Research

Study of learning environment simulation

• Principle Investigator  Jin Lee (Master's thesis, Inha University)

• Advisor Professor Seung Wan Hong, Inha University

• Related Publication Lee, J., Hong, S. W., & Cho, C. Y. (2024). Reinforcement-learning agents for architects' trade-offs in designing children's play environment: A qualitative comparative analysis. Design Studies, 91, 101248. https://doi.org/10.1016/j.destud.2024.101248

• Funding details  This thesis was supported by the National Research Foundation of Korea (NRF) grants funded by the Ministry of Education (Grant Number NRF-2021R1A2C1004608), and the Korea Agency for Infrastructure Technology Advancement(KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant Number 22AATD-C163269-02). 

This study is interdisciplinary research based on the intersection of the three circles: (1) Architectural Design, (2) Reinforcement-Learning Powered Agent-based Simulation, and (3) Children’s Learning Environment. A comparative experiment is conducted with fourteen professional architects to investigate the empirical effects of the RL agents-powered simulation. The participants produced fourteen children’s learning environment designs using habitual design methods (without simulation) and the children’s play behavior simulation model (with simulation). 

Based on the interview with the survey and recorded video data,  the study analyzed the experiment design results in terms of (1) design trade-offs, (2) motivation, and (3) quantifiable performance. Eventually, the study found that the RL-powered agent-based simulation enables the architects to shift their perspective from adult to child-centric and ultimately optimize holistic design performance.

In the first section, the participants performed the kindergarten design task with their habitual methods, including 3D modeling, diagramming, and sketching. In the second section, they explored their initial design using the simulation and modified it based on the observation. In the experiments, the designers and the operators were separated to let the participants focus on their design processes.