Document Type
Conference Proceeding
Publication Title
BuildSys 2017 - Proceedings of the 4th ACM International Conference on Systems for Energy-Efficient Built Environments
Abstract
In this poster we introduce the Commercial Occupancy Dataset (COD), a high-resolution long-term dataset of occupancy traces in a commercial office building spanning 9 months and covering zone-level occupancy for five different spaces containing more than 90,000 enter/exit events over this time period. Occupancy data in a building contains rich spatial-temporal information about the users and their usage of the space and facilities. However, obtaining accurate occupancy data is a very challenging task due to the limitation of existing sensing technologies. A novel depth-imaging based solution to estimate occupancy counts was deployed in five doorways of an office building to generate the dataset. The dataset has high resolutions in all three major dimensions: temporal, spatial and occupancy state. This allows applications such as building energy simulation, occupancy modeling and human-in-the-loop HVAC control which enhance energy efficiency and human comfort. Towards this objective, we provide a case study to demonstrate the utility of understanding occupancy.
DOI
10.1145/3137133.3141471
Publication Date
11-8-2017
Recommended Citation
Liu, Kin Sum; Francis, Jonathan; Pinto, Elvin Vindel; and Shelton, Charles, "Poster abstract: COD - A dataset of commercial building occupancy traces" (2017). University Faculty Publications and Creative Works. 241.
https://digitalcommons.calvin.edu/calvin_facultypubs/241