Solving the Manufacturing Cell Design Problem Using Human Behavior-Based Algorithm Supported by Autonomous Search

Ricardo Soto, Broderick Crawford, Francisco Gonzalez, Emanuel Vega, Carlos Castro, Fernando Paredes

Research output: Contribution to journalArticlepeer-review

8 Scopus citations

Abstract

The manufacturing Cell Design Problem (MCDP) is a classical optimization problem that finds application in lines of manufacture. The problem consist in distributing machines in cells, where the parts processed by each machine travels in the production process in such a way that productivity is improved. To solve the MCDP we employ a novel metaheuristic, which is inspired by actions, attitudes, and conducts that people normally have in life, named Human behavior-based optimization (HBBO). An individual try to evolve in life by trying his best in order to be a better human being with a brilliant future, successful at life, and be an example for others. We couple the HBBO with Autonomous Search (AS), which allows the modification of internal components of our approach when exposed to changing external forces and opportunities. We compare our HBBO-AS with the classic HBBO and an implementation using IRace, which is a software package that allows us to automatize the configuration of an algorithm through automatic configuration procedures. Additionally, in order to test the competitiveness of our results, we compare with other algorithms proved to perform well solving the MCDP. We illustrate experimental results, where the proposed approach is able to obtain interesting performance and robustness in the 125 well-known instances of the MCDP.

Original languageEnglish
Article number8827495
Pages (from-to)132228-132239
Number of pages12
JournalIEEE Access
Volume7
DOIs
StatePublished - 2019

Keywords

  • Autonomous search
  • HBBO
  • manufacturing cell design problem
  • metaheuristics

Fingerprint

Dive into the research topics of 'Solving the Manufacturing Cell Design Problem Using Human Behavior-Based Algorithm Supported by Autonomous Search'. Together they form a unique fingerprint.

Cite this