ETO Meets Scheduling: Learning Key Knowledge from Single-Objective Problems to Multi-Objective Problem

TL;DR

This paper introduces a novel evolutionary transfer optimization framework that learns and transfers key knowledge from single-objective scheduling problems to multi-objective problems, enhancing performance in complex combinatorial scheduling tasks.

Contribution

It is the first work to apply ETO to complex multi-objective scheduling problems by transferring knowledge from single-objective problems, specifically in permutation flow shop scheduling.

Findings

Empirical validation shows the effectiveness of the ETO-PFSP framework.

Knowledge transfer improves scheduling performance.

Potential for green and intelligent scheduling applications.

Abstract

Evolutionary transfer optimization(ETO) serves as "a new frontier in evolutionary computation research", which will avoid zero reuse of experience and knowledge from solved problems in traditional evolutionary computation. In scheduling applications via ETO, a highly competitive "meeting" framework between them could be constituted towards both intelligent scheduling and green scheduling, especially for carbon neutrality within the context of China. To the best of our knowledge, our study on scheduling here, is the 1st work of ETO for complex optimization when multiobjective problem "meets" single-objective problems in combinatorial case (not multitasking optimization). More specifically, key knowledge like positional building blocks clustered, could be learned and transferred for permutation flow shop scheduling problem (PFSP). Empirical studies on well-studied benchmarks validate relatively firm effectiveness and great potential of our proposed ETO-PFSP framework.