Back to search

Rochester Institute of Technology

Modeling the impact of behavioral changes and heterogeneity on U.S. energy demand

Abstract

dc:description.abstract

Energy consumption is fundamental to the functioning of modern society, powering essential activities such as transportation, cooking, and working. The drivers of energy demand are influenced by an interplay of physical characteristics, climatic conditions, policy decisions, and behavioral patterns. These behavioral patterns are shaped by socio-demographic factors such as age, income, and employment status, which introduce variability in energy usage. Additionally, the COVID-19 pandemic has introduced changes in daily behaviors, which is persisting into the post-pandemic era and further influencing changes in energy consumption. Understanding and quantifying the diversity and changes in societal behaviors is crucial for energy management and planning. Nevertheless, this task is particularly challenging due to the dynamic nature of behavior and the variations across different population segments. This dissertation aims to reduce the knowledge gap on how behavioral (time-use) changes and socio-demographic variations impact energy demand, thereby informing energy policy and planning decisions. First, we show that time-use changes due to digitalization, especially increase in teleworking, shifted energy demand during COVID-19. This impact persisted into the post-COVID-19 era to change where and how energy is being consumed. We examine and forecast energy consumption for three sectors: residential, non-residential, and transportation, with two scenarios: (1) temporary shift in telework due to COVID-19 (14% teleworking in 2030) and (2) permanent shift in telework that increases with historical trends (34%). Due to COVID- 19, increasing telework by 20 % will lead to 3.6 % net energy demand reduction in 2030, mostly due to reduction in non-residential sector. Although, telework shifts energy demand to home usage, this is offset by reductions in other sectors. Thus, it is important to account for potential behavior changes in energy modeling and forecasting. Second, since there is time-use heterogeneity due to specific socio–demographic characteristics, such as teleworking as shown in the first part, in this second part, we examine and simulate residential energy behavior for different socio-demographic groups. Our study stochastically models 10 residential activity schedules related to energy consumption across 24 distinct socio- demographic segments, combining age, income, and employment status. Subsequently, validating these simulations against “ground truth” schedules. Our finding reveals time-use differences stemming from nuances in combinations of age, income, and employment status, with varying degrees of influence. In line with the Justice40 Initiative for equitable energy transitions, our study emphasizes the need to incorporate diverse socio-demographic behaviors into energy models. This ensures that disadvantaged communities are not underrepresented, especially as we found differences in usage patterns between low-income, full-time workers and their higher-income counterparts. Therefore, our findings underscore the need for detailed inclusion of these characteristics in energy models to facilitate tailored and effective energy reduction strategies across diverse populations. Lastly, now that we are in the post-pandemic era, we explore the change in residential energy consumption by comparing pre- and post-COVID-19 behaviors using empirical time-use data from 2019 and 2022. Results showed an increase in usage of devices and home appliances. Linking these behavioral changes to associating devices and appliances, while controlling for energy efficiency and technological changes, increase in behavior for various activities contribute to 1.76% of increase in residential energy consumption. This increase resulted from more time spent cooking, gaming, working, and using electronic devices for other purposes at home. These results highlight the importance of updating energy models to reflect the current behaviors for a more effective energy planning and policy development in the post-COVID era. This dissertation contributes to a better understanding of time-use changes and heterogeneity in energy consumption behavior, which can aid policy makers and planners in making informed decisions in energy management and tailor both technology and behavior interventions to suit specific groups and prepare for future scenarios as we transition into the “new normal” post- COVID-19.

Degree

thesis:*
Name thesis:degree_name
Sustainability (Ph.D.)
Level thesis:degree_level
Dissertation
Discipline thesis:degree_discipline
Sustainability, Department of

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Phoung, Sinoun
Contributors dc:contributor
  • Eric Williams

Subjects

dc:subject × 5

Identifiers

dc:identifier.*
Repository record dc:identifier
https://repository.rit.edu/theses/11911

Chain of custody

source
Harvested from
Rochester Institute of Technology
Base URL
repository.rit.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Phoung, Sinoun. Modeling the impact of behavioral changes and heterogeneity on U.S. energy demand. Dissertation thesis, https://repository.rit.edu/theses/11911