Empirical validation of the spatial-choice modelling approach to user simulation

Tae Wan Kim, Seung Hyun Cha

Research output: Contribution to journalArticleResearchpeer-review

Abstract

This empirical study sought to validate the spatial-choice modelling approach to user simulation. This validation is challenging, because spatial-choice behaviour is difficult to observe in real-life settings. To overcome this challenge, the study assembled 1137 activity cases from time-use diaries kept by university students and then extracted 74 space-rejection cases (i.e. events when students were unable to take up a particular space-option and so moved to another one). We assessed the attributes of the space-options in the rejection cases by means of a site investigation, predicted space-use probabilities using spatial-choice models, and assessed forecasting accuracy by a Brier score. The spatial-choice models scored 0.35, demonstrating higher forecasting accuracy than two alternative methods, namely space-use analysis (0.50) and the closest distance method (0.84). This study contributes to user simulation by validating the accuracy of the spatial-choice models based on revealed spatial-choice data in real-life settings and comparing it with other methods.

Original languageEnglish
Pages (from-to)313-322
Number of pages10
JournalArchitectural Science Review
Volume62
Issue number4
DOIs
StatePublished - 2019 Jul 4

Fingerprint

Students

Keywords

  • User simulation
  • brier score
  • occupancy prediction
  • occupant behaviour
  • space utilization
  • spatial-choice modelling

Cite this

@article{0fd0f1e8719b43f9b8431510d23ccc07,
title = "Empirical validation of the spatial-choice modelling approach to user simulation",
abstract = "This empirical study sought to validate the spatial-choice modelling approach to user simulation. This validation is challenging, because spatial-choice behaviour is difficult to observe in real-life settings. To overcome this challenge, the study assembled 1137 activity cases from time-use diaries kept by university students and then extracted 74 space-rejection cases (i.e. events when students were unable to take up a particular space-option and so moved to another one). We assessed the attributes of the space-options in the rejection cases by means of a site investigation, predicted space-use probabilities using spatial-choice models, and assessed forecasting accuracy by a Brier score. The spatial-choice models scored 0.35, demonstrating higher forecasting accuracy than two alternative methods, namely space-use analysis (0.50) and the closest distance method (0.84). This study contributes to user simulation by validating the accuracy of the spatial-choice models based on revealed spatial-choice data in real-life settings and comparing it with other methods.",
keywords = "User simulation, brier score, occupancy prediction, occupant behaviour, space utilization, spatial-choice modelling",
author = "Kim, {Tae Wan} and Cha, {Seung Hyun}",
year = "2019",
month = "7",
day = "4",
doi = "10.1080/00038628.2019.1625299",
language = "English",
volume = "62",
pages = "313--322",
journal = "Architectural Science Review",
issn = "0003-8628",
number = "4",

}

Empirical validation of the spatial-choice modelling approach to user simulation. / Kim, Tae Wan; Cha, Seung Hyun.

In: Architectural Science Review, Vol. 62, No. 4, 04.07.2019, p. 313-322.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Empirical validation of the spatial-choice modelling approach to user simulation

AU - Kim, Tae Wan

AU - Cha, Seung Hyun

PY - 2019/7/4

Y1 - 2019/7/4

N2 - This empirical study sought to validate the spatial-choice modelling approach to user simulation. This validation is challenging, because spatial-choice behaviour is difficult to observe in real-life settings. To overcome this challenge, the study assembled 1137 activity cases from time-use diaries kept by university students and then extracted 74 space-rejection cases (i.e. events when students were unable to take up a particular space-option and so moved to another one). We assessed the attributes of the space-options in the rejection cases by means of a site investigation, predicted space-use probabilities using spatial-choice models, and assessed forecasting accuracy by a Brier score. The spatial-choice models scored 0.35, demonstrating higher forecasting accuracy than two alternative methods, namely space-use analysis (0.50) and the closest distance method (0.84). This study contributes to user simulation by validating the accuracy of the spatial-choice models based on revealed spatial-choice data in real-life settings and comparing it with other methods.

AB - This empirical study sought to validate the spatial-choice modelling approach to user simulation. This validation is challenging, because spatial-choice behaviour is difficult to observe in real-life settings. To overcome this challenge, the study assembled 1137 activity cases from time-use diaries kept by university students and then extracted 74 space-rejection cases (i.e. events when students were unable to take up a particular space-option and so moved to another one). We assessed the attributes of the space-options in the rejection cases by means of a site investigation, predicted space-use probabilities using spatial-choice models, and assessed forecasting accuracy by a Brier score. The spatial-choice models scored 0.35, demonstrating higher forecasting accuracy than two alternative methods, namely space-use analysis (0.50) and the closest distance method (0.84). This study contributes to user simulation by validating the accuracy of the spatial-choice models based on revealed spatial-choice data in real-life settings and comparing it with other methods.

KW - User simulation

KW - brier score

KW - occupancy prediction

KW - occupant behaviour

KW - space utilization

KW - spatial-choice modelling

UR - http://www.scopus.com/inward/record.url?scp=85066999296&partnerID=8YFLogxK

U2 - 10.1080/00038628.2019.1625299

DO - 10.1080/00038628.2019.1625299

M3 - Article

VL - 62

SP - 313

EP - 322

JO - Architectural Science Review

JF - Architectural Science Review

SN - 0003-8628

IS - 4

ER -