Evolutionary local search for solving the office space allocation problem

Office Space Allocation (OSA) is the task of correctly allocating the spatial resources of an institution to a set of entities by minimising the wastage of space and the violation of additional constraints. In this paper, an evolutionary local search algorithm is presented to tackle this problem. Th...

Full description

Bibliographic Details
Main Authors: Ulker, Ozgur, Landa-Silva, Dario
Format: Conference or Workshop Item
Published: IEEE press 2012
Subjects:
Online Access:https://eprints.nottingham.ac.uk/32128/
_version_ 1848794340033298432
author Ulker, Ozgur
Landa-Silva, Dario
author_facet Ulker, Ozgur
Landa-Silva, Dario
author_sort Ulker, Ozgur
building Nottingham Research Data Repository
collection Online Access
description Office Space Allocation (OSA) is the task of correctly allocating the spatial resources of an institution to a set of entities by minimising the wastage of space and the violation of additional constraints. In this paper, an evolutionary local search algorithm is presented to tackle this problem. The evolutionary components of the algorithm include standard crossover and mutation operators and a relatively small population of individuals. The offspring produced by the evolutionary operators are subjected to a short but intense local search process. A very fast cost calculation method tailored for searching a large section of the search space is implemented. Extensive experimentation is carried out related to several parameters of the algorithm: the mutation rate, the population size, the length of the local search procedure after each mutation, hence the balance between the evolutionary and the local search stages, and the level of greediness of the local search process. The final results on 72 different data instances show that this hybrid evolutionary algorithm is very competitive with an integer programming model.
first_indexed 2025-11-14T19:14:38Z
format Conference or Workshop Item
id nottingham-32128
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:14:38Z
publishDate 2012
publisher IEEE press
recordtype eprints
repository_type Digital Repository
spelling nottingham-321282020-05-04T20:21:39Z https://eprints.nottingham.ac.uk/32128/ Evolutionary local search for solving the office space allocation problem Ulker, Ozgur Landa-Silva, Dario Office Space Allocation (OSA) is the task of correctly allocating the spatial resources of an institution to a set of entities by minimising the wastage of space and the violation of additional constraints. In this paper, an evolutionary local search algorithm is presented to tackle this problem. The evolutionary components of the algorithm include standard crossover and mutation operators and a relatively small population of individuals. The offspring produced by the evolutionary operators are subjected to a short but intense local search process. A very fast cost calculation method tailored for searching a large section of the search space is implemented. Extensive experimentation is carried out related to several parameters of the algorithm: the mutation rate, the population size, the length of the local search procedure after each mutation, hence the balance between the evolutionary and the local search stages, and the level of greediness of the local search process. The final results on 72 different data instances show that this hybrid evolutionary algorithm is very competitive with an integer programming model. IEEE press 2012-06 Conference or Workshop Item PeerReviewed Ulker, Ozgur and Landa-Silva, Dario (2012) Evolutionary local search for solving the office space allocation problem. In: 2012 IEEE Congress on Evolutionary Computation (CEC 2012), 10-15 June 2012, Brisbane, Australia. Space allocation Hybrid evolutionary algorithms Hybrid metaheuristics Local search http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6253009
spellingShingle Space allocation
Hybrid evolutionary algorithms
Hybrid metaheuristics
Local search
Ulker, Ozgur
Landa-Silva, Dario
Evolutionary local search for solving the office space allocation problem
title Evolutionary local search for solving the office space allocation problem
title_full Evolutionary local search for solving the office space allocation problem
title_fullStr Evolutionary local search for solving the office space allocation problem
title_full_unstemmed Evolutionary local search for solving the office space allocation problem
title_short Evolutionary local search for solving the office space allocation problem
title_sort evolutionary local search for solving the office space allocation problem
topic Space allocation
Hybrid evolutionary algorithms
Hybrid metaheuristics
Local search
url https://eprints.nottingham.ac.uk/32128/
https://eprints.nottingham.ac.uk/32128/