledge of "carbon neutrality" from China. To the best of our knowledge, our paper on scheduling here, serves as the 1st work of a class of ETO frameworks when multiobjective optimization problem "meets" single-objective optimization problems in discrete case (not multitasking optimization). More specifically, key knowledge conveyed for industrial applications, like positional building blocks with genetic algorithm based settings, could be used via the new core transfer mechanism and learning techniques for permutation flow shop scheduling problem(PFSP). Extensive studies on well-studied benchmarks validate firm effectiveness and great universality of our proposed ETO-PFSP framework empirically. Our investigations (1) enrich the ETO frameworks, (2) contribute to the classical and fundamental theory of building block for genetic algorithms and memetic algorithms, and (3) head towards the paradigm shift of evolutionary scheduling via learning by proposal and practice of paradigm of "knowledge and building-block based scheduling" (KAB2S) for "industrial intelligence" in China.

翻译：作为“进化计算研究的新前沿”，进化迁移优化（ETO）将克服进化计算研究中传统范式中对已解决问题相关经验与知识的零复用。在基于ETO的调度应用中，可构建一种极具吸引力且高度竞争的“融合”框架，同时满足智能调度与绿色调度的需求，尤其契合中国提出的“碳中和”国际承诺。据我们所知，本文关于调度的研究，是离散情形下（非多任务优化）多目标优化问题与单目标优化问题“相遇”的一类ETO框架的首项工作。具体而言，通过基于遗传算法设置的定位构建块等面向工业应用的关键知识，可借助新型核心迁移机制与学习技术用于置换流水车间调度问题（PFSP）。对经典基准问题的广泛实验验证了所提ETO-PFSP框架的经验有效性及普适性。本研究：(1) 丰富了ETO框架体系；(2) 贡献于遗传算法与模因算法中构建块的经典基础理论；(3) 通过提出并实践“基于知识与构建块的调度”（KAB2S）范式，推动了中国“工业智能”领域进化调度的范式转变。