A multi-criteria based selection method using non-dominated sorting for genetic algorithm based design
- PDF / 4,049,103 Bytes
- 39 Pages / 439.37 x 666.142 pts Page_size
- 30 Downloads / 237 Views
A multi‑criteria based selection method using non‑dominated sorting for genetic algorithm based design Erkan Gunpinar1 · Shahroz Khan2 Received: 28 May 2019 / Revised: 20 November 2019 / Accepted: 20 November 2019 © Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract The paper presents a generative design approach, particularly for simulation-driven designs, using a genetic algorithm (GA), which is structured based on a novel offspring selection strategy. The proposed selection approach commences while enumerating the offsprings generated from the selected parents. Afterwards, a set of eminent offsprings is selected from the enumerated ones based on the following merit criteria: space-fillingness to generate as many distinct offsprings as possible, resemblance/non-resemblance of offsprings to the good/bad individuals, noncollapsingness to produce diverse simulation results and constrain-handling for the selection of offsprings satisfying design constraints. The selection problem itself is formulated as a multi-objective optimization problem. A greedy technique is employed based on non-dominated sorting, pruning, and selecting the representative solution. According to the experiments performed using three different application scenarios, namely simulation-driven product design, mechanical design and usercentred product design, the proposed selection technique outperforms the baseline GA selection techniques, such as tournament and ranking selections. Keywords Computer-aided design · Genetic algorithm · Mating selection · Multiobjective optimization
1 Introduction Optimization is the process of finding an alternative that is as fully perfect, functional, or effective as possible. A designer comes up with a new idea and tries different variations on an initial concept to improve it. However, he/she may not always anticipate all possible variations, as his/her intuition is limited. Therefore, * Erkan Gunpinar [email protected] 1
School of Mechanical Engineering, Istanbul Technical University, Inonu St. No. 65, Gumussuyu, 34437 Beyoglu, Istanbul, Turkey
2
Department of Naval Architecture, University of Strathclyde, Glasgow, UK
13
Vol.:(0123456789)
E. Gunpinar, S. Khan
an algorithm-driven design process can empower designers and achieve the desired objectives within given constraints. Genetic algorithm (GA) is an optimization technique based on the principles of genetics and natural selection. It can be employed in various engineering tasks such as design and computer-aided engineering. Starting with an initial population consisting of distinct designs and their fitness values, the population evolves under the specified selection rules. The work in this paper focuses on the selection technique of GA that is used in crossover mating. Rather than employing a probabilistic-based selection technique, as used in the baseline techniques (such as ranking and tournament selections), a systematic selection approach is employed. Offsprings generated in this way can better scan the design s
Data Loading...
