Floor plan optimization for indoor environment based on multimodal data

Designing an optimal indoor space is challenging in interior architecture. The optimal space design requires a comprehensive analysis of the living situation of residents in a space. However, it is extremely difficult to collect data from the space where daily life occurs. Many spatial analysis sensors are required because various daily life data must be collected precisely. Hence, it is difficult for indoor space designers to use the daily-life information of users when managing indoor layouts or floor plans. In this paper, we introduce a technique to solve this problem: simple mobile application (app) logs are used to identify the daily-life patterns of users in an indoor space, and the results are used to create the optimal space layout. We collect and process key information from the mobile app logs and Google app servers to generate a high-dimensional dataset required for user behavior analysis. Subsequently, we suggest a floor plan that minimizes the living cost using a two-dimensional genetic algorithm. Our method will facilitate the spatial analysis of currently inhabited indoor space and reduce the space utilization feedback costs of users.