Cozie & Oura thermal comfort study

Research Project, 2024

Personal comfort models represent a significant advancement in the realm of thermal comfort research, moving away from generic, one-size-fits-all models like PMV (Predicted Mean Vote)
and adaptive models. The key distinction lies in their individual-centric approach. Unlike traditional models that rely on averaged responses from a group, personal comfort models are tailored to
each participant, creating a unique model that is both trained and tested on an individual basis.
This approach ensures a more accurate and personalized understanding of thermal comfort. Despite their individualistic nature, when aggregated, these models can effectively predict the
thermal preferences of larger groups sharing the same environment. This capability is crucial for designing comfortable and energy-efficient buildings and public spaces. One of the most innovative aspects of personal comfort models is their flexibility in terms of data
input. They can harness a wide variety of data sources, including environmental parameters and onboard sensors found in wearable devices and smartphones. This multi-dimensional data
gathering enables a comprehensive analysis of factors affecting an individual's thermal comfort.

However, developing these models is not without challenges. One of the primary obstacles is the need for continuous and reliable feedback from participants. To address this, the Cozie project was initiated. Cozie is an open-source application designed for smartwatches, including Fitbit (Versa 2 and Ionic) and Apple smartwatches. It facilitates the collection of real-time data by allowing participants to conveniently complete 'Right-Here-Right-Now' surveys directly from their smartwatches. This method, known as ecological momentary assessment, is a significant leap forward in data collection for thermal comfort research.
The collected data through Cozie is not limited to subjective feedback. It can be enriched with environmental data from wireless sensors and physiological data from the smartwatch itself, offering a holistic view of the factors influencing thermal comfort.
The planned longitudinal field study aims to further explore and validate the effectiveness of personal comfort models. By engaging 10-15 participants in a 6-month study, spanning spring
and autumn, and utilizing the Cozie app for real-time data collection, the study seeks to gather comprehensive and diverse data. This will not only test the robustness of the personal comfort models but also provide valuable insights into seasonal variations in thermal comfort preferences. In summary, personal comfort models, supported by innovative data collection methods like Cozie, represent a significant step forward in understanding and catering to individual thermal
comfort needs. By combining individual feedback with a wide array of environmental and physiological data, these models have the potential to revolutionize the way we approach thermal comfort in various settings.