Data-Driven Evaluation and Optimization of Office Layouts Using Beacon-Based Behavioral Monitoring of Workers

This study proposes a data-driven method for evaluating and optimizing office layouts based on worker behavioral monitoring data obtained from Bluetooth Low Energy (BLE) beacons. First, a worker behavior model considering office floor plans and furniture layouts is developed to refine location data obtained from beacons through correction and interpolation processes. Using the refined data, worker activities such as spatial movements, seating states, and interpersonal interactions are extracted. The accuracy of the activity extraction method is verified through observational surveys conducted in an actual office environment. Next, the spatiotemporal distributions of worker activities are analyzed using kernel density estimation to reveal differences in activity patterns depending on worker attributes and time periods. Based on these activity data, an office layout evaluation index is defined using indicators representing inter-department collaboration intensity and shared-space usage frequency. Finally, a genetic algorithm is applied to search for office layouts that maximize the proposed evaluation index. The results demonstrate that the proposed framework enables the quantitative evaluation and generation of efficient office layouts based on empirical behavioral data, thereby providing a practical approach for supporting data-driven workplace planning in smart office environments.