Systematic construction of algorithm portfolios for a maintenance scheduling problem

Ahmad Almakhlafi; Joshua Knowles

doi:10.1109/CEC.2013.6557577

Systematic construction of algorithm portfolios for a maintenance scheduling problem

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

We investigate how combinations of evolutionary algorithms (portfolios) can be constructed efficiently for solving optimization problem instances drawn from a distribution. We consider selection methods ranging in intricacy and based on different principles including a technique for efficient tuning of metaheuristics, a racing algorithm. The selectors are used here to optimize instances of the Preventive Maintenance Scheduling Problem (PMSP) in power generation. Experiments show different behaviors of selectors in term of computational time needed to choose constituent algorithms and the performance of the generated portfolios at optimizing previously unseen PMSP instances. The racing selector offers a good trade-off between the computational time and the performance. © 2013 IEEE.

Original language	English
Title of host publication	2013 IEEE Congress on Evolutionary Computation, CEC 2013\|IEEE Congr. Evol. Comput., CEC
Publisher	IEEE
Pages	245-252
Number of pages	7
ISBN (Print)	9781479904549
DOIs	https://doi.org/10.1109/CEC.2013.6557577
Publication status	Published - 2013
Event	2013 IEEE Congress on Evolutionary Computation, CEC 2013 - Cancun, Mexico Duration: 20 Jun 2013 → 23 Jun 2013

Conference

Conference	2013 IEEE Congress on Evolutionary Computation, CEC 2013
Country/Territory	Mexico
City	Cancun
Period	20/06/13 → 23/06/13

Keywords

Algorithm Portfolio
Algorithm Selection
Genetic Algorithms
Maintenance Scheduling
Racing Algorithm

Access to Document

10.1109/CEC.2013.6557577

Cite this

@inproceedings{b1a2d4aa80ef4a36ad9103c0cfb24ccd,

title = "Systematic construction of algorithm portfolios for a maintenance scheduling problem",

abstract = "We investigate how combinations of evolutionary algorithms (portfolios) can be constructed efficiently for solving optimization problem instances drawn from a distribution. We consider selection methods ranging in intricacy and based on different principles including a technique for efficient tuning of metaheuristics, a racing algorithm. The selectors are used here to optimize instances of the Preventive Maintenance Scheduling Problem (PMSP) in power generation. Experiments show different behaviors of selectors in term of computational time needed to choose constituent algorithms and the performance of the generated portfolios at optimizing previously unseen PMSP instances. The racing selector offers a good trade-off between the computational time and the performance. {\textcopyright} 2013 IEEE.",

keywords = "Algorithm Portfolio, Algorithm Selection, Genetic Algorithms, Maintenance Scheduling, Racing Algorithm",

author = "Ahmad Almakhlafi and Joshua Knowles",

note = "(Accepted 23/4/2013); 2013 IEEE Congress on Evolutionary Computation, CEC 2013 ; Conference date: 20-06-2013 Through 23-06-2013",

year = "2013",

doi = "10.1109/CEC.2013.6557577",

language = "English",

isbn = "9781479904549",

pages = "245--252",

booktitle = "2013 IEEE Congress on Evolutionary Computation, CEC 2013|IEEE Congr. Evol. Comput., CEC",

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - Systematic construction of algorithm portfolios for a maintenance scheduling problem

AU - Almakhlafi, Ahmad

AU - Knowles, Joshua

N1 - (Accepted 23/4/2013)

PY - 2013

Y1 - 2013

N2 - We investigate how combinations of evolutionary algorithms (portfolios) can be constructed efficiently for solving optimization problem instances drawn from a distribution. We consider selection methods ranging in intricacy and based on different principles including a technique for efficient tuning of metaheuristics, a racing algorithm. The selectors are used here to optimize instances of the Preventive Maintenance Scheduling Problem (PMSP) in power generation. Experiments show different behaviors of selectors in term of computational time needed to choose constituent algorithms and the performance of the generated portfolios at optimizing previously unseen PMSP instances. The racing selector offers a good trade-off between the computational time and the performance. © 2013 IEEE.

AB - We investigate how combinations of evolutionary algorithms (portfolios) can be constructed efficiently for solving optimization problem instances drawn from a distribution. We consider selection methods ranging in intricacy and based on different principles including a technique for efficient tuning of metaheuristics, a racing algorithm. The selectors are used here to optimize instances of the Preventive Maintenance Scheduling Problem (PMSP) in power generation. Experiments show different behaviors of selectors in term of computational time needed to choose constituent algorithms and the performance of the generated portfolios at optimizing previously unseen PMSP instances. The racing selector offers a good trade-off between the computational time and the performance. © 2013 IEEE.

KW - Algorithm Portfolio

KW - Algorithm Selection

KW - Genetic Algorithms

KW - Maintenance Scheduling

KW - Racing Algorithm

U2 - 10.1109/CEC.2013.6557577

DO - 10.1109/CEC.2013.6557577

M3 - Conference contribution

SN - 9781479904549

SP - 245

EP - 252

BT - 2013 IEEE Congress on Evolutionary Computation, CEC 2013|IEEE Congr. Evol. Comput., CEC

PB - IEEE

T2 - 2013 IEEE Congress on Evolutionary Computation, CEC 2013

Y2 - 20 June 2013 through 23 June 2013

ER -

Systematic construction of algorithm portfolios for a maintenance scheduling problem

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this