EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Computational Framework for Integrating Activity-Based Space Utilization Optimization and Office Layout Generation

Yoan Song, Jae Young Jang, Gisung Han, Jin-Kook Lee, Seung Hyun Cha

Journal of Computational Design and Engineering · 20.11.2026

DOI: 10.1093/jcde/qwag103 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computational Design and Engineering, online erschienen am 20.11.2026. Verfasst von Yoan Song und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Space utilization is an important metric in office design and operation. It is typically measured after occupancy, and pre-occupancy optimization is less common. Few studies have incorporated activity-based space utilization into computational office layout generation. This study proposes a computational framework that integrates activity-based space utilization optimization with constrained office layout generation. A lexicographic multi-objective mixed-integer linear programming model uses user-activity and space-type data to assign activity units to room instances. It first minimizes deviation from a target utilization range and then minimizes the difference between the highest and lowest room utilization values. Subsequent area allocation completes the room program. A spatial relationship graph built from the room program represents room affinities and node priorities and guides a Generative Design search within two fixed-depth room bands along a corridor. The framework is demonstrated in the KAIST N5 Basic Experiment & Research Building through room-count controls, demand scenarios, layout-method comparisons at matched search budgets, and sensitivity analyses. The proposed program contains 19 rooms, whereas the recorded 16-room baseline follows a different activity-mapping and load-sharing convention, so the two are compared as planning configurations rather than as equivalent alternatives: maximum modeled utilization is 74.61%, and the population standard deviation is 16.97 percentage points, compared with 132.18% and 25.07 percentage points in the baseline. These utilization differences reflect room provision, activity assignment, and load sharing, all determined before geometric placement. The comparison therefore leaves the effect of optimization alone undetermined. The selected layout satisfies the implemented geometric checks but retains two required-proximity violations, and independent evaluation of the discrete parameter grid reproduces the selected result. The framework makes the utilization consequences of activity assumptions explicit for early-stage office planning, with additional rooms and spatial reserve as the associated trade-off; effects on occupant satisfaction, operating costs, and broader architectural performance require further evaluation.

Zitieren

APA

Song, Y., Jang, J. Y., Han, G., Lee, J., & Cha, S. H. (2026). A Computational Framework for Integrating Activity-Based Space Utilization Optimization and Office Layout Generation. Journal of Computational Design and Engineering. https://doi.org/10.1093/jcde/qwag103

Vancouver

Song Y, Jang JY, Han G, Lee J, Cha SH. A Computational Framework for Integrating Activity-Based Space Utilization Optimization and Office Layout Generation. Journal of Computational Design and Engineering. 2026. doi:10.1093/jcde/qwag103

BibTeX
@article{Song2026,
  author = {Song, Yoan and Jang, Jae Young and Han, Gisung and Lee, Jin-Kook and Cha, Seung Hyun},
  title = {A Computational Framework for Integrating Activity-Based Space Utilization Optimization and Office Layout Generation},
  journal = {Journal of Computational Design and Engineering},
  year = {2026},
  doi = {10.1093/jcde/qwag103}
}
Kostenlos starten →