An AI-Assisted Qualitative Analysis of Energy Consumption Awareness and Usage Practices

Our experiment shows that the primary reason for students electricity usage is the actual behavior and not necessarily the technology. There is a clear awareness-behavior gap the students know what the behaviors are but fail to adjust their behavior. Our rule based system was able to predict electricity usage and classify the users, providing a range of actionable recommendations that were feasible and affordable for the students to implement without any complex devices or models. The use of dashboards with visualizations helped to develop better understanding of electricity usage and encouraged awareness and saving behaviors among students.   Limitations: The data was collected from a survey which is not a purely objective measure; data collected might be biased or contain errors based on students’ judgment. Our rule based system provides estimation to the value and the relationship between the parameters are not fully represented. We did not get the opportunity to collect actual usage data with sensors or smart meters. This study was done within a specific group of students therefore not entirely applicable in general or different context.   Take-aways: Behavior play significant role in saving electricity. Basic AI based systems can be developed and use to enhance students’ consciousness and effectiveness about their electricity consumption. The combination of behavioral study and technology is key to achieving an efficient and sustainable usage of electricity.  

[1] W. Abrahamse and L. Steg, “Residential electricity consumption behavior: Influencing factors, related theories and intervention strategies,” Renewable and Sustainable Energy Reviews, 2017. [2] A. Amasyali and N. M. El-Gohary, “Machine learning for estimation of building energy consumption and performance: A review,” Renewable and Sustainable Energy Reviews, 2018. [3] S. S. Sahoo et al., “IoT-based analysis for smart energy management,” arXiv preprint arXiv:2311.18643, 2023. [4] K. Kong et al., “Short-term load forecasting using deep neural networks,” arXiv preprint arXiv:1811.03242, 2018. [5] Z. Wei et al., “A systematic review of deep learning and machine learning for building energy,” arXiv preprint arXiv:2202.12269, 2022. [6] Y. Chen et al., “Machine learning approach to uncovering residential energy consumption patterns based on socioeconomic and smart meter data,” arXiv preprint arXiv:2104.05154, 2021. [7] E. Costanza et al., “Socially intelligent interfaces for increased energy awareness in the home,” arXiv preprint arXiv:2106.15297, 2021. [8] N. A. M. Salleh et al., “The behaviour of households towards electricity consumption: A case study at Seremban,” 2021. [9] S. Darby, “Reducing electrical energy consumption through behaviour changes,” 2006. [10] M. Zhang et al., “Energy-saving behavior and its impact on energy consumption,” Sustainability, 2023.