Multi-Objective parameter-less population pyramid for solving industrial process planning problems

Abstract

Evolutionary methods are effective tools for obtaining high-quality results when solving hard practical problems. Linkage learning may increase their effectiveness. One of the state-of-the-art methods that employ linkage learning is the Parameter-less Population Pyramid (P3). P3 is dedicated to solving singleobjective problems in discrete domains. Recent research shows that P3 is highly competitive when addressing problems with so-called overlapping blocks, which are typical for practical problems. In this paper, we consider a multi-objective industrial process planning problem that arises from practice and is NP-hard.To handle it, we propose a multi-objective version of P3. The extensive research shows that our proposition outperforms the competing methods for the considered practical problem and typical multi-objective benchmarks.

Hybrid linkage learning for permutation optimization with Gene-pool optimal mixing evolutionary algorithms

“…Finally, the root of an LT contains all genes. More information and examples considering the LT creation process may be found in [25,28,34]. Other means of computing a DSM from a population are possible as well.…”

“…PFSP is an overlapping problem (all genes are dependent on each other, although the strength of the dependency may differ). For such problems, using a single LT during a whole optimization process is unfavorable [25] and may prevent an EA from reaching high-quality results. P4 is less affected by the lack of linkage diversity than LT-GOMEA because it generates and maintains a more diverse population, which may partially alleviate the lack of linkage diversity.…”

A self‐organizing weighted optimization based framework for large‐scale multi‐objective optimization

“…In recent years, multi-objective optimization problems (MOPs), consisting of two or more conflicting objectives, are commonly seen in many real-world engineering applications [1][2][3], such as power allocation [4], dynamic resource strategy [5], shop scheduling [6], chemical engineering [7] and crashworthiness design [8]. In general, there is no single solution that can optimize all the conflicting objectives simultaneously.…”