Evolving optimal spatial allocation policies for complex and uncertain environments

Marta Vallejo; David W. Corne; Verena Rieser

doi:10.1007/978-3-662-44440-5_21

Evolving optimal spatial allocation policies for complex and uncertain environments

Marta Vallejo^*
, David W. Corne
, Verena Rieser

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Citations (Scopus)

Abstract

Urban green spaces play a crucial role in the creation of healthy environments in densely populated areas. Agent-based systems are commonly used to model processes such as green-space allocation. In some cases, this systems delegate their spatial assignation to optimisation techniques to find optimal solutions. However, the computational time complexity and the uncertainty linked with long-term plans limit their use. In this paper we explore an approach that makes use of a statistical model which emulates the agent-based system’s behaviour based on a limited number of prior simulations to inform a Genetic Algorithm. The approach is tested on a urban growth simulation, in which the overall goal is to find policies that maximise the inhabitants’ satisfaction. We find that the model-driven approximation is effective at leading the evolutionary algorithm towards optimal policies.

Original language	English
Title of host publication	Communications in Computer and Information Science
Publisher	Springer
Pages	351-369
Number of pages	19
Volume	449
ISBN (Print)	9783662444399
DOIs	https://doi.org/10.1007/978-3-662-44440-5_21
Publication status	Published - 1 Jan 2014
Event	5th International Conference on Agents and Artificial Intelligence - Barcelona, United Kingdom Duration: 15 Feb 2013 → 18 Feb 2013

Publication series

Name	Communications in Computer and Information Science
Volume	449
ISSN (Print)	18650929

Conference

Conference	5th International Conference on Agents and Artificial Intelligence
Abbreviated title	ICAART 2013
Country/Territory	United Kingdom
City	Barcelona
Period	15/02/13 → 18/02/13

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 11 Sustainable Cities and Communities

Keywords

Agent-based model
Genetic algorithm
Green space planning
Optimisation
Statistical model
Uncertainty

ASJC Scopus subject areas

General Computer Science

Access to Document

10.1007/978-3-662-44440-5_21

Cite this

@inproceedings{5398601d79d54859becdec37a988ab32,

title = "Evolving optimal spatial allocation policies for complex and uncertain environments",

abstract = "Urban green spaces play a crucial role in the creation of healthy environments in densely populated areas. Agent-based systems are commonly used to model processes such as green-space allocation. In some cases, this systems delegate their spatial assignation to optimisation techniques to find optimal solutions. However, the computational time complexity and the uncertainty linked with long-term plans limit their use. In this paper we explore an approach that makes use of a statistical model which emulates the agent-based system{\textquoteright}s behaviour based on a limited number of prior simulations to inform a Genetic Algorithm. The approach is tested on a urban growth simulation, in which the overall goal is to find policies that maximise the inhabitants{\textquoteright} satisfaction. We find that the model-driven approximation is effective at leading the evolutionary algorithm towards optimal policies.",

keywords = "Agent-based model, Genetic algorithm, Green space planning, Optimisation, Statistical model, Uncertainty",

author = "Marta Vallejo and Corne, \{David W.\} and Verena Rieser",

year = "2014",

month = jan,

day = "1",

doi = "10.1007/978-3-662-44440-5\_21",

language = "English",

isbn = "9783662444399",

volume = "449",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "351--369",

booktitle = "Communications in Computer and Information Science",

note = "5th International Conference on Agents and Artificial Intelligence, ICAART 2013 ; Conference date: 15-02-2013 Through 18-02-2013",

}

Vallejo, M , Corne, DW & Rieser, V 2014, Evolving optimal spatial allocation policies for complex and uncertain environments. in Communications in Computer and Information Science. vol. 449, Communications in Computer and Information Science, vol. 449, Springer, pp. 351-369, 5th International Conference on Agents and Artificial Intelligence, Barcelona, United Kingdom, 15/02/13. https://doi.org/10.1007/978-3-662-44440-5_21

Evolving optimal spatial allocation policies for complex and uncertain environments. / Vallejo, Marta ; Corne, David W.; Rieser, Verena.
Communications in Computer and Information Science. Vol. 449 Springer, 2014. p. 351-369 (Communications in Computer and Information Science; Vol. 449).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Evolving optimal spatial allocation policies for complex and uncertain environments

AU - Vallejo, Marta

AU - Corne, David W.

AU - Rieser, Verena

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Urban green spaces play a crucial role in the creation of healthy environments in densely populated areas. Agent-based systems are commonly used to model processes such as green-space allocation. In some cases, this systems delegate their spatial assignation to optimisation techniques to find optimal solutions. However, the computational time complexity and the uncertainty linked with long-term plans limit their use. In this paper we explore an approach that makes use of a statistical model which emulates the agent-based system’s behaviour based on a limited number of prior simulations to inform a Genetic Algorithm. The approach is tested on a urban growth simulation, in which the overall goal is to find policies that maximise the inhabitants’ satisfaction. We find that the model-driven approximation is effective at leading the evolutionary algorithm towards optimal policies.

AB - Urban green spaces play a crucial role in the creation of healthy environments in densely populated areas. Agent-based systems are commonly used to model processes such as green-space allocation. In some cases, this systems delegate their spatial assignation to optimisation techniques to find optimal solutions. However, the computational time complexity and the uncertainty linked with long-term plans limit their use. In this paper we explore an approach that makes use of a statistical model which emulates the agent-based system’s behaviour based on a limited number of prior simulations to inform a Genetic Algorithm. The approach is tested on a urban growth simulation, in which the overall goal is to find policies that maximise the inhabitants’ satisfaction. We find that the model-driven approximation is effective at leading the evolutionary algorithm towards optimal policies.

KW - Agent-based model

KW - Genetic algorithm

KW - Green space planning

KW - Optimisation

KW - Statistical model

KW - Uncertainty

U2 - 10.1007/978-3-662-44440-5_21

DO - 10.1007/978-3-662-44440-5_21

M3 - Conference contribution

AN - SCOPUS:84916634153

SN - 9783662444399

VL - 449

T3 - Communications in Computer and Information Science

SP - 351

EP - 369

BT - Communications in Computer and Information Science

PB - Springer

T2 - 5th International Conference on Agents and Artificial Intelligence

Y2 - 15 February 2013 through 18 February 2013

ER -

Evolving optimal spatial allocation policies for complex and uncertain environments

Abstract

Publication series

Conference

UN SDGs

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this