Smart Lighting and Student Performance: A Novel System Design, Implementation, and Effects in Classrooms

Presented by: Baoshi Sun, MASc student, Systems Design Engineering 

Abstract: As one of the most essential factors of learning environment, lighting in classroom has been found to have significant impact on student performance. Moreover, brightness level and correlated color temperature (CCT) are the two key luminous properties that have been examined in many relevant studies. And researchers were increasingly focusing on the diversity of luminous requirements under different learning context. However, knowledge regarding the optimum lighting configuration (the combination of brightness and CCT) for some specific learning context is still insufficient due to the complexity of reality, including learning context, classroom environment, demographic characteristics of students and user preferences.

To enrich the pertinent knowledge of both engineering and academia, three major works were conducted in this study. Firstly, a context-based smart lighting system was designed and implemented. This system has been tested in more than one hundred classrooms from about ten schools. It turned out to be an advanced but practical solution. Secondly, the data of a field study for examining the effect of different lighting settings on student academic performance were collected and analyzed. The field study involved twelve classrooms, 568 students of grade one and grade two from one elementary school in China. The results showed that students in context-based lighting environment significantly improved more on both Language and Mathematics than those in standard lighting environment. Interestingly, although no significant effect of gender was reported via MANOVA, the separated t-tests indicated that the lighting environment had significant effect on female, but not on male. Regarding user operation preference, it was out of expectation that no significant difference was found. Lastly, an innovative indoor environmental data-processing framework was proposed. This framework can automatically optimize lighting configuration for different learning context by gathering and analyzing a variety of classroom data and student data, including learning context, illumination settings, environmental data, student performance and some demographic information. It made it possible to shift the research practice from traditional controlled laboratory experiments to emerging Big Data and machine learning methods.

Although this study was only a preliminary work towards the best lighting settings in classrooms, it established a solid foundation (the smart lighting system) and embarked on a novel approach (the self-optimizable framework) for research in this area. Some ideas and topics for future study were also discussed.

Date and Time: November 19, 2021, 10 am to 10:45 am.

Location: MS Teams, see link below.