Thermal comfort prediction/optimisation in buildings using data analysis and machine learning

Study level


Topic status

We're looking for students to study this topic.


Dr Sara Omrani
Lecturer in Virtual Design & Construction
Division / Faculty
Science and Engineering Faculty


Despite the claims, recent studies show that currently used thermal comfort prediction models, such as the Predicted Mean Vote (PMV) model, may not entirely represent actual comfort level of occupants due to parameters such as cultural and environmental differences. These models, however, are the basis for control parameters of indoor thermal conditions of most buildings.

This discrepancy between occupants’ preference and building operation conditions can have several negative consequences such as high energy consumption, efficiency reduction and health issues.

This research aims to tailor thermal comfort conditions of each building based on the preference of its occupants taking into account additional parameters such as cultural and environmental conditions.

Research activities

The expected research activities for this project are:

  • data collection including occupants’ votes and thermal comfort parameters such as temperature, air velocity and humidity
  • analysis of occupant preferences in relation to indoor and outdoor thermal conditions
  • developing a predictive thermal comfort model based on the collected data.


A predictive comfort model that can be adjusted based on the occupant’s vote.

Skills and experience

To be considered for this project, it would be desirable if you have skills and/or experience in coding.


You may be able to apply for a research scholarship in our annual scholarship round.

Annual scholarship round



Contact the supervisor for more information.