An adaptive evolutionary multi-objective approach based on simulated annealing

A multi-objective optimization problem can be solved by decomposing it into one or more single objective subproblems in some multi-objective metaheuristic algorithms. Each subproblem corresponds to one weighted aggregation function. For example, MOEA/D is an evolutionary multi-objective optimization...

Full description

Bibliographic Details
Main Authors: Li, Hui, Landa-Silva, Dario
Format: Article
Published: MIT Press 2011
Subjects:
Online Access:https://eprints.nottingham.ac.uk/32605/
_version_ 1848794447709470720
author Li, Hui
Landa-Silva, Dario
author_facet Li, Hui
Landa-Silva, Dario
author_sort Li, Hui
building Nottingham Research Data Repository
collection Online Access
description A multi-objective optimization problem can be solved by decomposing it into one or more single objective subproblems in some multi-objective metaheuristic algorithms. Each subproblem corresponds to one weighted aggregation function. For example, MOEA/D is an evolutionary multi-objective optimization (EMO) algorithm that attempts to optimize multiple subproblems simultaneously by evolving a population of solutions. However, the performance of MOEA/D highly depends on the initial setting and diversity of the weight vectors. In this paper, we present an improved version of MOEA/D, called EMOSA, which incorporates an advanced local search technique (simulated annealing) and adapts the search directions (weight vectors) corresponding to various subproblems. In EMOSA, the weight vector of each subproblem is adaptively modified at the lowest temperature in order to diversify the search towards the unexplored parts of the Pareto-optimal front. Our computational results show that EMOSA outperforms six other well-established multi-objective metaheuristic algorithms on both the (constrained)multi-objective knapsack problemand the (unconstrained) multi-objective traveling salesman problem. Moreover, the effects of the main algorithmic components and parameter sensitivities on the search performance of EMOSA are experimentally investigated.
first_indexed 2025-11-14T19:16:21Z
format Article
id nottingham-32605
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:16:21Z
publishDate 2011
publisher MIT Press
recordtype eprints
repository_type Digital Repository
spelling nottingham-326052020-05-04T20:23:59Z https://eprints.nottingham.ac.uk/32605/ An adaptive evolutionary multi-objective approach based on simulated annealing Li, Hui Landa-Silva, Dario A multi-objective optimization problem can be solved by decomposing it into one or more single objective subproblems in some multi-objective metaheuristic algorithms. Each subproblem corresponds to one weighted aggregation function. For example, MOEA/D is an evolutionary multi-objective optimization (EMO) algorithm that attempts to optimize multiple subproblems simultaneously by evolving a population of solutions. However, the performance of MOEA/D highly depends on the initial setting and diversity of the weight vectors. In this paper, we present an improved version of MOEA/D, called EMOSA, which incorporates an advanced local search technique (simulated annealing) and adapts the search directions (weight vectors) corresponding to various subproblems. In EMOSA, the weight vector of each subproblem is adaptively modified at the lowest temperature in order to diversify the search towards the unexplored parts of the Pareto-optimal front. Our computational results show that EMOSA outperforms six other well-established multi-objective metaheuristic algorithms on both the (constrained)multi-objective knapsack problemand the (unconstrained) multi-objective traveling salesman problem. Moreover, the effects of the main algorithmic components and parameter sensitivities on the search performance of EMOSA are experimentally investigated. MIT Press 2011 Article PeerReviewed Li, Hui and Landa-Silva, Dario (2011) An adaptive evolutionary multi-objective approach based on simulated annealing. Evolutionary Computation, 19 (4). pp. 561-595. ISSN 1530-9304 multiobjective optimization simulated annealing local search combinatorial optimization http://www.mitpressjournals.org/doi/abs/10.1162/EVCO_a_00038#.Vv_I3vkrJpg doi:10.1162/EVCO_a_00038 doi:10.1162/EVCO_a_00038
spellingShingle multiobjective optimization
simulated annealing
local search
combinatorial optimization
Li, Hui
Landa-Silva, Dario
An adaptive evolutionary multi-objective approach based on simulated annealing
title An adaptive evolutionary multi-objective approach based on simulated annealing
title_full An adaptive evolutionary multi-objective approach based on simulated annealing
title_fullStr An adaptive evolutionary multi-objective approach based on simulated annealing
title_full_unstemmed An adaptive evolutionary multi-objective approach based on simulated annealing
title_short An adaptive evolutionary multi-objective approach based on simulated annealing
title_sort adaptive evolutionary multi-objective approach based on simulated annealing
topic multiobjective optimization
simulated annealing
local search
combinatorial optimization
url https://eprints.nottingham.ac.uk/32605/
https://eprints.nottingham.ac.uk/32605/
https://eprints.nottingham.ac.uk/32605/